Boards and Shareholders in European Listed Companies PDF

BOARDS AND SHAREHOLDERS IN
EUROPEAN LISTED COMPANIES
With contributions by distinguished scholars from legal and financial

backgrounds, this collection of essays analyses four main topics in the
corporate governance of European listed firms: (i) board structure, com-
position and functioning and their interaction with ownership structure;
(ii) board remuneration; (iii) shareholder activism; and (iv) corporate
governance disclosure based on the ‘comply or explain’ approach.
The authors provide new comparative evidence and analyse its impli-
cations for the policy debate. They challenge the conventional wisdom
that corporate governance in European firms was systematically dysfunc-
tional. While proposals aimed at increasing disclosure and accountability
are usually well grounded, caution is suggested when bringing forward
regulatory changes with respect to proposals targeting specific gover-
nance arrangements, especially in the fields of board composition and
shareholder activism. They argue that the ‘comply or explain’ principle
should be retained and that further efforts should be exercised to enhance
disclosure.
massimo belcredi is Professor of Corporate Finance at the Università

Cattolica of Milan. He has written numerous books and articles in the
fields of corporate finance, corporate governance, ownership and board
structure, law and economics.
guido ferrarini is Professor of Business Law and Capital Markets
Law at the University of Genoa, and Director of the Genoa Centre for
Law and Finance. Among other important roles, he was an adviser to the
Corporate Governance Committee of the Italian Stock Exchange. He has
published widely on the topics of corporate governance, financial law,
corporate law and business law.
international corporate law and financial
market regulation
Corporate law and financial market regulation matter. The global financial
crisis has challenged many of the fundamental concepts underlying corporate
law and financial regulation; but crisis and reform has long been a feature of
these fields. A burgeoning and sophisticated scholarship now challenges and
contextualises the contested relationship between law, markets and compa-
nies, domestically and internationally. This Series informs and leads the
scholarly and policy debate by publishing cutting-edge, timely and critical
examinations of the most pressing and important questions in the field.
Series Editors
Professor Eilìs Ferran, University of Cambridge
Professor Niamh Moloney, London School of Economics and Political
Science
Professor Howell Jackson, Harvard Law School
Editorial Board
Professor Marco Becht, Professor of Finance and Economics at Université
Libre de Bruxelles and Executive Director of the European Corporate
Governance Institute (ECGI)
Professor Brian Cheffins, S.J. Berwin Professor of Corporate Law at the
Faculty of Law, University of Cambridge
Professor Paul Davies, Allen & Overy Professor of Corporate Law and
Professorial Fellow of Jesus College, University of Oxford
Professor Luca Enriques, Visiting Professor, Harvard Law School
Professor Guido Ferrarini, Professor of Business Law at the University of
Genoa and Fellow of the European Corporate Governance Institute (ECGI)
Professor Jennifer Hill, Professor of Corporate Law at Sydney Law School
Professor Klaus J. Hopt, Emeritus Scientific Member, Max Planck Institute of
Comparative and International Private Law, Hamburg
Professor Hideki Kanda, Professor of Law at the University of Tokyo
Professor Colin Mayer, Peter Moores Professor of Management Studies at the
Saïd Business School and Director of the Oxford Financial Research Centre
James Palmer, Partner of Herbert Smith, London
Professor Michel Tison, Professor at the Financial Law Institute of the
University of Ghent
Andrew Whittaker, General Counsel to the Board at the UK Financial
Services Authority
Professor Eddy Wymeersch, former Chairman of the Committee of
European Securities Regulators (CESR); former Chairman of the IOSCO
European Regional Committee, and Professor of Commercial Law,
University of Ghent.
BOARDS AND
SHAREHOLDERS IN
EUROPEAN LISTED
COMPANIES
Facts, context and post-crisis reforms
A research project promoted by Emittenti Titoli S.p.A.
Edited by
MASSIMO BELCREDI
and
GUIDO FERRARINI
EMITTENTI TITOLI S.p.A.
Emittenti Titoli is a company promoted by Assonime and created in 1998. Its shareholders are
some of the main non-financial Italian listed firms and their controlling holding companies.
Emittenti Titoli promotes the development of the securities market in the interest of Italian
issuers. After having acquired a 6.5% participation in Borsa Italiana, Emittenti Titoli contributed
to define both the governance of the Italian Stock Exchange and its listing rules, trying to
counterbalance the influence of intermediaries. Following the acquisition of Borsa Italiana by
the London Stock Exchange Group, Emittenti Titoli is currently the first Italian shareholder of
LSE, holding 1.6% of share capital. Emittenti Titoli publishes, jointly with Assonime, an annual
analysis on the corporate governance of Italian listed companies and on the state of implemen-
tation of the Italian Governance Code. Emittenti Titoli is led by a Board of Directors composed of
15 members, chaired (since 2012) by Luigi Abete.
University Printing House, Cambridge CB2 8BS, United Kingdom
Published in the United States of America by Cambridge University Press, New York
Cambridge University Press is part of the University of Cambridge.
It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107040564
© Emittenti Titoli S.p.A.
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2013
Printed in the United Kingdom by Clays, St Ives plc
A catalogue record for this publication is available from the British Library
ISBN 978-1-107-04056-4 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
FOREWORD
The papers collected in this volume, written by a group of leading

European scholars, are the result of a research project promoted by
Emittenti Titoli.
In recent years, the academic debate focused on the relationship
between corporate governance and the financial crisis. It is still unclear
whether, and to what extent, dysfunctional corporate governance has
contributed to the recent financial crisis. Nonetheless, a number of
policy proposals have been put forward to redress the most obvious
failures. In particular, the European Commission published two Green
Papers, in 2010 and 2011, respectively, targeting corporate governance in
financial institutions and remuneration policies and the EU corporate
governance framework. In December 2012, on the basis of its reflection
and of the results of previous consultations, the European Commission
published an Action Plan outlining future initiatives in the areas of
European company law and corporate governance.
This volume analyses four main topics in the corporate governance of
European listed firms: (i) board structure and composition and their
interaction with ownership structure; (ii) board remuneration; (iii)
shareholder activism; and (iv) corporate governance disclosure based
on the ‘comply or explain’ approach. For each of them, the authors
provide new evidence and analyse its implications for the policy debate.
In the main, they challenge the conventional wisdom that corporate
governance in European firms was systematically dysfunctional and,
therefore, they suggest caution in bringing forward regulatory changes.
Basically, while proposals aimed at increasing disclosure and account-
ability are usually well-grounded, caution is needed with respect to
proposals targeting specific governance arrangements (especially in the
fields of board composition and shareholder activism). Similarly, they
argue that the ‘comply or explain’ principle should be retained, but that
further efforts should be exercised to enhance disclosure.
v
vi foreword
Emittenti Titoli, a company promoted by Assonime, and whose capital

is held by the most important Italian non-financial companies, is happy to
offer the results of this research project to the international financial
community in order to further stimulate the debate on corporate
governance.
Luigi Abete
Chairman, Emittenti Titoli
CONTENTS
Foreword page v
Figures ix
Tables x
Contributors xiii
1 Corporate boards, incentive pay and shareholder activism in

Europe: main issues and policy perspectives 1
m a s s im o be l cr e d i a n d g uid o f er r a r i n i
2 European corporate governance codes and their
effectiveness 67
eddy wymeersch
3 Restructuring in family firms: a tale of two crises 143
c h r i s t i a n a n d r e s , l o r en zo ca p r i o a n d
ettore croci
4 Corporate boards in Europe: size, independence and gender
diversity 191
d a n i e l f e r r e i r a a n d to m k i r ch m a i e r
5 Board on Task: developing a comprehensive understanding of
the performance of boards 225
jaap winter and erik van de loo
6 Directors’ remuneration before and after the crisis: measuring
the impact of reforms in Europe 251
r o b e r t o ba r o n t i n i, s t e f a n o b o z z i , g u i d o
ferrarini and maria-cristina ungureanu
vii
viii contents
7 Shareholder engagement at European general

meetings 315
luc renneboog and peter szilagyi
8 Board elections and shareholder activism: the Italian
experiment 365
massimo belcredi, stefano bozzi and
car m in e d i n oia
Index 423
FIGURES
4.1 Time trends in board characteristics: European Union, 2000–10 page 195
4.2 Time trends in board size: European Union, 2000–10, stable samples 196
4.3 Time trends in board independence: European Union, 2000–10, stable
samples 197
4.4 Time trends in board gender diversity: European Union, 2000–10, stable
samples 198
4.5 Time trends in board characteristics: United States, 2000–10 199
5.1 The Group lens 237
6.1(a) Compliance in 2007. Financial vs non-financial companies 283
6.1(b) Compliance in 2010. Financial vs non-financial companies 283
ix
TABLES
3.1 Performance, family ownership and crisis (777 companies) page 160
3.2 Performance, family ownership and crisis (regressions) 167
3.3 Performance, family CEOs and crisis 171
3.4 Investments, downsizing and increase in size 175
3.5 Crises, wages and employment 180
4.1 Board size across countries (2010) 200
4.2 Board independence across countries (2010) 201
4.3 Board gender diversity across countries (2010) 202
4.4 One-tier versus two-tier board structures 205
4.5 Corporate board size in Europe: the impact of firm characteristics,
industries and countries (2010) 207
4.6 Corporate board independence in Europe: the impact of firm
characteristics, industries and countries (2010) 210
4.7 Board gender diversity in Europe: the impact of firm characteristics,
industries and countries (2010) 214
4.8 Board size in 2010 and firm characteristics in 2007 216
4.9 Board independence in 2010 and firm characteristics in 2007 218
4.10 Board gender diversity in 2010 and firm characteristics in 2007 219
5.1 Matrix of board interaction 241
6.1 Say-on-pay regulations in various jurisdictions 261
6.2 Remuneration characteristics and expected effect on European
firms 263
6.3 Criteria describing the governance and disclosure of remuneration
practices 267
6.4(a) Characteristics of the firms included in the sample for 2007 and 2010:
whole sample 270
6.4(b) Characteristics of the firms included in the sample for 2007 and 2010:
sample of financial firms 272
6.4(c) Characteristics of the firms included in the sample for 2007 and 2010:
sample of non-financial firms 274
6.5 Country-specific evolution of the 15 criteria on remuneration and
governance characteristics 277
6.6 Governance, disclosure variables and firms’ characteristics 286
x
tables xi
6.7 Governance, disclosure variables and firms’ ownership
characteristics 290
6.8(a) Mean (median) total compensation of the board of directors 292
6.8(b) Mean (median) total compensation of the CEO 294
6.9(a) Composition of CEO mean and median pay and stock-based incentive
portfolio: whole sample 297
6.9(b) Composition of CEO mean and median pay and stock-based incentive
portfolio: sample of non-financial firms 298
6.9(c) Composition of CEO mean and median pay and stock-based incentive
portfolio: sample of financial firms 299
6.10(a) Regression analysis of determinants of CEO total
compensation 300
6.10(b) Regression analysis of determinants of board total
compensation 301
7.1 The use of control-enhancing mechanisms 322
7.2 Statutory requirements with respect to general meetings 324
7.3 Number of management and shareholder proposals in Europe by country
and year 330
7.4 Votes for management and shareholder proposals in Europe 332
7.5 Number of shareholder proposals and votes for the proposals in the
US 338
7.6 Financial performance and ownership characteristics of the sample
firms 341
7.7 Country-level shareholder rights and corporate governance 342
7.8 Regressions explaining the votes for management proposals 345
7.9 Determinants of shareholder proposal submissions 350
7.10 Regressions explaining the votes for shareholder proposals 353
8.1 Descriptive statistics: firm characteristics 384
8.2(a) Descriptive statistics: ownership structure according to the identity of the
ultimate shareholder 385
8.2(b) Descriptive statistics: ownership structure according to the identity of the
8.3(a) Descriptive statistics: board elections according to the identity of the
8.3(b) Descriptive statistics: board elections according to the identity of the
8.4 Determinants of the decision to submit a ‘minority’ slate (ownership
defined in terms of concentration) 396
defined in terms of ultimate shareholder identity) 398
concentration and voting rules) 400
xii tables
8.7 Determinants of the decision to submit a ‘mutual fund’ slate (ownership
defined in terms of concentration) 405
defined in terms of ultimate shareholder identity) 407
concentration and voting rules) 409
CONTRIBUTORS
christian andres is Professor of Empirical Corporate Finance, Otto

Beisheim School of Management, WHU.
roberto barontini is Professor of Corporate Finance, Scuola
Superiore Sant’Anna in Pisa and Director of the Masters’ Course in
Innovation, Management and Service Engineering (MAINS).
massimo belcredi is Professor of Corporate Finance, Università
Cattolica of Milan and an independent director of Arca SGR and Erg.
stefano bozzi is Associate Professor of Corporate Finance, Università
Cattolica of Milan.
lorenzo caprio is Professor of Corporate Finance, Università Cattolica
of Milan and an independent director of Sogefi.
ettore croci is a Lecturer in Corporate Finance, Università Cattolica
of Milan.
carmine di noia is Head of the Capital Market and Listed Companies
Unit and Deputy Director General at Assonime.
guido ferrarini is Professor of Business Law, University of Genoa
and Director of Genoa Centre for Law and Finance. He is founder,
director and fellow of the European Corporate Governance Institute
(ECGI), Brussels.
daniel ferreira is Professor of Finance at the London School of
Economics, Director of the Ph.D. Programme in Finance and Research
Fellow of CEPR and ECGI.
tom kirchmaier is a Lecturer in Business Economics and Strategy at
Manchester Business School and Fellow of the Financial Markets Group,
London School of Economics.
xiii
xiv contributors
luc renneeboog is Professor of Corporate Finance, Tilburg University

and Director of Graduate Studies, CentER for Economic Research.
peter szilagyi is a Lecturer in Finance at the Judge Business School,
University of Cambridge.
maria cristina ungureanu is Advisor Corporate Governance at
Sodalin, a global provider of corporate governance consulting, share-
holder transactions and institutional investor relations. She is also fellow
of the Genoa Centre for Law and Finance.
eric van de loo is Professor of Leadership VU Amsterdam and
TiasNimbas Tilburg, Visiting Clinical Professor of Leadership INSEAD
and Tun Ismail Ali Chair of Leadership in Kuala Lumpur.
jaap winter is Professor of Corporate Governance, Duisenberg
School of Finance, Amsterdam, and Professor of International Company
Law in Amsterdam.
eddy wymeersch is Chairman of the Public Interest Oversight Board
in Madrid and Board Member of Euroclear SA and of the Association for
the Financial Markets in Europe (AFME).
1
Corporate boards, incentive pay and shareholder

activism in Europe: main issues and policy
perspectives
massimo belcredi and guido ferrarini
1. Introduction*
1.1. Purpose and scope
In this chapter, we offer an overview of the present volume, placing the
same in the context of recent European Union (EU) reforms and of
corporate governance theory and summarising the main outcomes of the
following chapters. In addition, we offer some policy perspectives – as to
boards, incentive pay and shareholder activism – based on the theoret-
ical and empirical outcomes of the research project of which this volume
is the product. In drawing this broad picture, we underline particu-
larly that variances in ownership structures of listed companies and in
the adoption of either a shareholder value or a stakeholder approach
have pervasive implications for corporate governance issues. For exam-
ple, board composition criteria may reflect a stakeholder orientation,
such as that found in the German codetermination system (Schmidt
2004). Also the board’s function, the role of independent directors and
incentive pay arrangements may vary depending on whether diffuse
shareholders or blockholders own the company. Similarly, diffuse own-
ership companies represent the natural setting for shareholder activism,
which may not be a cost-effective solution in controlled corporations.1
* The analysis across the volume refers to EU and Member State regulation as of 15 January
2013.
1
Within this context, it is debated whether additional reform, aimed at stimulating
activism of institutional investors (such as, for instance, the adoption of cumulative,
proportional or slate voting in corporate elections), may be useful (see Section 6.3.2.
below and Chapter 8).
1
2 massimo belcredi and guido ferrarini
In general, we assume that boards are an essential mechanism for

directing the company and monitoring the agency costs of management,
while incentive pay is important to align the interests of professional
managers with those of shareholders. Moreover, we assume that share-
holder activism can work as a useful complement to these governance
mechanisms by exercising pressure on boards and holding them
accountable for the performance of their monitoring functions.
However, the effectiveness of similar mechanisms depends on a variety
of factors, including the quality of corporate law and its enforcement, the
degree to which private codes of best practice are complied with, and
the institutional context in which boards and shareholders operate. In
particular, ownership structures in a given system or company affect the
equilibrium between the corporate governance mechanisms that we
analyse in this volume. While mainstream global corporate governance
is heavily influenced by the model of the Berle and Means corporation,
an analysis of the European context requires a less biased approach in
order to catch the richness of governance models and diversified experi-
ences (as particularly shown by the study of family firms in Chapter 3).
In the remainder of this Chapter, we introduce recent reform initia-
tives and the variety of corporate governance systems in Europe, sketch-
ing out the alternative between shareholder and stakeholder governance
and the specificities of bank governance. In Section 2, we outline the
main tools for controlling agency costs, including market mechanisms,
corporate law, codes of best practice and the ‘comply or explain’
approach, and bank prudential regulation. In Section 3, we analyse the
impact of ownership structures on agency costs and comment on
Chapter 3 on family firms in Europe. In Section 4, we examine the theory
and practice of boards, in light of EU law and soft law and of the analysis
in Chapters 4 and 5 on board size, independence and gender diversity
and also of the limitations inherent to a ‘law and economics’ approach.
In Section 5, we examine the theory and practice of incentive pay, in light
of EU soft law and banking regulation, and summarise the outcomes of
an empirical analysis on pay practices in large European listed compan-
ies included in Chapter 6. In Section 6, we analyse shareholder activism
in Europe and summarise the outcomes of two empirical contributions
(one on activism in the EU and the United States (US), the other on
activism in Italian corporate elections) contained in Chapters 7 and 8.
In Section 7, we outline some policy considerations on the topics con-
sidered in the previous four sections. Section 8 draws some general
conclusions.
boards, incentive pay, shareholder activism 3
1.2. EU reform
In the present section, we review the legal reforms that have affected EU
corporate governance since the beginning of the current century. These
reforms addressed the main corporate governance failures which gov-
ernments and legislators identified in the 2001–2 corporate scandals and
the 2008 financial crisis (Enriques and Volpin 2007; Bainbridge 2012).
Similar failures affected both the internal governance structures of
corporations – including those relating to the audit of accounts – and
the essential mechanisms for capital market efficiency, such as securities
underwriters, financial analysts and rating agencies (Gilson and
Kraakman 2003; Skeel 2011). This chapter focuses mainly on corporate
boards and shareholders, in line with the remainder of this volume.
Indeed, boards have a key governance role and perform monitoring
and advisory tasks with respect to firms’ managers. Shareholders have
fundamental governance rights, including that of appointing the board,
which derive from their function as residual risk-bearers. In line with
recent Commission Green Papers, this chapter and the whole volume
take into consideration both shareholder activism (which occurs mainly
in diffuse ownership companies) and the protection of minority share-
holders (which typically concerns controlled corporations).
1.2.1. After Enron

The ‘new economy’ bubble highlighted serious corporate governance
shortcomings, mainly related to internal controls, executive remunera-
tion and external auditors (Coffee 2005). Corporate frauds and account-
ing failures had been made easier by lack of appropriate internal controls
for which the firms’ managers and directors were generally responsible.
Moreover, stock options and other incentives were aggressively resorted
to, contributing to managers manipulating share prices through false
information relative to their firms’ financial performance. The auditors
and other gatekeepers, such as investment bankers, business lawyers and
rating agencies, largely contributed to the first crisis of this century (i.e.
the corporate scandals era), by covering frauds and aiding insolvent
companies to conceal their true financial conditions (Coffee 2002;
Gordon 2002; Miller 2004; Ferrarini and Giudici 2006).
Wide reforms were sought both at EU and domestic levels, often
modelled along the US Sarbanes-Oxley Act, which had, however, been
enacted in a remarkably brief period, with minimal legislative processing
(Bainbridge 2012). The European response to the financial scandals was
relatively less hasty, given that the epicentre of the 2001–2 turmoil had
been the US, and also considering the more complex political process for
EU legislation. Moreover, the final response in Europe was not as strong
and pervasive as that in the US (Ferrarini et al. 2004). The EU Action
Plan was out in the 2003 Communication from the Commission on
Modernising Company Law and Enhancing Corporate Governance in the
European Union,2 which was prepared on the basis of a report by the
High Level Group of company law experts appointed by Commissioner
Bolkestein and chaired by Jaap Winter (the Winter Report).3 The
Commission’s Action Plan envisaged four main pillars for corporate
governance reform.
(i) The first referred to enhancing corporate governance disclosure,
with the argument that more than forty corporate governance
codes had been adopted in Europe, their contents being widely
convergent; however, ‘information barriers’ undermined
shareholders’ ability to evaluate the governance of companies.
The Commission proposed that companies be required to include
in their annual reports and accounts a comprehensive corporate
governance statement covering the key elements of their corporate
governance structures and practices. This statement should carry a
reference to a code on corporate governance, designated for use at
national level that the company complies with, or in relation to
which it explains deviations. This proposal led to the adoption in
2006 of the new Article 46a of Directive 78/660/EEC on the annual
accounts of certain types of companies, which required companies
with securities admitted to a regulated market to publish a corpor-
ate governance statement in their annual report.4 The content
and implementation of the EU ‘comply or explain’ principle are
2
See Communication from the Commission to the Council and the European Parliament,
Modernising Company Law and Enhancing Corporate Governance in the European
Union – A Plan to Move Forward, Brussels, 21.5.2003, COM(2003) 284 final.
3
See the Report by the High Level Group of Company Law Experts, A Modern Regulatory
Framework for Company Law in Europe, Brussels, 4 November 2002.
4
See Article 1, para. 7 of Directive 2006/46/EC of the European Parliament and of the
Council of 14 June 2006 amending Council Directives 78/660/EEC on the annual
accounts of certain types of companies, 83/349/EEC on consolidated accounts, 86/635/
EEC on the annual accounts and consolidated accounts of banks and other financial
institutions and 91/674/EEC on the annual accounts and consolidated accounts of
insurance undertakings, O.J. 16.8.2006, L 224/1.
analysed briefly in the following paragraph, and more extensively in

Chapter 2.
(ii) The second pillar contemplated strengthening shareholders’ rights
in terms of both electronic access to information and procedural
rights (to ask questions, table resolutions, vote in absentia, and
participate in general meetings via electronic means). The
Commission proposed that the facilities relevant for the exercise
of similar rights should be offered to shareholders throughout the
EU, while specific problems related to cross-border voting should
be resolved urgently. This led to the adoption of the Shareholder
Rights Directive,5 which is analysed briefly in Section 6 and in
Chapter 7.
(iii) The third pillar involved modernising the board of directors. First, as
to board composition, non-executive or supervisory directors who,
in the majority, are independent, should take decisions in key areas
where executive directors have conflicts of interest – such as remu-
neration and supervision of the audit of company accounts. Second,
the directors’ remuneration regime should require disclosure of
remuneration policy and remuneration details of individual direc-
tors in the annual accounts; prior approval by the shareholder
meeting of share and share option schemes in which directors
participate; and proper recognition in the annual accounts of the
costs of such schemes for the company. Third, the collective
responsibility of all board members for key financial and non-
financial statements should be clearly recognised under national
legal systems.
The proposals relative to board composition found detailed
specification in the Commission Recommendation of 15 February
2005 on the role of non-executive or supervisory directors of listed
companies and on the committees of the (supervisory) board6
(commented upon briefly under Section 4.2.); the proposals relative
to directors’ remuneration found specification in the Commission
Recommendation of 14 December 2004 fostering an appropriate
regime for the remuneration of directors of listed companies (see
Section 5.2. and Chapter 6); and those on collective responsibility
5
Directive 2007/36/EC of the European Parliament and of the Council of 11 July 2007 on
the exercise of certain rights of shareholders in listed companies, O.J. 14.7.2007, L 184/17.
6
O.J. 25.2.2005, L 52/51.
were translated into Articles 50b and 50c of Directive 78/660/EEC

on the annual accounts of certain types of companies.7
(iv) The fourth pillar involved co-ordinating corporate governance
efforts of Member States, with reference both to the development
of national corporate governance codes and to the monitoring and
enforcement of compliance and disclosure (a topic dealt with in
Chapter 2).
These four pillars fundamentally marked two areas for corporate govern-
ance reform – boards and shareholder rights – which are intercon-
nected to the extent that companies are run in the interest of
shareholders and the latter monitor board governance and appoint and
remove directors. The Commission further suggested two main paths for
EU reform, which were subsequently implemented through directives or
recommendations: disclosure of corporate governance structures and
functioning (including those concerning directors’ remuneration); and
setting of standards for board and remuneration practices, and for
shareholders’ information and rights.
1.2.2. The recent financial crisis

It is uncertain whether and to what extent corporate governance con-
tributed to the recent financial crisis. While policymakers generally offer
a positive answer (Kirkpatrick 2009), the topic is still debated amongst
academics. For sure, a distinction should be made between financial
institutions – banks in particular – and other companies, given that the
former were at the epicentre of the financial crisis, both in the US and in
Europe, while non-financial companies were affected by the crisis but
did not show risk-management or other governance failures similar to
those experienced by financial institutions (Cheffins 2009). Moreover,
empirical research has proven that banks which failed in the crisis had
adopted ‘good’ corporate governance standards (Beltratti and Stulz
2012). However, other research has shown that banks which fared better
in the crisis had better risk-management systems in place, suggesting
that the criteria defining ‘good’ governance need to be reconsidered
(Ellul and Yerramilli 2012). The European Commission sided with
governments and international organisations arguing that corporate
governance had failed in the crisis, but appropriately distinguished
7
See Article 1, para. 8 of Directive 2006/46/EC (n. 4 above), inserting a new Section 10A
(Duty and liability for drawing up and publishing the annual accounts and the annual
report) in the Directive on annual accounts.
between financial institutions and other firms. Therefore, two Green

Papers were published, one in 2010 on Corporate Governance in
Financial Institutions and Remuneration policies8 and the other in
2011 on The EU Corporate Governance Framework.9
The 2010 Green Paper was part of a programme for reforming the
regulatory and supervisory framework of financial markets announced
in a Commission Communication of 4 March 2009,10 which was based
on the conclusions of the de Larosière Report.11 In the Green Paper’s
introduction, the Commission stated:
As highlighted by the de Larosière report, it is clear that boards of

directors, like supervisory authorities, rarely comprehended either the
nature or scale of the risks they were facing. In many cases, the share-
holders did not properly perform their role as owners of the companies.
Although corporate governance did not directly cause the crisis, the lack
of effective control mechanisms contributed significantly to excessive
risk-taking on the part of financial institutions.
This statement helps understand the remaining contents of the Green

Paper, which include the role and composition of the (supervisory)
board; risk management as a key aspect of corporate governance;
and appropriate shareholder monitoring and the role of supervisory
authorities with respect to the internal governance of financial institu-
tions. We pay some attention to the specificities of bank governance in
Section 1.3.2. and to the role of banking regulation and supervision in
Section 2.4. However, the discussion found in the 2010 Green Paper
largely overlaps with the analysis developed in the 2011 Green Paper, so
that they can be bundled in our analysis.
Indeed, the 2011 Green Paper extends the arguments applicable
to financial institutions to other firms, assuming that ‘corporate
governance is one means to curb harmful short-termism and excessive
risk-taking’ for firms in general and suggesting that the Green Paper
should ‘assess the effectiveness of the current corporate governance
framework for European companies.’ Similar to the 2003 Commission
Communication on Modernising Company Law, the 2011 Green Paper
focuses on the board of directors, emphasising that ‘effective boards are
needed to challenge executive management’; on shareholders, arguing
8
COM(2010) 284 final. 9
COM(2011) 164 final. 10 COM(2009) 114 final.
11
Report of the High-Level Group on Financial Supervision in the EU published on
25 February 2009, available at http://ec.europa.eu/internal_market/finances/docs/
de_larosiere_report_en.pdf.
that they must ‘engage with companies and hold management to account
for its performance’; and on the ‘comply or explain’ approach, claiming
that the informative quality of explanations published by companies is
‘not satisfactory’ and the monitoring of the codes’ application is
‘insufficient’. We shall make specific references to the 2011 Green
Paper throughout the present chapter, highlighting some of its main
features in connection with the individual topics touched upon in our
analysis.
1.3. Varieties of corporate governance

As anticipated, variances in European corporate governance are import-
ant and depend mainly on the ownership structures of listed companies
and the national systems’ adherence to either a shareholder or a stake-
holder approach (Hansmann and Kraakman 2001; Clarke and Chanlat
2009; Kraakman et al. 2009). In this Section, we outline the key differ-
ences between shareholder and stakeholder governance, focusing on
scholarly definitions and positions taken by EU policy documents. We
also present the core specificities of bank governance, which determine
the regulation and supervision of board structures and functions, and the
reorientation of the relevant criteria for the protection of stakeholders
(depositors) and the financial system (systemic risk) rather than for mere
shareholder wealth maximisation.
1.3.1. Shareholder v. stakeholder governance

There is no clear-cut, generally accepted definition of corporate govern-
ance. Many definitions are found in the academic literature and in
codes of best practice, but differences, though rarely spelled out, are
substantial. The dominant approach in the financial literature (Tirole
2006) focuses on the relationship between firms and suppliers of funds
(debt and equity). An oft-cited work argues that ‘corporate governance
deals with the ways in which suppliers of finance to corporations assure
themselves of getting a return to their investment’ (Shleifer and Vishny
1997). In other words, corporate governance concerns how corporate
insiders can credibly commit to return funds to investors, so as to attract
outside financing. Suppliers of debt and equity may benefit from several
control mechanisms, based on either legal protection (through contract
and/or regulation) or sheer power deriving from concentration of claims.
A similar view is sometimes criticised as being too narrow, for other
stakeholders (employees, clients, local communities) have an interest in
how the firm is run (Blair 1995; Blair and Stout 2001). Becht et al. (2002)
offer a broad definition under which ‘corporate governance is concerned
with the resolution of collective action problems among dispersed
investors and the reconciliation of conflicts of interest between various
corporate claimholders.’ These definitions imply that corporate govern-
ance is a ‘common agency’ problem, involving an agent (the Chief
Executive Officer, CEO) and multiple principals (shareholders, cred-
itors, employees, clients). Since the firm is a nexus of contracts (Jensen
and Meckling 1976) and contracts are incomplete, managerial discretion
arises and governance mechanisms are needed to allocate power and
create incentives. However, the presence of multiple principals blurs
corporate objectives and may ultimately compound agency problems,
providing the management with an ad hoc rationale to explain any
decision whatsoever (Williamson 1985; Tirole 2006). In a similar setting,
regulation may shift part of the discretionary powers to the regulator,
who will find a ‘political’ solution to these trade-offs.
Recent EU policy documents are rather ambivalent and fluctuate
between the two approaches. The 2011 Green Paper remarks that cor-
porate governance is traditionally defined (a) as the system by which
companies are directed and controlled and (b) as a set of relationships
between a company’s management, its board, its shareholders and other
stakeholders. The first part of the definition echoes the shareholder
approach already followed in the UK by the Cadbury Report, which
emphasises the respective roles and responsibilities of boards and share-
holders. The board should set the company’s strategic aims, provide the
leadership to put them into effect, supervise the management of the
business and report to the shareholders. Shareholders appoint (and
possibly remove) the directors. Under this approach, corporate govern-
ance centres on the agency relation between boards (agents) and share-
holders (principals). Other stakeholders are protected by contracts and/
or regulation (concerning bankruptcy, competition, labour, etc.), rather
than by traditional corporate governance institutions. However, share-
holder primacy has come under closer scrutiny in the last few years,
particularly in financial institutions, where corporate governance
arrangements have been criticised for distorting managerial incentives
and/or contributing to the financial crisis (Kirkpatrick 2009; Beltratti
and Stulz 2012; Fahlenbrach and Stulz 2011; Admati et al. 2012; Becht
et al. 2012).
The second part of the Green Paper’s definition reflects a stakeholder
view, similar to that found in the Organization for Economic Co-operation
and Devolopment (OECD) Principles of Corporate Governance. These

Principles highlight that (a) different classes of shareholders may exist
and need to be treated in an equitable manner and (b) other stakeholders
may possess rights established by law or through mutual agreements,
which may also extend to corporate governance institutions (e.g.
employees may obtain board representation and have a say in specific
corporate decisions). From a similar perspective, corporate governance
institutions do not exclusively concern the relationship between man-
agers and (undifferentiated) shareholders. Rather, they must solve the
potential trade-offs between different kinds of agency problems, which
may justify regulating, for instance, the composition and role of the
board of directors.
The question therefore arises whether and to what extent the board
and/or shareholders’ powers should be regulated to reflect other
stakeholders’ interest. From a comparative perspective, the answers to
this question are diverse, as shown by the fact that workers’ participation
in boards is required in some countries, while special rules have been
adopted internationally for the corporate governance of financial insti-
tutions. In general, corporate governance institutions vary considerably
across countries and types of firms, with differences that are persistent
and largely dependent on specific institutional contexts (Bebchuk and
Roe 1999).
1.3.2. Bank governance

Banks are different from other firms for several reasons that matter from
a corporate governance perspective (Adams and Mehran 2003; Macey
and O’Hara 2003; Mülbert 2010; Ferrarini and Ungureanu 2011). First,
they are more influential than other firms, with the consequence that the
conflict between shareholders and fixed claimants, which is present in all
corporations, is more acute. Second, banks’ liabilities are largely issued as
demand deposits, while their assets, such as loans, have longer matur-
ities. The mismatch between liquid liabilities and illiquid assets may
become a problem in a crisis situation, as we saw vividly in the recent
financial turmoil, when bank runs took place at large institutions, threat-
ening the stability of the whole financial system. Third, despite contri-
buting to the prevention of bank runs, deposit insurance generates moral
hazard by incentivising shareholders and managers of insured institu-
tions to engage in excessive risk taking (Corrigan 1982; 2000). Moral
hazard is exacerbated when a bank approaches insolvency, because
shareholders do not internalise the losses from risky investments, but
instead benefit from potential gains (for example, by having an implicit

put option at strike price zero) (Macey and Miller 1992; Polo 2007).
While risk taking by non-bank corporations close to insolvency is con-
strained by market forces and contractual undertakings, banks in a
similar condition can continue to attract liquidity, thanks to deposit
insurance (Macey and O’Hara 2003; Sorkin 2009). Fourth, asset sub-
stitution is easier in banks than in non-financial firms (Levine 2004).
This allows for more rapid risk shifting, which further increases agency
costs between shareholders and stakeholders (and bondholders and
depositors in particular). In addition, banks are more opaque – it is
difficult to assess their risk profile and stability. Information asymmet-
ries, particularly for depositors, hamper market discipline and, in turn,
increase managers’ moral hazard.
For all these reasons, ‘good’ corporate governance (that is, aligning the
interests of managers and shareholders) may lead bank managers to
engage in more risky activities (Laeven and Levine 2009), since a major
part of the losses would be externalised to stakeholders, while gains
would be internalised by shareholders and managers (if properly aligned
by the right incentives). Prudential regulation and supervision aim to
reduce the excessive risk propensity of shareholders and managers in
order to guarantee the safety and soundness of banks. An exogenous
regulatory cost is allocated on excessively risky behaviour of bank man-
agers, reducing agency costs between shareholders and stakeholders.
Recent empirical research confirms that ‘good’ governance may not be
enough for bank soundness. Beltratti and Stulz (2012) investigate
whether banks’ poor performance in the recent crisis was the outcome
of a financial tsunami that hit them unexpectedly, or of some banks being
more inclined to experience large losses. The authors analyse possible
determinants (regulation, corporate governance, balance sheet and
income characteristics) of bank performance measured by stock returns
during the crisis for a sample of ninety-eight large banks across the
world, of which nineteen are US banks. They find no evidence for the
thesis advanced in a report by the OECD12 that the financial crisis, to an
important extent, can be attributed to failures and weaknesses in corpor-
ate governance arrangements (Kirkpatrick 2009). In particular, they
find no evidence that banks with better governance performed better
12
OECD, ‘Corporate Governance and the Financial Crisis: Key Findings and Main
Messages’, June 2009.
during the crisis. On the contrary, banks with more pro-shareholder

boards performed worse.
Adams (2009) reaches similar results assessing to what extent the crisis
can be attributed to bad governance of financial firms. She shows that banks
receiving bailout money from the US government under the Troubled Asset
Relief Program (TARP) had more independent boards, larger boards, more
outside directorships for board members, and greater incentive pay for
CEOs than non-TARP banks. Except for the finding of more independent
boards, these results are consistent with the idea that TARP banks had
worse governance. However, Adams finds it striking that TARP banks had
boards that were more independent. One explanation could be that inde-
pendent directors are less likely to have in-depth knowledge of their banks
and the financial expertise to understand complex transactions like securi-
tizations. In other words, greater independence may be detrimental for a
bank board because a more independent board will not have sufficient
expertise to monitor the actions of the CEO.
The criteria for examining corporate governance employed by the
studies mentioned above are open to debate. For instance, independent
directors are used as a proxy for good monitoring by the board, but this
monitoring depends on professional qualities and levels of engagement
in board activities that are not necessarily captured by current definitions
of independence (Ferrarini and Ungureanu 2011). Similarly, interna-
tional corporate governance indexes make reference to aspects such as
internal controls, which do not necessarily reflect the detailed require-
ments for proper monitoring of complex risk-management processes by
a bank board (Bhagat et al. 2008; Stulz 2008). Thus, while establishing a
prima facie case for excluding corporate governance as a main determin-
ant of the crisis, the above studies cannot be used for asserting that what
appeared to be ‘good’ governance at banks which failed was satisfactory
in practice and in no need of reform. A similar statement calls for proof
that banks failed despite best monitoring efforts deployed by their
boards, a proof no doubt difficult to offer, particularly in light of the
egregious risk-management failures seen in most troubled banks (Senior
Supervisors Group 2008; Stulz 2008). Moreover, recent empirical
research proves that banks that had strong risk-control systems in
place – as measured by the importance attached to the risk-management
function within the organisation and, in particular, by the existence and
role of the Chief Risk Officer – were more judicious in their exposure to
risky financial instruments before the crisis and, generally, fared better
post-crisis (Becht et al. 2012; Ellul and Yerramilli 2012).
2. Controlling agency costs

Agency problems stem from the information asymmetries characterising
modern business, which create an opportunity for principals to hire
better-informed agents.13 However, specialisation comes at a cost. The
delegation of discretionary powers, which are necessary to exploit the
agents’ superior capabilities, carries conflicts of interest. Agency costs
include those of writing and enforcing contracts. First, there are the costs
of structuring, monitoring and bonding contracts with conflicted agents.
Second, output is lost whenever the costs of full enforcement would
exceed the benefits (Fama and Jensen 1983). Several mechanisms and
institutions keep agency problems under control. In this Section we
consider the impact of product and financial markets; the role of corpor-
ate law; soft law and the related ‘comply or explain’ mechanism; and the
impact of prudential regulation on banks’ internal governance.
However, two preliminary remarks are necessary with reference to
corporate law and its impact on European corporate governance. The
first is that the EU dimension of the topic adds an additional complexity,
to the extent that not all cases in which corporate law has a role to play
are also cases in which EU intervention is appropriate. Under the sub-
sidiarity principle (Article 5 Treaty on European Union (TEU), legal
harmonisation should only occur when national legislation is unfit to
address existing cross-border externalities (ECLE 2011). This explains
why the role of European corporate law is rather limited and its impact
on corporate governance overall modest with respect to the role played
by national legislation and case law (Enriques and Volpin 2007). The
second remark is that EU law acknowledges the importance of soft law
in corporate governance and attempts, particularly through disclosure
(‘comply or explain’), to enhance the role of private codes. This reflects a
general trend in Europe, given that codes of best practice are widely
employed to address corporate governance issues in Member States; on
the other hand, it may also be seen as a reflection of the inherent limits of
13
Agency problems come in many guises. Tirole (2006) offers the following classification:
(a) insufficient effort, such as leisure on the job and inefficient allocation of work time to
various tasks; (b) extravagant investments, like suboptimal allocation of capital – i.e.
negative NPV decisions – due to conflicts of interest; (c) entrenchment strategies,
including actions taken by the managers to secure their own position, without regard
to the impact of the same on company value; and (d) self-dealing, ranging from benign
to illegal activities, such as consumption of perquisites, tunnelling and other behaviours
including thievery. Roe (2005) groups agency costs in two main categories: ‘stealing and
shirking’, i.e. expropriation and waste of resources.
EU powers in this area, since EU legislation can easily cover disclosure by

European listed issuers, but would find it more problematic directly to
address the typical agency issues affecting internal corporate govern-
ance. The recourse by the Commission to non-binding instruments,
such as the 2004, 2005 and 2009 Recommendations examined in this
chapter, confirms this approach (Armour and Ringe 2011).
2.1. Market solutions

Competition in the product and factor markets may reduce the most
serious agency costs, to the extent that inefficient firms do not survive. In
other words, competition has a disciplinary function, pushing firms and
managerial teams to seek efficient performance (Fama 1980).14 Financial
markets also play a role in reducing agency costs. A firm tapping the
market for new resources is subject to the scrutiny of potential investors.
Therefore, it issues new information about its current management and
perspectives and possibly about corporate governance arrangements. In
addition, market prices generate incentives to value maximisation. If
agency costs are perceived as low by investors, the price of the firm’s
securities will be enhanced. Furthermore, well-developed financial mar-
kets allow re-packaging expected cash-flows and restructuring the set of
financing contracts, so as to minimise agency costs (Barnea et al. 1981).15
The market for corporate control concurs to reduce agency costs. Both
theory and evidence support the idea that hostile takeovers may solve
governance problems (Manne 1965; Jensen 1988; Scharfstein 1988).
14
The true extent to which agency costs are limited by product markets is disputed. Jensen
and Meckling (1976) argue that: ‘If my competitors all incur agency costs equal to or
greater than mine I will not be eliminated from the market by their competition’.
Jagannathan and Srinivasan (2000) produce evidence consistent with a disciplinary
role of competition in product markets.
15
Financial structure decisions may reflect the relative pros and cons of debt and equity in
controlling conflicts of interest: debt is more appropriate where free cash flow produc-
tion is high, since it forces management to seek approval (and re-financing) for new
investment projects; on the opposite, equity financing is more appropriate where free
cash flow production is lower (and/or unstable), since the risk of leniency in corporate
decisions is naturally lower and lower leverage allows to reduce the risk of costly
bankruptcy. An inefficient financial structure (implying higher than necessary agency
costs) may be easily restructured by the firm’s management or by a large investor buying
out – at market prices – all the securities issued by the firm, which could then switch to
the most efficient solution.
Takeover targets are often poorly performing firms, and their managers
are removed once the takeover succeeds. A different view, focusing on
the UK, argues that hostile takeovers are not so much about correcting
poor performance, but changing the strategy of middle-of-the-road
performers, so that they become top performers (Franks and Mayer
1996; Mayer 2013). In general, unfriendly takeovers are widely seen as
a corporate governance mechanism directed to control managerial dis-
cretion where ownership is dispersed (Easterbrook and Fischel 1991). At
the same time, bidder decisions may also be affected by agency problems
(Masulis et al. 2007), while hostile takeovers may transfer wealth from
stakeholders to shareholders of target firms (Shleifer and Summers
1988). However, in corporate systems like those prevalent in continental
Europe, where controlling shareholders are often the norm in listed
companies, the role of hostile takeovers is naturally limited, while man-
datory bids contribute to protecting minority investors by granting
the same a right of exit in change of control situations (Ferrarini and
Miller 2010).
The market for managerial labour may also play an important role, for
individual managers are disciplined by competition from within and
outside the firm. Compensation packages for managers, both incumbent
and recruited on the job market, represent a market price for their
services. If remuneration fully reflected a manager’s past/expected per-
formance, including possible misbehaviour, the value of human capital
would be adjusted accordingly and the moral hazard problem would
disappear (Jensen and Meckling 1976). However, the managerial labour
market does not exert this disciplinary role perfectly, especially when
managers have a short residual work life (Fama 1980). Moreover, the
idea that remuneration is the result of arm’s length contracting has been
criticised recently, to the extent that the setting of pay may be influenced
by the executives through capturing the board or as a result of informa-
tion asymmetry (Bebchuk et al. 2002; Bebchuk and Fried 2004), espe-
cially where shareholders are weak and dispersed (see Sections 3 and 5
below).
2.2. Corporate law

Market solutions do not eliminate agency problems altogether in the real
world because financial, product and labour markets are not perfectly
information-efficient. If prices do not incorporate the information avail-
able to individual agents in a timely and correct manner, they will not
provide a complete solution to agency problems (Barnea et al. 1981).

Indeed, distorted prices produce a distorted set of incentives, aggravating
the agency problems that the relevant markets would otherwise reduce.
As we shall see in Section 5, for instance, CEO incentive pay packages
may give rise to agency problems rather than reducing the ones that they
were intended to cure.
Similar market failures explain why corporate law affords protection
to outside investors, such as shareholders and creditors. In general,
corporate law sets the requirements and limits to contracts that may be
entered into by private parties. State powers are also available to the same
parties for enforcing contractual performance and/or the collection of
damages for non-performance. All of this affects both the kinds of
contracts that are executed and the extent to which contracting is relied
upon (Jensen and Meckling 1976). Of course, this mechanism is more
effective to the extent that the contracts at issue are ‘complete.’
However, legal protection may go far beyond guaranteeing compli-
ance with contractual clauses explicitly stipulated by the parties (Armour
et al. 2009b). Mandatory rules requiring or prohibiting some types of
agents’ behaviour may be dictated (Coffee 1989; Gordon 1989). As a
result, agents and principals do not need to negotiate detailed provisions
in their contracts, and transaction costs are minimised. Furthermore,
when discretion is given to an agent, the law offers standards (rather than
rules) against which the agents’ behaviour will be adjudicated ex post
(Kaplow 1992). Fiduciary duties – like the duty of care and that of
loyalty, as specified by the courts – provide a set of incentives even in
the absence of a contractual clause (Easterbrook and Fischel 1991).16
Nonetheless, there are limits to corporate law as a mechanism for
controlling agency costs other than ‘stealing.’ No doubt, well-structured
(and thoroughly enforced) corporate law provisions may deter control-
ling shareholders from diverting value to themselves and managers from
putting firm assets into their own pockets. However, corporate law is less
effective in preventing the sheer mismanagement of corporate resources
(‘shirking’) (Roe 2002). The US business judgment rule and its equiva-
lents in European jurisdictions typically insulate directors and manag-
ers from courts’ interference, absent fraud or conflict of interest,
16
The protection afforded by legal standards of conduct is lower than that offered by rules.
Since standards are general, their enforcement is problematic. Their aggressive enforce-
ment may discourage risk taking and favour conformism, ultimately damaging the
principles.
exempting them from ex post legal scrutiny (Hopt 2011).17 Indeed,

courts regularly second-guessing managers would be a cure worse than
the disease, for judges are generally less informed than managers and
lack the incentives to take business decisions. Furthermore, judges may
be affected by hindsight bias, finding reckless ex post managerial conduct
which was perfectly reasonable when performed. As a result, if system-
atic review of business decisions by the courts were permitted, the
incentives inherent to agency relationships would inevitably be distorted
(Rock and Wachter 2001).
2.3. ‘Comply or explain’

Since informational asymmetry characterises agency relationships, dis-
closure is crucial for controlling the related costs. Ex ante disclosure
allows prospective principals to select agents on the basis of their intrin-
sic qualities and to better decide on which terms the agency relationship
should be entered into. Ex post disclosure is crucial for enforcement, as
principals can better detect contract violations, or deviations from the
expected standards of conduct. Even in the absence of a breach, informed
principals can revise their expectations about the risks and rewards of the
agency relationship and take appropriate actions (Mahoney 1995).
The ‘comply-or-explain’ principle – which is widely applied in Europe
and was harmonised under the 2006 Directive cited above (Section
1.2.1.) – reflects this ‘governance’ function of disclosure (Kraakman
2004). Listed companies must state whether they apply a corporate
governance code, specify if they comply with its provisions and, in case
of non-compliance, explain the reason for their choice. The need for
disclosure, combined with obvious reputational concerns – most firms
want to appear good at corporate governance, or at least do not want to
appear non-compliant with best practices – push companies to comply
with a code that, however, remains voluntary in nature. Therefore,
disclosure performs a ‘legislative’ function by lending support to soft
law and its enforcement (ibid.) At the same time, the flexibility of soft law
is protected, to the extent that a code can be easily displaced, provided
17
In Germany the business judgement rule is embodied in statute law rather than being
solely a creation of the courts (as in the US and other countries). In countries like the
UK, the business judgement rule is not stated explicitly, but seems to emerge from the
courts’ lack of willingness to review management business decisions in the absence of
conflicts of interest (Enriques et al. 2009).
that a motivation is given for non-compliance with one or more of its

provisions.
These and other key issues of corporate governance codes are analysed
by Wymeersch in Chapter 2. Corporate governance codes are usually the
outcome of ‘private’ initiatives. At the same time, they respond to the
public interest and are considered as an alternative to public regulation.
However, codes reflect mainly the concerns of business leaders, address-
ing issues confronted by the same as board members or vis-à-vis share-
holders. Their business bias may explain the declining trust in corporate
governance codes by the political world, save for cases in which the codes
are promoted by securities regulators or under the aegis of governments.
In a few jurisdictions, two layers of recommendations or codes have been
adopted for boards and shareholders, respectively, the latter referring
particularly to institutional investors (an issue that will be further con-
sidered at Section 6.2.3 below).
The ‘comply or explain’ principle is ambiguous and has stirred debate
with respect to its place in the legal system. Wymeersch favours a broad
interpretation, arguing that a company should explain if and how it
complies with a corporate governance code and, in the case of non-
compliance, give the reasons for this and the solutions adopted as an
alternative. The principle at issue caters to the private autonomy of
companies. As a result, some of the main pillars of today’s corporate
governance – such as independent directors, audit committees and lead
directors – derive from corporate governance best practices rather than
regulation. However, the same freedom makes the code system fragile.
Much depends on what explanation is deemed as ‘proper’ in a given
system. Company reports frequently include boilerplate explanations,
carried over from year to year; however, a similar practice should be
rejected. In several jurisdictions guidelines exist about the appropriate-
ness of an explanation.
In most countries, certain entities systematically analyse corporate
governance statements. The nature of these monitoring bodies and
the scope of their action differ considerably. Usually the substance of
disclosure and explanations are not verified, as this would require ques-
tioning corporate boards and analysing the reasons given for non-
compliance. As a result, monitoring is generally limited to statistical
analysis and comments. Moreover, individual breaches and the
company’s identity are kept confidential, generally for fear of commit-
ting libel and slander. Moreover, publication of the breaches per se could
be considered as a sanction, triggering human rights concerns. On the
other side, public authorities are reluctant to lend their assistance to

enforcement of the codes, which are private in nature.
Wymeersch concludes that further Europe-wide harmonisation is
problematic. Different ownership and governance structures, as well as
different legal regimes, counsel avoiding a uniform approach to corpor-
ate governance issues. Rather, corporate governance commissions
should better explore how they can learn from each other and possibly
align their recommendations and terminology. At the same time, com-
panies should streamline their governance practices and disclosures.
European business associations could usefully support the convergence
of best practices. High-level principles, reflecting the common denom-
inator amongst best practices, might then be developed, but the national
standard setters should remain free to adopt only those which fit best to
their legal order.
2.4. Bank prudential regulation

As already noted, banks are different from other enterprises to the extent
that even ‘good’ corporate governance (that is, aligning the interests of
managers and shareholders) may lead bank managers to engage in more
risky activities. Given high leverage and other special features of banks, a
major part of the losses would be externalised to stakeholders, while
gains would be fully internalised by shareholders and managers (if
properly aligned by the right incentives). As a result, prudential regu-
lation and supervision aim to reduce the excessive risk propensity of
shareholders and managers in view of guaranteeing the safety and
soundness of banks. In a similar framework, corporate governance
works as a complement to prudential regulation by contributing to
keep risk management under control. This explains why banking super-
visors have become so interested in corporate governance in the last
decade (Basel Committee 2010).
By fixing the standards under which bank boards should operate in
their monitoring activities vis-à-vis the managers and by supervising
their implementation in practice, bank regulators indirectly control risk
taking by banks and assure their safety and soundness. As a result, the
corporate governance of banks (and financial institutions in general) is
clearly directed not only to maximise shareholders’ wealth, but also to
protect the interests of depositors (and other stakeholders) and to pre-
vent systemic risk in all cases in which these could materialise (large
institutions, interconnected ones, etc.) (Becht et al. 2012). As underlined
by the 2010 Green Paper, ‘it is therefore the responsibility of the board of
directors, under the supervision of the shareholders, to set the tone and
in particular to define the strategy, risk profile and appetite for risk of the
institutions it is governing’.
3. Ownership
The differences in ownership structures amongst listed companies in
Europe need to be emphasised: diffuse shareholders are prevalent in the
UK and Ireland, while controlling shareholders are the norm in other
countries (Barca and Becht 2001; McCahery et al. 2002; Gordon and
Roe 2004). The importance of these differences on regulatory grounds is
highlighted by the 2011 Green Paper, where the European Commission
discusses the issue of shareholder ‘engagement’ – which is understood as
engaging in a dialogue with the company’s board and using shareholder
rights to improve the governance of the investee company – mainly with
reference to diffuse ownership companies. The Commission then intro-
duces the topic of minority shareholder protection by stating that
‘minority shareholder engagement is difficult in companies with con-
trolling shareholders, which remain the predominant governance model
in European companies’. The Commission also comments that similar
difficulties may make the ‘comply or explain’ mechanism much less
effective, hypothesising that legal rules may be needed for either reserv-
ing some of the board seats to minority shareholders (a theme analysed
in Chapter 8) or controlling related party transactions.
In this Section, after sketching the different types of agency problems
deriving from the two main ownership structures, we consider the
special case of family companies, which show interesting dissimilarities
from other companies controlled by non-family blockholders.
3.1. Dispersed v. concentrated ownership

Agency problems may arise either between managers and shareholders
(as a class) or between controlling shareholders (as agents) and minority
shareholders (as principals). When shareholders are dispersed, an
appropriate set of constraints is required to guarantee that self-interested
managers – who have discretion over the allocation of the company’s
resources – act primarily in the shareholders’ interest. Alternatively, one
or more investors may acquire a large equity stake (Shleifer and Vishny
1997). The ensuing concentration of claims makes concerted action
amongst investors easier, given that transaction costs are reduced, while
blockholders are entitled to a higher (proportionate) share of the
expected benefits. However, the interests of blockholders are not always
aligned with those of the remaining investors. Indeed, the dominant
shareholders (and the managers appointed by the same) may use their
discretion to expropriate minority investors and get a disproportionate
share of the firm’s benefits.
Ownership structures vary across countries and firms. In the UK, US
and other common law countries, ownership is typically dispersed and
separate from control (La Porta et al. 1999). In the rest of the world, large
shareholdings of some kind are the norm: ownership is typically con-
centrated in the hands of families and the State (Claessens et al. 2000;
Becht and Mayer 2001; Faccio and Lang 2002).18 Consequently, different
countries generally witness different kinds of agency problems (Roe
2005).
A third category of agency costs may be identified with regard to the
relationship between the controllers of a company (as agents) and non-
shareholder stakeholders (Armour et al. 2009b; ECLE 2011). However,
not all relationships of this kind are easily defined in terms of agency.19
While debt contracts fit an agency perspective, the same cannot be said
for other relationships such as those with the firm’s clients or local
communities. Nonetheless, contracts with stakeholders and the applic-
able regulatory framework may have an impact on corporate governance
to the extent that the relevant prohibitions and/or obligations directly
or indirectly affect the firm’s directors and shareholders (Braithwaite
2008).
The interaction between ownership structures and total agency costs is
widely discussed in the economic literature. According to some scholars (La
Porta et al. 1997; 1998; 2000), ownership concentration leads to suboptimal
diversification. When a firm goes public, the founder should therefore
relinquish control altogether, provided that institutions are available for
18
Precise numbers may vary according to sample size, reference years and methodology of
analysis. However, a clear distinction may be traced between the UK, US and a handful
of other countries, on one hand, where the average (or median) largest shareholding
block is below the conventional 10% threshold, and continental European (and Asian)
countries, where the average (or median) largest block is much higher (between 25% and
50%) and allows control of the decisions of the general meeting.
19
Jensen and Meckling (1976) define an agency relationship ‘as a contract under which
one or more persons (the principal(s)) engage another person (the agent) to perform
some service on their behalf which involves delegating some decision making authority
to the agent’.
keeping managerial agency costs under control. Consequently, good

investor protection leads to both ownership dispersion and higher firm
values. This ‘law matters’ theory of corporate governance has been criticised
from different perspectives. First, the underlying legal analysis and the
measures of investor protection adopted are not always accurate (Cools
2005; Armour et al. 2009a). Second, the theory at issue does not fit the
evidence available for a number of countries (Cheffins 2001; Coffee 2001;
Dyck and Zingales 2002; Gilson 2006). Third, it is unclear whether share-
holding blocks persist in some institutional contexts because minority
shareholders fear the controlling ones or because they fear the managers,
who might dissipate shareholder value if the controlling shareholders dis-
appeared (Roe 2002).
Briefly, an optimal ownership structure is not easily found, which may
be due to the complexities of the ‘common agency’ problem. It has also
been argued that ownership structures, as well as corporate governance
institutions in general, are path-dependent (Bebchuk and Roe 1999), i.e.
their pros and cons may depend on a country’s existing pattern of
corporate structures and institutions.20 Therefore, the optimal solution
to the ‘common agency’ problem may be country-specific, when not
specific to the individual firm.
In all cases, ownership structure decisions involve a choice between
alternative sets of agency problems. The same is true for institutions
aimed at keeping these problems under control. A given mechanism may
mitigate one type of agency problem, but reinforce another: for instance,
entitling shareholders to remove the managers may mitigate the agency
problems of shareholders as a class, but reinforce those of minority
shareholders (ECLE 2011).
3.2. The case of family firms

In Chapter 3, Andres, Caprio and Croci analyse how family-controlled
firms compare with non-family firms in responding to crises. On one
side, they confirm what has already been acknowledged, i.e. that family
firms in Europe generally outperform non-family firms (Barontini and
Caprio 2006; Maury 2006; Sraer and Thesmar 2007; Andres 2008; Franks
et al. 2012). On the other, they provide new information on the ways in
which family firms behave in booms and busts. Their findings are in
20
Which may also include historical accidents, due to non-CG factors, such as wars,
upheavals and other less dramatic ‘political’ influences (Morck and Steier 2005).
stark contrast with the private benefits hypothesis, which assumes that
ownership remains concentrated in the hands of families where low
investor protection allows the same to extract higher private benefits.
Andres, Caprio and Croci show that family firms react to downturns
more efficiently than non-family firms, as the former adjust their invest-
ment decisions more quickly. They also show that the engagement of
long-term investors does not necessarily produce more stable perform-
ance and investments, contrary to what is assumed by most literature.
The better performance of family firms derives from their more efficient
investment policy, which includes rapid downsizing in crises. Moreover,
family firms apparently do not take advantage from a crisis to expro-
priate minority investors.
Andres, Caprio and Croci find evidence that family firms reacted to
the credit crisis by reducing their workforce and wages. This could imply
the break-up of long-term implicit contracts with employees and a
possible wealth transfer from labour to shareholders (Shleifer and
Summers 1988). Similar adjustments would be more difficult to carry
out quickly if employees owned a significant fraction of the equity capital
and/or if they were represented on the board of directors. Employees’
ownership and/or board membership, despite being deeply rooted in
some Member States, may work as a double-edged sword during crises.
On one hand, they could lead to a smoother transition; on the other, they
could prevent or slow down the restructuring of ailing firms.
These results suggest that ownership structures in different countries
may be determined by causes other than by the degree of investor
protection prevalent in each country. The complexity of corporate gov-
ernance arrangements can scarcely be captured by a simple measure of
investor protection or a ‘governance index’. Moreover, similar arrange-
ments, in order to be effective, should fit the underlying legal and
institutional structure, rather than be dictated by the same. The simple
transplant of corporate governance solutions may be ineffective and
could even backfire, where the regulatory and institutional contexts are
not receptive.
4. Boards
Whatever the firm’s ownership structure, both the markets and corpor-
ate law provide incomplete solutions to agency problems, which are too
complex to be solved solely through ex ante mechanisms. Discretionary
powers, which are the essence of agency relationships, survive in a world
of incomplete contracts. This leaves room for governance mechanisms

allocating decisional powers ex post, i.e. after the contract has been
stipulated, in a state-contingent manner. These governance mechanisms
are characterised by flexibility, for they allow new information generated
after the making of the corporate contract to be exploited in the manage-
ment of the firm (Williamson 1988). The first mechanism of this type is
the board of directors which, in the two-tier system of governance
foreseen in some European countries, finds its equivalent in the super-
visory board.
4.1. Theory
Given contractual incompleteness, the (supervisory) board is entrusted
with the required discretion to take the core business decisions and
monitor the managers on behalf of the shareholders (and possibly
other stakeholders). Boards are found in all jurisdictions and all types
of organisations (profit and non-profit), and were generally developed
before specific legal provisions were introduced to regulate them. Boards
can therefore be regarded as a market solution to agency problems, i.e. an
endogenously determined institution that helps keep agency costs under
control (Hermalin and Weisbach 2003).
Board discretion covers the monitoring of managerial actions and the
taking of high-profile decisions, which should not be left to the managers
alone. In Williamson’s (2008) words, boards are meant to ‘serve as
vigilant monitors and as active participants in the management of the
corporation’. Their monitoring regards corporate organisation and
management performance. It also includes the ‘hiring and firing’ of the
CEO and other key executives and the setting of their incentives and
compensation packages. The monitoring extends to the information
flows to investors, such as financial statements, event-related price-
sensitive information, etc. The board’s management role mainly relates
to fundamental corporate actions, such as the approval of major business
transactions and of corporate strategy and relevant plans. Other board
roles are the offering of advice to the managers and networking with
other firms and institutions.
The board’s appointment gives rise to a discrete agency relationship
under which agents (directors) monitor other agents (managers) and to
the ensuing conflicts of interest. Nonetheless, the board is usually con-
sidered a successful governance mechanism because of its collegial
nature, which increases the information set collectively available to the
monitors and grants superior decision-making under a number of cir-

cumstances (Bainbridge 2002). Board collegiality also makes bribery of
delegated monitors more expensive and easier to detect (Hermalin and
Weisbach 2003).
The theoretical framework for the board as a governance mechanism
is straightforward. Distant shareholders lack information and focus.
They could not run the company directly, truly understand its business,
select and motivate the CEO. Rather, they entrust the board to direct the
company’s business, hire the executives and delegate day-to-day man-
agement to the same. However, problems often arise in practice, for the
board may be captive to senior managers and/or controlling sharehold-
ers. In companies with dispersed shareholders, the CEO may influence
the selection of board candidates and easily dominate the board by
controlling information flows. In controlled companies, majority share-
holders appoint the board and can either dismiss or simply not renew the
appointment of directors who do not follow their directives.
That boards may depart in practice from their theoretical model
should not lead to their replacement as alternatives – such as direct
monitoring by investors – would likely cause a net loss, at least in general
(Williamson 2008). Moreover, in egregious cases of underperformance,
the market for corporate control already allows investors to replace the
board through a hostile takeover.
4.2. Practice
The (supervisory) board is widely accepted as a governance mechanism
and presents common features internationally. However, its composi-
tion and structure, and the allocation of powers between the board and
the general meeting of shareholders differ across countries and change
over time. Boards rely on either non-executive or supervisory directors,
depending on whether they reflect a one- or two-tier board structure
(Hopt 2011). Some board members must also comply with independ-
ence requirements – aiming to assure their objectivity of judgment – and
with certain professional requirements, particularly accounting and
financial experience (Gordon 2007). More recently, board composition
requirements were introduced in some countries to promote gender
diversity. In general, the organisation of boards greatly improved over
the last twenty years, after the Cadbury Report in the UK marked the soft
law approach to corporate governance reform, which was also followed
in Continental Europe (Weil Gotshal & Manges 2002). Whether
organisational reforms also translated into effective improvements of

boards’ functioning is still an open question, one that it is difficult to
answer in general terms (Williamson 2008).
The main criteria embodied in European corporate governance codes
concerning the composition and organisation of boards (both one- and
two-tier) are usefully summarised in the Commission Recommendation
on the role of non-executive or supervisory directors of listed companies
and on the committees of the (supervisory) board.21 Section II of the
Recommendation deals with the presence and role of non-executive
(supervisory) directors on (supervisory) boards. One of the core criteria
is that boards should have an appropriate balance of executive (manag-
ing) and non-executive (supervisory) directors, so that no individual or
small group of individuals can dominate decision-making within the
relevant bodies. Another criterion is that a sufficient number of inde-
pendent non-executive (supervisory) directors should be elected to the
(supervisory) board of companies, to ensure that any material conflict of
interest involving directors would be properly dealt with.
Furthermore, boards should be organised in such a way that a suffi-
cient number of independent non-executive (supervisory) directors play
an effective role in key areas where the potential for conflict of interest is
particularly high. To this end, nomination, remuneration and audit
committees should be created within the (supervisory) board, where
that board plays a role in the areas of nomination, remuneration and
audit under national law. Interestingly, every year the (supervisory)
board should carry out an evaluation of its own performance, which
should encompass an assessment of its membership, organisation and
operation as a group, an evaluation of the competence and effectiveness
of each board member and of the board committees, and an assessment
of how well the board has performed against any performance objectives
which have been set.
Section III of the Recommendation deals with the profile of non-
executive (supervisory) directors, including their qualifications and
independence. The (supervisory) board should determine its desired
composition in relation to the company’s structure and activities, and
evaluate this periodically. The (supervisory) board should ensure that it
is composed of members who, as a whole, have the required diversity of
knowledge, judgement and experience to accomplish their tasks prop-
erly. The members of the audit committee should, collectively, have a
21
See n. 6 above and accompanying text.
recent and relevant background and experience in finance and accounting

in listed companies appropriate to the company’s activities. As to inde-
pendence, a director should be considered independent only if he is free of
any business, family or other relationship with the company, its controlling
shareholder or the management of either, that could otherwise create
conflicts of interest such as to impair his judgement. A number of criteria
for the assessment of the independence of directors are indicated in the
guidance set out in Annex II to the Recommendation. However, the
determination of what constitutes independence is fundamentally an issue
for the (supervisory) board itself to determine.
The 2011 Green Paper specifically addressed the issues of professional,
gender and international diversity on the board. The Feedback Statement
on the consultation22 shows that most respondents agreed on the import-
ance of board diversity; however, no consensus seems to exist on the
need to increase diversity through regulation (or on the most appropri-
ate regulatory instruments). In particular, the majority of respondents
rejected the idea of listed companies being required to ensure a better
gender balance in boards. Nonetheless, gender diversity has been
addressed by a recent Directive proposal setting, for the under-
represented sex, of a minimum target of 40 per cent by 2020 as to the
non-executive directors of publicly listed companies (2018 for those of
listed public undertakings).23 Board diversity and its possible effects on
board effectiveness, firm value and performance are still a controversial
issue (Carter et al. 2003; Adams and Ferreira 2009; Ahern and Dittmar
2012); the evidence available so far has proved inconclusive and is often
plagued by endogeneity problems (Adams et al. 2010).
4.3. Empirical analysis

In Chapter 4 Ferreira and Kirchmaier analyse a cross-section of
board characteristics – board size, independence and gender diversity –
across Europe. Country characteristics explain a significant part of the
variation in board independence and gender diversity, suggesting that
22
See Feedback Statement Summary of Responses to the Commission Green Paper on the
EU Corporate Governance Framework, November 2011, available on the Commission’s
website, www.ec.europa.eu.
23
See the Proposal for a Directive of the European Parliament and of the Council on
improving the gender balance among non-executive directors of companies listed on
stock exchanges and related measures available on the Commission’s website, www.ec.
europa.eu.
country-level governance rules play an important role in the determin-

ation of these variables. In contrast, board size is mostly explained by
firm and industry characteristics. Differences in board structure between
European and US firms are seemingly persistent. In the US almost three
out of four directors are independent, while in Europe (particularly
Continental Europe) only a minority of directors are independent.
This might be associated, at least in part, with differences in ownership
structure.
Board characteristics are not necessarily stable over time. European
firms have reduced the size and increased independence of their boards
over the last decade. This is partly due to a composition effect, since the
coverage of small firms by commercial databases used for research
purposes has increased over time. However, Ferreira and Kirchmaier
show that European firms that performed poorly during the crisis chose
to reduce both the size and independence of their boards. This might be
a response to the evidence relating pro-shareholder boards in financial
firms to poor performance in the crisis, which clearly questions the
conventional wisdom that more board independence is always benefi-
cial. Listed firms appear to consider the (changing) wishes of investors
when proposing new board candidates, with the result that solutions to
corporate governance issues evolve over time. A related question is
whether board composition (like other issues of corporate governance)
may be subject to fashions and fads, having less to do with value
creation than with the influence of consultants and ‘experts’ responding
to their own incentives. Further research could usefully address this
aspect.
Board gender diversity is a fairly recent topic in the policy debate.
Ferreira and Kirchmaier show that, despite being on the rise, diversity is
still limited in Europe, except for countries having quota-based regula-
tion of board membership. In fact, regulation is the single most import-
ant factor explaining differences in board gender diversity across
European countries. However, it is unclear whether gender diversity is
a corporate governance issue or one concerning equal opportunities.
More diverse boards are not clearly superior in terms of economic
performance. In the case of superior performance, it is difficult to
establish whether diversity was the cause or the effect thereof. Ferreira
and Kirchmaier show that cultural norms are correlated with female
participation in non-executive positions even after controlling for labour
force participation. This raises the question of whether policies targeting
boardroom diversity directly are sustainable in the long run.
Furthermore, while protection of equal opportunities can hardly be

disputed, it is difficult to see what is special about boards (of listed
companies). Why should shareholders (and boards) be trusted to make
the correct (i.e. value-maximizing) choices in all respects but for gender
diversity? A claim that boards are entrenched would have implications
going far beyond gender diversity. From another viewpoint, why does
equal access deserve more protection in board elections of business
corporations than in other areas of human activity (such as non-profit
institutions, regulatory agencies, parliaments, etc.)?
4.4. Limits of a quantitative approach

Of course, all theoretical models, including board models, form an
incomplete description of reality. Directors in particular have a com-
plex objective-function, which may be only partially described
through theoretical (agency) models. The risk of overestimating the
contribution provided by ‘hard’ sciences to the analysis of human
interaction is always present. In a similar vein, Winter and van de
Loo argue in Chapter 5 that lawyers and (financial) economists often
follow a narrow approach to the functioning of boards, which cannot
fully explain board performance in reality and the factors determin-
ing it. A comprehensive and integrated approach touching upon
behavioural aspects may better describe boards as social institutions,
i.e. as an organisation through which people co-operate, debate and
take decisions in view of certain corporate objectives. The behaviour
of directors, either as individuals or as a group, unavoidably affects
board performance.
Winter and van de Loo therefore suggest an alternative and broader
description of the interaction between executives and non-executives
within boards. They advance, in particular, a new concept – the board
‘on task’ – which should be used to understand and assess board per-
formance in practice, whilst opening new perspectives to research on
board performance. Winter and van de Loo also warn against the over-
confidence generated by current research on boards, for empirical ana-
lysis may suffer from incomplete data availability and the presence of
non-measurable factors. Noting that ‘actual board performance occurs
in a black box that cannot be observed by outsiders’, they recommend
caution in deriving policy implications from studies relying purely on
‘hard’ data (a recommendation that we no doubt follow in this volume
when formulating policy suggestions).
5. Incentive pay
While there are multiple characterisations of the executive pay question
(e.g. Loewenstein (1996) describing executive pay as a wealth transfer
issue), the dominant model examines executive pay in terms of the
principal/agent relationship and incentives. In this section, after briefly
analysing the main theories related to incentive pay and agency costs, we
review recent EU reform of executive pay, distinguishing between non-
financial firms and financial institutions (banks in particular) and sum-
marise the outcomes of the empirical study on pay practices at large
European companies described in Chapter 6.
5.1. Two views

The principal/agent model generates two competing views of executive
pay. According to the first, executive pay remedies the agency costs
generated by the misalignment of management and shareholder interests
in the dispersed ownership company. Shareholders in dispersed owner-
ship systems have only a fractional interest in the firm profits, are not
fully incentivised to discipline and have limited opportunities to monitor
management (Jensen and Meckling 1976). Management’s unobserved
actions, particularly where personally costly decisions (e.g. laying off
employees) and privately beneficial activities (e.g. consuming perqui-
sites) are involved, can prejudice shareholder wealth and give rise to
agency costs.
Whether, and the extent to which, a manager will fully pursue
shareholders’ interests depends on how that manager is incentivised.
Agency theory suggests that the performance-based pay contract, which
links pay to shareholder wealth via performance indicators such as share
prices or accounting-based targets, is a powerful way of attracting,
retaining and motivating managers to pursue the shareholders’ agenda
(Jensen and Murphy 1990; 2004; Conyon and Leech 1994; Hall and
Liebman 1997). In the dispersed ownership context, this paradigm has
dominated the pay debate and pay practices since the early 1990s and
still enjoys considerable support as making management more sensitive
to shareholders’ interests (Holmstrom and Kaplan 2003; Kraakman
2004).
However, executive pay can also be regarded as an agency cost in itself,
providing a powerful and opaque device for self-dealing by conflicted
managers (Bebchuk et al. 2002; Hill and Yablon 2002; Bebchuk and Fried
2004). In practice, pay is not set by the shareholders; rather, it is set on

their behalf by the board of directors, which should align shareholder
and managerial incentives (Jensen and Murphy 2004). Nonetheless, a
conflicted board may use the pay-setting process to influence pay and
extract rents in the form of pay in excess of that which would be optimal
for shareholders, given weaknesses in the design of pay contracts and in
their supporting governance structures (Bebchuk et al. 2002; Bebchuk
and Fried 2004). In other words, executive pay raises an additional
agency problem: how can the effectiveness of the executive pay contract
as a remedy for manager/shareholder agency costs be protected from
conflicts between the board, as pay-setter, and shareholders (Jensen and
Murphy 2004)? The equity-based incentive contract may, as post-Enron
scholarship argues, deepen conflicts of interest between shareholders
and management by generating perverse management incentives to
manipulate financial disclosure (particularly earnings) and distort
share prices, which can lead to catastrophic corporate failures (Coffee
2002; Gordon 2002; Ribstein 2003). The consequences of such a cycle of
ever higher share prices and their impact on pay have been examined as
‘the agency costs of overvalued equity’ (Jensen and Murphy 2004).
The relationship between agency problems and the executive pay
incentive contract takes on an additional complexity in continental
European firms, characterised by concentrated shareholdings and
long-term shareholder commitment (Ferrarini and Moloney 2004;
2005). Here, incentives and conflicts change. Here, concentration of
control (possibly intensified by cross shareholdings, pyramidal owner-
ship structures, proxy voting by financial institutions connected to the
company, and voting pacts) recasts the agency problem which executive
pay is designed to resolve. The agency costs which trouble the dispersed
ownership company are reduced, as block-holding shareholders have
both incentives and resources to monitor managers effectively (Garrido
and Rojo 2003). As a result, there is less need for an incentive contract to
control the conflict between management and shareholder interests,
which is remedied by executive pay. There is also less probability of an
agency problem deriving from executive pay contracts.
In concentrated ownership conditions, however, different profiles arise,
which have varying implications for executive pay and the management of
conflicts of interest. Where a shareholder/owner manages the company,
the need for an incentive contract, in principle, recedes, as the owner
is incentivised by his own equity interest. Where a professional/outside
manager performs management activities for the owner-shareholder (who
may also be a non-executive director), the owner monitors the manager’s

performance directly, thereby reducing the need for an incentive con-
tract. Nonetheless, monitoring by the owner may not be sufficient, given
that not all actions taken by the professional manager are easily observ-
able, so that an incentive contract can further align managerial interests
with those of shareholders. Rather, protection may be needed to prevent
collusion between blockholders and managers on pay-bargaining, given
that controlling shareholders might overcompensate the managers to
reward their cooperation in the extraction of private benefits from the
company.
5.2. Non-financial firms

The 2004 Commission Recommendation was the EU’s first attempt to
address best practice with respect to pay governance. It used disclosure
and shareholder voice mechanisms to support efficient pay and recom-
mended disclosure of company pay policy and process, either in a
distinct remuneration report or in the annual report; detailed disclosure
concerning individual directors’ pay; a shareholder vote on company pay
policy, either binding or advisory; and prior approval of share-based
schemes. The Commission also recommended disclosing the mandate
and composition of the remuneration committee, and the names of the
external consultants whose services have been used in setting the remu-
neration policy. The role of the board, its independence in the pay
process and the creation of remuneration committees were addressed
in the 2005 Recommendation on the role of non-executive directors,
where the Commission highlighted remuneration as an area in which the
‘potential for conflict of interest is particularly high’.
Member States were entitled to adopt these two Recommendations
either through legislation or (as was generally the case) soft law, typically
represented by a corporate governance code and the related ‘comply or
explain’ principle.24 However, the effectiveness of the ‘comply or explain’
mechanisms with respect to executive remuneration and relevant
investor monitoring proved doubtful, also given the low levels of
24
European Commission, Report on the application by Member States of the EU of the
Commission Recommendation on directors’ remuneration (2007) (SEC(2007) 1022);
European Commission, Report on the application by the Member States of the EU of
the Commission Recommendation on the role of non-executive or supervisory directors of
listed companies and on the committees of the (supervisory) board (2007) (COM SEC
(2007) 1021).
conformity with the Recommendations at Europe’s largest companies

(Ferrarini et al. 2010). As a result, significant differences continued to
exist across Member States’ regulatory regimes and pay governance
practices. No doubt, all of this could not per se be taken as an argument
for legal reform. On one side, a failure of self-regulation would need to be
proven, showing that non-conformity concerns provisions which com-
panies should mandatorily follow, such as those on disclosure (as further
suggested in Section 7.4.). On the other, non-compliance could depend
on the fact that soft law might either not entirely match the needs of
companies and/or investors’ expectations, or require more time to be
fully assimilated by corporate practice.
Although switching some of the focus on pay structure, the
Commission’s 2009 Recommendation on directors’ pay at non-financial
companies identified major weaknesses in the existing disclosure prac-
tices, emphasising the need for the remuneration statement to be clear
and easily understandable and providing greater detail on how disclo-
sure of performance-related pay should be implemented. It also set
further related requirements, i.e. that an explanation be provided con-
cerning how performance criteria relate to firms’ long term interests and
that sufficient information be provided concerning termination pay-
ments, vesting and the peer groups on which the remuneration policy
is based.
Remuneration committees are also more extensively considered under
the 2009 Recommendation, which addresses their composition, role and
functioning. The criteria suggested are relatively non-controversial: the
committee should periodically review the remuneration policy for direc-
tors; exercise independent judgment and integrity; address conflicts of
interests concerning consultants by ensuring that they do not at the same
time advise the human resources department or the executive directors.
The committee should also report to the shareholders on its functions.
However, the implementation of the 2009 Recommendation in the
Member States was only partial, as argued by Ferrarini et al. (2010)
showing that firms’ disclosure levels still vary from country to country
and are strongly dependent on the existence of either regulations or best
practice guidelines in each Member State. Firms widely comply with
binding rules, but only partially follow guidelines. In the absence of
binding rules, firms appear reluctant to provide full disclosure of remu-
neration, particularly of the pay/performance link and termination pay-
ments. It is not easy to compare how Europe’s largest companies
approach executive pay and, in particular, performance conditions. As
to the governance process, while the remuneration committee is gener-

ally well established, composition problems (including independence)
remain and the ‘say on pay’ mechanism is underdeveloped.
The present discussion on remuneration at non-financial firms con-
cerns the alternative between mandatory rules and soft law, specifically
in the areas of pay disclosure, remuneration report and shareholders’
vote on pay. Supporters of mandatory disclosure argue that the same
contributes to establish a level playing field internationally, improving
comparability of information between companies in different member
states (Bhagat et al. 2008; Posner 2009). Standardisation of the format in
which disclosure is provided also supports better monitoring and pos-
itive externalities, given that it is difficult to compare remuneration
across companies (Ferrarini et al. 2010). On the other side, critics
claim that mandatory disclosure interferes with board decisions on
executive remuneration, affects the privacy of directors and could have
a ratcheting effect on remuneration levels.25
The European Commission’s Green Paper on the corporate govern-
ance framework specifically considers mandatory disclosure and ‘say
on pay’. In particular, the Commission investigates whether it should be
mandatory to put the remuneration policy and the remuneration report
to a shareholder vote (however, leaving open the question as to whether
this vote should be binding or advisory); and whether disclosure of a
company’s remuneration policy, the annual remuneration report and
individual remuneration of directors should be enacted by the Member
States through mandatory legislation. The feedback on the consulta-
tion26 shows a wide consensus for mandatory disclosure, while manda-
tory say-on-pay turns out to be more controversial; furthermore, many
respondents who are in favour of a mandatory shareholder vote add that
such vote should be advisory only.
5.3. Financial institutions (banks in particular)

The financial crisis has reshaped the context within which executive pay
at financial institutions is examined (Bebchuk and Spamann 2010;
Ferrarini 2012). First, the rescue of large banks by governments investing
taxpayers’ money enhanced public resentment against the excessive pay-
outs at the helm of international banks. As a consequence, executive pay
was drastically reduced and bonuses almost disappeared at financial
25 26
See Feedback Statement, n. 22 above. Ibid.
institutions rescued by the states, whilst compensation structures were

tightly regulated to avoid paying taxpayers’ money to undeserving execu-
tives (Ferrarini and Ungureanu 2010). Soon similar initiatives were also
voluntarily adopted by sound banks in an effort to pre-empt investors’
and authorities’ concerns for inappropriate risk management. Several
regulators extended the treatment originally conceived for bankers’ pay
at rescued institutions to all financial institutions. Second, the national
measures adopted by the governments rescuing banks in crisis led to the
generation of the international Financial Stability Board (FSB) Principles
and Standards on compensation practices at financial institutions
(Ferrarini and Ungureanu 2011).
The FSB issued these Principles following coordinated action by the
G20 governments, which rapidly responded to heavy political pressure
deriving, both domestically and internationally, from the financial crisis
and repeated bank failures. Through swift adoption, authorities intended
to show that reforms of the international financial system were timely
put in place with respect to executive compensation. The Principles are
addressed to ‘significant financial institutions’, which are considered to
deserve an internationally uniform regime. Some principles are not new
to the extent that they require a balanced pay structure and long-term
approach, alignment of pay with performance, independence of the pay-
setting process and disclosure of remuneration policies. Relatively new is
the emphasis on effective alignment of compensation with prudent risk-
taking and compensation practices that reduce employees’ incentives to
take excessive risk.
The Principles cover four main compensation areas: governance,
structure, disclosure and supervision. As to compensation governance,
they incorporate well-known best practices concerning the strategic and
supervisory role of the board. In addition, they reflect post-crisis empha-
sis on bank risk management and monitoring by the board of directors,
who should determine the risk appetite of the firm. They reiterate the
role of the remuneration committee, also requiring its liaison with the
risk committee to ensure compliance with the relevant requirements.
Compensation structures are considered along lines that, to a large
extent, reflect general best practices already adopted before the crisis.
While pre-crisis practices mainly emphasised the alignment of
managers’ incentives with shareholder wealth maximisation, the FSB
Principles break new ground by requiring financial institutions to align
compensation with prudent risk taking. Accordingly compensation
needs adjustment with all types of risk, including those considered
difficult to measure, such as liquidity risk, reputation risk, and capital

cost. Deferment of compensation, traditionally used as a retention
mechanism, has been introduced to make compensation pay-out sched-
ules sensitive to the time horizon of risks. Furthermore, the Principles
require that a substantial portion of variable compensation be awarded
in shares or share-linked instruments, as long as the same create incen-
tives aligned with long-term value creation and the time horizons of risk.
Malus and clawback mechanisms could further enable boards to reduce
or reclaim variable compensation paid on the basis of results that are
unrepresentative of the company performance over the long term or
later prove to have been misstated.
The Principles also consider ‘guaranteed’ bonuses as conflicting with
sound risk management and the pay-for-performance principle, whilst
severance packages should be related to performance achieved over time
and designed in a way that does not reward failure. While before the
crisis disclosure was seen as one of the main mechanisms for aligning
managers with shareholder interests, after the crisis remuneration dis-
closure is considered to benefit not only shareholders, but also other
stakeholders (e.g. creditors, employees, financial supervisors), at least in
financial institutions. The FSB Principles add new items of disclosure,
such as deferral, share-based incentives and criteria for risk adjustment.
Moreover, disclosure should identify the relevant risk-management and
control systems and facilitate the work of supervisors in this area. In the
case of a failure by a firm to implement ‘sound’ compensation policies,
prompt remedial action should be taken by supervisors, and appropriate
corrective measures should be adopted to offset any additional risk that
may result from non-compliance or partial compliance with the relevant
provisions.
The Principles represent a reasonable political compromise between
the various interests at stake in the area of compensation, incorporating
traditional criteria and adapting these to new circumstances which have
emerged from the financial crisis. They were implemented along differ-
ent models (Financial Stability Board 2010). In many jurisdictions, the
model includes a mix of regulation and supervisory oversight, with new
regulations often supported by supervisory guidance that illustrates how
the rules can be met. Other jurisdictions follow a primarily supervisory
approach to implementation, involving principles and guidance and the
associated supervisory reviews. To a great extent, legislative and regu-
latory responses depend on the type of equilibrium found in each
country between the different interests at stake. Where public criticism
of bankers and hostility to their remuneration practices are strong, the

risk of regulatory capture is lower and a tougher regime for executive pay
may emerge. Culture may contribute to similar outcomes, given that
high levels of executive pay are less tolerated in some countries (Levitt
2005; Posner 2009).
However, no domestic regulatory solution could be effective without
agreement at international level. One-sided reforms (i.e. adopted only by
some countries) do not prevent contagion from other countries choosing
not to regulate compensation at financial institutions. In addition, they
could jeopardise a country’s competitive position as a financial centre, by
determining a flow of financial firms’ headquarters and top managers to
other countries adopting a more liberal stance relative to executive
compensation (Ferrarini and Ungureanu 2011).
The EU adopted the FSB Principles through amendments to the
Capital Requirements Directive (CRD III), which took effect in
January 2011 and are further analysed in Chapter 6 of this volume.

In Chapter 6 Barontini, Bozzi, Ferrarini and Ungureanu analyse the
evolution of various aspects of remuneration structure, governance and
disclosure among large European firms (both financial and non-
financial). They show that the implementation of EU recommendations
concerning remuneration governance and disclosure (of both remuner-
ation policy and individual compensation) has increased over the last
few years. Compliance with European standards for governance varia-
bles (existence and independence of remuneration committees and
consultants), already relatively good before the crisis, has further
increased; general information about the remuneration policy and dis-
closure of individual pay (including also termination agreements) has
increased remarkably over time, with Italian, German and French firms
showing the fastest progress on disclosure practices, toward the best
practice model represented by the UK; information on other aspects
(forward-looking information and details of stock-based compensation)
is still lagging behind. The general picture shows, nonetheless, that
compliance is on the rise, and is significantly affected by firm size,
industry (higher in financial companies, which have been targeted by
specific regulations), ownership concentration (higher in companies
without a control blockholder) and country.
Barontini, Bozzi, Ferrarini and Ungureanu also analyse the dynamics

of the level and structure of remuneration packages (for the whole board
and for CEOs) before and after the crisis. They show that remuneration
packages are remarkably variable across countries, reflecting the differ-
ences in board structure and, possibly, the national job markets for
managerial talent. These differences may possibly be attributed, at least
in part, to differences of the institutional context, allowing segmentation
to persist. International mobility of CEOs is still relatively uncommon
across Europe.
Directors’ pay in Europe is lower than in the US, and makes less use of
stock-based compensation (25 per cent against 50 per cent in the US).
However, the variable remuneration of EU CEOs is definitely non-
trivial, amounting to 60 per cent of total compensation (decreasing to
54 per cent in 2010, due to the consistent reduction in the amount of
cash-based variable pay). The structure of CEO pay (in terms of the
relative weight of fixed/variable components) is affected by firm size,
growth opportunities and past firm performance. The recourse to stock-
based compensation is lower in closely held companies; this is consistent
with the hypothesis that stock-based incentives are less important where
control blockholders exert a monitoring role and also with blockholders
being more sensitive to the implicit cost (in terms of dilution) of new
share issues.
After the crisis, directors’ remuneration has decreased remarkably in
financial institutions (especially for the CEOs), while it has slightly
increased in non-financial companies. The decrease is substantially due
to the cash portion of variable compensation (bonuses), while other
components have remained more or less stable (this is true for both
fixed salary and stock-based compensation; however, stock grants are
apparently becoming more popular, at the expense of stock options).
The change in the proportion of incentives detected in 2010 for financial
firms seems to be only partially explained by the negative performance in
the 2007–10 period. It is totally plausible that a concurrent factor was the
pressure exerted on financial firms by the national and international
regulators for a rethinking of their compensation structure. In fact, the
changes observed in the pay structure of financial firms go in the
direction indicated by the regulators, pushing for an ‘adequate’ balance
of variable and fixed components and for a portion of variable compen-
sation being deferred and awarded, at least partially, in shares or share-
linked instruments.
6. Shareholder activism
According to Gillan and Starks (1998), a shareholder activist is ‘an
investor who tries to change the status quo through “voice”, without a
change in control of the firm’. Armour and Cheffins (2009) propose a
more specific definition of activism as ‘the exercise and enforcement of
rights by minority shareholders with the objective of enhancing share-
holder value over the long term’. In this section, we examine the main
strategies for shareholder activism. We then explore some key aspects of
the legal framework for shareholder activism and EU reform proposals.
Finally, we comment on the outcomes of empirical analysis in Chapters 7
and 8.
6.1. Types and role of activism

Activist strategies are by no means uniform.27 Individual investors
usually hold small equity stakes and submit proposals to the general
meeting concerning governance (including social responsibility) issues.
Institutional investors are diverse and track different trading styles and
regulatory models. Consequently, they respond to different incentives
and skills as to active monitoring. While mutual fund and public pension
fund activism tends to be incidental and ex post (Kahan and Rock
2007),28 hedge fund activism is strategic and ex ante. Hedge fund man-
agers first determine whether a company would benefit from activism,
then take a position and become active. They prefer short-time strat-
egies, possibly including a public challenge to management. Hedge fund
activism is often labelled as ‘offensive’, while the strategies of other
institutional investors are defined as ‘defensive’ (Armour and Cheffins
2009). The incentives of active investors are not necessarily aligned with
those of shareholders as a class. While pension funds are criticised for
not being sufficiently interventionist, hedge funds are sometimes
criticised for being too active (ECLE 2011) or for acting for the ‘wrong’
reason.
27
The identity of shareholder activists and the focus of their efforts changed over time.
Until the end of the 1970s, shareholder activism was mainly practised by individual
investors. The 1980s saw a mounting involvement of institutional investors (public
pension funds, in particular) and corporate raiders. Starting in the 1990s, hedge funds
have taken the lead (Gillan and Starks 1998; 2007).
28
These institutions are long-period investors which prefer quiet negotiations with com-
pany management to high-profile initiatives (Becht et al. 2009).
Activism interferes with corporate decision-making and, in a sense,

contradicts delegation and specialisation. The decision to become active
should be based on a comparison between expected costs and benefits.
Since costs are predictable,29 while benefits are uncertain and limited
(since insurgents will receive only a fraction of the improvements in
shareholder returns generated by their efforts), activism is the exception
rather than the rule (Easterbrook and Fischel 1991).
The role of activism in controlling agency costs is controversial.
Bebchuk (2005) and Harris and Raviv (2008) claim that proxy proposals
by active shareholders mitigate managerial agency problems. Other
scholars argue that activists lack the capacity and ability to engage in,
or even evaluate corporate decision-making, which should be the sole
responsibility of the board (Woidtke 2002; Prevost et al. 2009). In the
wake of the financial crisis, active investors are often considered as a
possible complement or substitute for other corporate governance insti-
tutions (such as boards of directors and takeovers) in controlling agency
costs. Gilson and Gordon (2011) convincingly argue, however, that
‘specialisation’ is needed to develop the skills required by activism and
overcome the problems of what they define as ‘agency capitalism’.30 In
particular, a new set of actors is required to complement the diversified
investing and portfolio optimisation that institutional investors engage
in. These actors would develop the skills to identify governance short-
falls, acquire a position in a company, and then present to ‘reticent
institutions’ their value proposition: ‘the institutions will vote in favour
of the specialised actors perspective if the issue is framed in a compelling
way. From this perspective, the overall obligation to beneficial owners is
split between the portfolio management undertaken by institutional
29
Activism implies both direct and indirect costs. The former relate to the time spent by
senior executives and the out-of-pocket expenses for the selection of board candidates,
coordination with other shareholders, proxy solicitation and other campaign efforts.
Indirect costs are less visible, but nonetheless substantial, and include limitations to
trading implied by market abuse regulation, suboptimal diversification (where activism
requires a large and/or long-term investment in the company), legal liability for acting in
concert and potential litigation costs (Pozen 2003).
30
Gilson and Gordon (2011) argue that investment managers have no private incentive
proactively to address governance and performance problems, and therefore do not
engage in that activity, even if it would benefit their beneficiaries. This gap between the
clients’ and the fund’s interests represents an agency cost that locks in another agency
cost: managerial slack at the portfolio companies. Together these are the ‘agency costs of
agency capitalism’, which ‘result in the chronic undervaluation of governance rights’.
investors, and the active monitoring of portfolio company strategy and

execution undertaken by activist investors’ (Gilson and Gordon 2011).
6.2. Regulatory impact and reform proposals

Regulation may favour or hinder activism. Roe (1994) claims that share-
holder apathy in the US is largely due to limitations imposed on institu-
tional investors, notably under the rules on ‘acting in concert’. Rules
applicable to shareholder participation and voting in general meetings
may also affect investors’ behaviour. In this section, we analyse similar
rules included in the Shareholder Rights Directive and the national
regimes concerning the division of powers between boards and share-
holders. We then briefly consider EU policy perspectives as to share-
holder ‘engagement’.
6.2.1. Shareholder rights

The 2003 Communication on Modernising Company Law proposed
strengthening shareholders’ rights along lines which were implemented
in 2007 by the Shareholder Rights Directive.31 As stated in the 3rd
considerandum of the Directive, effective shareholder control is a pre-
requisite to sound corporate governance; these should, therefore, be
facilitated and encouraged, while obstacles which deter shareholders
from voting (such as making the exercise of voting rights subject to the
blocking of shares during a certain period before the general meeting)
should be removed. In particular, certain minimum standards should be
introduced with a view of protecting investors and promoting the
smooth and effective exercise of shareholder rights attached to voting
shares (4th considerandum). The Directive was also intended to solve the
problems related to cross-border voting, given that significant propor-
tions of shares in listed companies are held by shareholders who do not
reside in the Member State in which the company has its registered office
(5th considerandum). In particular, non-resident shareholders should be
able to exercise their rights in relation to the general meeting as easily as
shareholders who reside in the home Member State of the company.
Obstacles which hinder the access of non-resident shareholders to the
information relevant to the general meeting and the exercise of voting
rights without physically attending the general meeting should, there-
fore, be removed.
31
See n. 5 above and accompanying text.
The main issues dealt with by the Directive concern the organisation
and functioning of the shareholder meeting, and touch upon issues such
as: (a) information prior to the general meeting and convocation of the
same; (b) right to put items on the agenda of the general meeting and to
table draft resolutions; (c) requirements for participation and voting in
the general meeting (excluding the need for a prior deposit of the shares);
(d) participation in the general meeting by electronic means; (e) proxy
voting, including the right to appoint a proxy holder and the limits which
may be introduced by Member States in order to address conflicts of
interest; (f) voting by correspondence; and (g) publication of the voting
results. On the whole, the Directive makes shareholders’ participation to
general meetings easier, particularly in cross-border situations, but does
not per se promote the ‘shareholder engagement’ which is envisaged by
the 2010 and 2011 Green Papers as an essential component of an
effective corporate governance environment. Indeed, the Directive har-
monises some important aspects of the regimes applicable to shareholder
rights, removes obstacles to the exercise of those rights and possibly
reduces the costs of the same, but does not act on the basic incentives for
institutional and other investors to engage in activism.
6.2.2. Shareholder powers

Shareholder engagement also depends on the substantive powers attrib-
uted to shareholders vis-à-vis the (supervisory) board under national
company law. On a comparative ground, the distinction is made between
‘board-centric’ and ‘shareholder-centric’ systems (Davies et al. 2012).
The latter are in principle more open to shareholder activism, even
though activist investors are also present and successful in board-centric
systems, particularly in the US (as shown in Chapter 7). Board-centric
systems (like those of Germany, Italy, the Netherlands and Poland),
reserve only certain key powers to the general meeting (Rock et al.
2009; Bruno and Ruggiero 2011). These powers are defined in the law
either by a catch-all clause (such as ‘economically important decisions’)
or by a catalogue of fundamental decisions, such as charter amendments,
share issuance, mergers, divisions, etc. In shareholder-centric systems,
like the UK, the division of powers between the board and the share-
holders is left to the articles of association, but the shareholders may
decide in all matters that lie in the competence of the board and may
change its decisions by reaching a 75 per cent majority of the votes
(Enriques et al. 2009).
However, the distinction between board-centric and shareholder-

centric systems may become blurred in practice, depending on the
power relationship between the board of a company and its sharehold-
ers. In the case of controlling shareholders, a weak board may ask the
general meeting to decide matters that are in its own competence. On the
other hand, if directors can easily be removed by shareholders (e.g.
without cause and possibly even without receiving any compensation
for departure), the board may become weak even in a formally board-
centric system, since shareholders will hold the ultimate decisional
power. This will, in turn, create an incentive for entrepreneurs to retain
a control stake when the firm goes public. When shareholders are diffuse,
the general meeting may be weakened either by absenteeism or by lack of
shareholder engagement, with the result that the board and the managers
de facto enjoy greater powers than those formally attributed to them
(Davies et al. 2012).
As a general rule, the appointment and removal of (supervisory)
directors are tasks for the general meeting, except for cases where some
board members are elected by the workforce (as in the German code-
termination system) or by a third party. When ownership is diffuse, the
role of the general meeting is often formal in practice, as shareholders
elect candidates that are proposed by the board (ibid.) Only exception-
ally are candidates proposed to the general meeting by the shareholders
themselves and successfully elected; an example is offered by the Italian
slate voting system (analysed in Chapter 8). On the other hand, in
controlled companies, the ultimate power rests with the general meeting,
and the role of the board is often formal in practice: candidates are
usually proposed by the board under controlling shareholders’ instruc-
tions, so that controlling shareholders ultimately select and appoint the
full board. This, in turn, creates an incentive to keep ownership con-
centrated in the first place.
It is debated whether additional protection should be granted to
minority investors in controlled companies. The 2011 Green Paper
asked whether ‘minority shareholders need additional rights to represent
their interests effectively in companies with controlling or dominant
shareholders’. A positive answer would justify the recourse to multiple-
winner voting systems, granting board representation to qualified
minority shareholders through cumulative voting (as in Poland), pro-
portional voting (as in Spain) or quotas (as in the Italian slate voting
system). However, as reported in the Feedback Statement on the con-
sultation, ‘the vast majority of respondents that provided an answer to
this question share the view that minority shareholders are already
sufficiently protected’.32 Many respondents advanced two arguments
in particular: one being that additional rights are only likely to increase
the potential for abuse by minority shareholders and are contrary to
shareholder equality; the other that minority shareholders do not form a
homogenous group.
The rules on removal of (supervisory) directors, to some extent, reflect
the characterisation of a system as either ‘board-centric’ or ‘shareholder-
centric’ (Davies et al. 2012). The most shareholder-friendly rule is
removal without cause and without compensation. Other rules, which
require compensation and/or a cause for removal, tend to be board-
friendly. In any case, the provisions on duration of office should also be
considered, with shorter terms foreseen in shareholder-friendly regimes
(one year in the UK and Sweden) and longer terms (between three and
five years) in more board-friendly ones (Italy, France and Germany).
On the whole, the division of powers between boards and shareholders
has an impact on the potential for activism. Shareholder-centric systems
offer, in principle, a broader scope for activism, while board-centric
systems may need regulatory support for activism to arise. The EU
regulatory debate shows that such regulatory support is not necessarily
justified. On one hand, a shareholder-centric system may not offer the
incentives sought by institutional (and other) investors to become active.
In other words, a greater potential for activism in company law does
not imply that shareholders will be interested in exploiting it (Black
1990). On the other hand, the ‘abuse’ argument means that activism may
be the result of a conflict of interests instead of the solution to it, i.e.
investors might become active for the ‘wrong’ reason.
6.2.3. Reform proposals

The 2011 Green Paper claims that the lack of shareholder engagement
in European listed companies may derive from widespread short-
termism of investors, including those who have long-term obligations
towards their beneficiaries (such as pension funds, life insurance com-
panies, state pension reserve funds and sovereign wealth funds) and
should therefore be interested in improving long-term returns to share-
holders. A similar stance of investors may reflect the short-termism of
modern capital markets, but also the agency problems in the relation-
ship between long-term investors and their asset managers, who may
32
Feedback Statement, n. 22 above.
not be adequately incentivised to seek long-term benefits for their

principals.33
The Commission, moreover, conjectured that short-termism may
derive from ‘regulatory bias’ and asked participants in the consultation
on the Green Paper to identify EU legal rules that could be changed to
prevent such behaviour. Interestingly, all respondents invited caution
before any action is taken.34 ECLE (2011), in particular, claimed that:
we do not have a very sophisticated understanding of the relationship

between investment strategies and intervention in portfolio companies
or the links between such intervention and the long-term success of
portfolio companies. Some intervention by investors with short-term
goals is good because it brings about change which long-term investors
also want but cannot themselves cheaply bring about. Sometimes long-
term support for a company means keeping inefficient incumbent man-
agement in place. But equally, the opposites of these propositions also
hold true in some cases.
Furthermore, the 2011 Green Paper highlights the lack of transparency

about the performance of fiduciary duties by asset managers, suggesting
that ‘information about the level of and scope of engagement with
investee companies that the asset owner expects the asset manager to
exercise, and reporting on engagement activities by the asset manager
could be beneficial’. A different but complementary proposal was made
by the 2010 Green Paper, which suggested disclosure by institutional
investors of their voting practices at shareholders’ meetings as a way to
motivate shareholder engagement. A good example of similar policy
proposals is offered by the UK Stewardship Code, which was first issued
by the Financial Reporting Council in July 2010 with the aim to enhance
the quality of engagement between institutional investors and compan-
ies. Principle 6 of this Code states that institutional investors should
have a clear policy on voting and disclosure of voting activity, while
Principle 7 provides that institutional investors should report periodi-
cally on their stewardship and voting activities.
33
The Commission states (at p. 12 of the Green Paper): ‘It appears that the way asset
managers’ performance is evaluated and the incentive structure of fees and commissions
encourage asset managers to seek short-term benefits.’ A similar point is made and
further explored by Gilson and Gordon (2011), as mentioned also at n. 30 above.
34
The rules more frequently cited were the following (Feedback Statement p. 12): Solvency
II (in particular the provisions not enabling long-term investors to keep long-term
provisions); MiFID (as regards high frequency trading); financial reporting (especially
quarterly reporting); accounting (mark-to-market and fair value accounting in general).
In general, however, the case for enhancing shareholder activism in

Europe through regulatory harmonisation is rather weak. First, the
decision to engage in activism should be left to individual investors,
who will proceed on the basis of a cost–benefit analysis (save for cases in
which their incentives are clearly distorted). Second, investors’ incen-
tives, in order to become active, crucially depend on the characteristics of
the firm, such as its ownership structure. In controlled companies,
minority shareholders may rationally choose to stay passive, knowing
that they can influence corporate decisions very little. Third, the diver-
sity of institutional contexts in the EU – for instance, the divergence of
national rules dealing with the distribution of powers between boards
and shareholders – undermines regulatory harmonisation, which would
have different impact across countries. In fact, EU regulation has so far
followed a less ambitious ‘enabling’ approach, aimed at removing some
impediments to activism from Member State regulation. The final say is
still left to the individual shareholders within the different national
regulatory frameworks.

6.3.1. Management and shareholder proposals
Renneboog and Szilagyi in Chapter 7 analyse management and share-
holder proposals at the general meetings of listed firms in a number of
European countries, comparing the relevant data with those collected for
the US. In Europe, activists target firms that underperform and are, at the
same time, subject to governance concerns. This suggests that, just as in
the US, shareholder activism may produce non-trivial control benefits.
Shareholder proposals, in particular, may be regarded as a useful dis-
ciplinary tool and their sponsors as valuable monitoring agents.
Shareholders, however, submit proposals much less frequently than in
the US, particularly in Continental Europe. Furthermore, their success in
terms of voting results is limited across Europe, irrespective of the issues
addressed. There is also no evidence that the recourse to shareholder
proposals is on the rise as a result of the financial crisis or the adoption of
the EU Shareholder Rights Directive. While it is too early to fully gauge
its effects, this Directive aimed at minimum harmonisation and left a
number of important aspects untouched.
Renneboog and Szilagyi argue that the different recourse to share-
holder proposals on the two sides of the Atlantic can be attributed to
differences in the cost of activism and the regime of shareholder pro-

posals, which are non-binding in the US, while binding in most of
Europe. Furthermore, incentives to become active crucially depend on
firm characteristics, in particular on ownership structure. This is con-
firmed by the fact that fundamental differences exist in the objectives of
shareholder activism between the UK, where shareholdings are dis-
persed and activists often use proposals to replace the board, and
Continental Europe, where ownership is concentrated and proposal
objectives are generally confined to governance issues.
Renneboog and Szilagyi suggest that there might be scope for further
harmonisation in the areas of investor coordination and voting. For
example, the EU rules on acting in concert may deserve clarification,
so as to reduce regulatory disparities across Member States and facilitate
effective monitoring. However, further harmonisation looks problem-
atic: it would probably require tilting the current balance of power
between shareholders and boards, which may be both unwarranted
and contrary to the subsidiarity principle. On the other hand, a tailor-
made intervention by national legislators could make the playing field
more uneven.
Policy choices in this field imply trade-offs. Making shareholder
coordination easier mitigates the agency problems between managers
and shareholders as a whole, but aggravates the agency problems
between different classes of shareholders (ECLE 2011). In fact, institu-
tional investors will find it easier to coordinate their voting ex ante.
However, blockholders may also benefit from a looser treatment of
concert actions, which may allow them to enhance their control powers
and exercise the same to the detriment of minority shareholders.
The need for more transparency also applies to the voting behaviour
of institutional investors and the role of proxy advisers. In this regard, it
remains to be seen whether the adoption of UK-style codes of conduct
(on a ‘comply-or-explain’ basis) is sufficient, or if specific areas require a
direct, regulatory intervention.35
6.3.2. The Italian slate voting system

Belcredi, Bozzi and Di Noia in Chapter 8 analyse board elections in Italy,
offering further insights into the pros and cons of shareholder activism.
Board elections came to the forefront after the financial crisis. In the US,
shareholders’ influence over board elections is – apparently – at a
35
The point is specifically analysed by Wymeersch, Chapter 2 below.
historical minimum, so that a number of regulatory proposals were put

forward to increase the role of shareholders (Gordon 2008; Kahan and
Rock 2011). In Europe, the adoption of multiple-winner voting rules
granting board representation to minority shareholders is one of the
measures mentioned by the Green Paper on the EU Corporate
Governance Framework to support the alignment of managerial incen-
tives, particularly in companies with a controlling shareholder. The
analysis of previous national experiences allows a better assessment of
similar regulatory proposals. The Italian case looks particularly interest-
ing in this regard, as the introduction of a multiple-winner system has
been quite effective in stimulating activism.
Belcredi, Bozzi and Di Noia show that the submission by minority
shareholders of a board candidates’ slate is associated with firm charac-
teristics (above all, ownership structure and firm size), while voting rules
are comparatively less relevant. In particular, shareholder activism is
affected very little by the quorum required to submit a list, which is, on
average, one-quarter of the stake held by the second-largest shareholder
(the ideal candidate to seek board representation). The identity of active
investors also varies with firm characteristics. Mutual funds are active in
a small number of blue chip and other well-established companies
(satisfying criteria of prudent investment), while individual minority
shareholders holding a relevant stake submit slates in family firms.
Mutual funds’ activism is affected by transaction costs and portfolio
composition and, possibly, by the ‘political returns’ expected from
becoming active. Furthermore, institutional investors face regulatory
hurdles (the ‘acting in concert’ regime being perhaps the most conspicuous),
which are not easily overcome save for a favourable stance taken by the
supervisory authorities.
These results have a number of implications for the policy debate on
activism (not necessarily limited to corporate elections). A multiple-
winner voting rule spurs activism in a minority of cases. Where apathy
is nonetheless prevalent, board representation might simply not be a
cost-effective way to monitor management. In addition, the impact of
regulation may differ according to firm characteristics. This is particu-
larly relevant at the EU level, since regulatory harmonisation would
encompass industrial and ownership structures which are quite diverse
across Member States.
Once a multiple-winner system is in place, however, activism is
relatively insensitive to the voting rules specifically adopted at company
level. Moreover, transaction costs of activism are likely to be substantial,
while benefits from additional monitoring are, at best, uncertain. The

incentives to activism depend on characteristics of the institutional
context (such as ownership structure, regulatory and supervisory
approaches, etc.) which may vary across Member States. As a result, no
clear case for EU harmonisation can be made. Decisions in this field are
best left to individual shareholders, who can fully appreciate costs and
benefits of alternative strategies.
7. Policy
It is unclear whether, and to what extent, dysfunctional corporate gov-
ernance has contributed to the recent financial crisis. In order to answer
this question, financial institutions should be distinguished from other
companies. Recent empirical studies show that corporate governance
may have contributed to excessive risk taking by banks, in the sense that
firms characterised by ‘good’ corporate governance fared worse during
the financial turmoil. One could infer that corporate governance has
been too successful in aligning managers’ incentives with the interests of
shareholders. However, it is also important to consider that financial
institutions are highly influential and that the agency costs of debt are
therefore important for them. These costs create a last-stage problem,
which materialises when the risk of default is non-trivial, leading
shareholders (and the managers appointed by the same) to deviate
from value maximisation. If regulation of risk taking by financial insti-
tutions is insufficient or ineffective, corporate governance may exacer-
bate managers’ and shareholders’ incentives to gamble with creditors’
money. In other words, corporate governance standards are not neces-
sarily ‘wrong’, but may create perverse incentives in firms which are not
properly regulated and supervised. As a result, banks may need better
prudential regulation and supervision rather than corporate governance
reform.
A different question is whether corporate governance standards are
correctly defined. No doubt corporate governance mechanisms have
intrinsic limitations. For example, independent directors may be fit to
supervise related party transactions, but less to control the conflicts of
interest between shareholders and creditors (for possible lack of profes-
sional skills and experience). We still know very little about the relative
merits of different mechanisms and should therefore be cautious in
extending corporate governance standards along a ‘one size fits all’
model. Codes of best practice play a key role in this regard by allowing
individual firms and countries to benefit from experience and improve

on their practices incrementally.
As to non-financial companies, the evidence of dysfunctional corpor-
ate governance (and of a possible causal link with the recent turmoil) is
even more limited. Consequently, the need for EU reform also finds
limited support. Let us consider the four main corporate governance
areas analysed in this volume (board structures, directors’ remuneration,
shareholder activism and ‘comply or explain’) and draw some policy
implications.
7.1. Board structures

The arguments and evidence provided in this volume suggest restraint in
the design of new standards for board structures. For instance, increas-
ing board diversity carries costs as well as benefits. If board members
differ as to nationality, language differences may determine communi-
cation problems which are no less dangerous than ‘groupthink’. No easy
recipe exists for board diversity. Optimal governance structures, to a
large extent, depend on firm-specific factors, the evaluation of which is
best left to shareholders. A similar argument can be advanced for
minority investors’ access to the boardroom, the benefits of which
depend on the ownership structure and size of companies.
No clear case for regulation can derive from anecdotal – and, so far,
unsystematic – evidence of market failures. Shareholders’ decisions have
not been proven to be systematically flawed and in need of correction.
Nor is it clear why national (or EU) legislators may be expected to
produce a superior outcome. Gender diversity is different to the extent
that the protection of the general interest to granting equal opportunities
to women is at stake, which, however, has little to do with shareholder
value. EU intervention in this regard may add a separate layer of rules
which are not necessarily fully consistent (as to substance and timing)
with national approaches to gender diversity. Therefore, it seems import-
ant at least to keep some flexibility in the formulation of uniform
standards. A general argument against EU regulation is that board
structures should vary depending on social and institutional features,
which greatly differ across Member States, including ownership struc-
tures; the board-centric or shareholder-centric orientation of each gov-
ernance system; the prevailing management culture; and other aspects of
the legal system, such as the quality of private and public enforcement.
Governance models are ‘sticky’ and path-dependent (Bebchuk and Roe
1999; Schmidt 2004), so that new rules on boards would yield different
results across Member States.
Similar arguments hold for non-binding standards. As the EU recom-
mendations on board structure, composition and functioning grow in
size and impact, their basis appears to be thinner, especially if it consists
of theoretical models rather than observable best practices. Moreover,
abstract analysis is easily bent to serve individual constituencies, so that
the new standards may be influenced by fashions and fads. No doubt any
adverse consequence of innovation as to codes of best practice is tem-
pered by their non-binding nature. This may lead to experimenting new
solutions, however dubious their merits. In addition, possible deviations
in practice from the new (arguable) standard may be exploited to sup-
port a call for binding rules, which would then crystallise solutions that
are weakly grounded.
7.2. Directors’ remuneration

Directors’ remuneration is still a hot topic in the policy debate. Once again,
a distinction should be made between financial institutions and other firms.
Many have regarded managerial compensation as one of the causes of
excessive risk-taking by financial institutions, if not as one of the determin-
ants of the financial crisis. Nonetheless, the available evidence shows that
managerial and shareholder interests were aligned in banks before the crisis
and that short-term incentives did not necessarily have an adverse impact
on bank performance during the crisis. One of the likely main reasons for
excessive risk taking was insufficient or ineffective prudential regulation
rather than flawed corporate governance.
However, some new rules concerning the disclosure, governance, level
and structure of managerial remuneration have been enacted in response
to the turmoil. Moreover, our evidence in this volume shows that the
level and structure of managerial compensation in European financial
institutions have, indeed, changed after the crisis. In particular, CEOs
have experienced a decrease of their cash bonuses, while other compon-
ents of remuneration remained substantially unchanged. Furthermore,
stock grants have apparently become more popular than stock options. It
is, however, difficult to assess whether pay-performance sensitivity has
increased or decreased as a result.
In non-financial companies leverage is generally much lower, so that
excessive risk taking, while troublesome, causes less concern. The main
worry is that managers may use their power to extract rents from the
company through their compensation to the detriment of shareholders.

‘Excessive’ compensation may derive in particular from unduly complex
structures adopted to ‘camouflage’ the true amounts paid and to avoid
shareholder scrutiny. There is no easy way to cope with this problem,
since the informational asymmetry inherent in the manager-shareholder
relationship is not easily overcome.
As a result, no clear case can be made for regulatory intervention
on the level/structure of managerial remuneration at non-financial
companies. For example, schemes that have somehow become popular,
such as malus and clawback clauses in compensation arrangements, are
less needed for this kind of firm. Two types of remedies have instead been
adopted at EU level, which favour either the corporate governance
structure (remuneration committees and say-on-pay) or remuneration
disclosure. Governance solutions are generally non-binding (like the EU
recommendation on remuneration committees), as they may also
depend on the underlying national law and require further experimen-
tation. Even when binding provisions are adopted at national level, as in
the case of say-on-pay, flexibility is often preserved through the adoption
of an ‘advisory’ vote (an unprecedented solution in some jurisdictions).
Our evidence shows that conformity to EU recommendations concern-
ing the governance of the remuneration process is generally good. We
see no need to change the current approach in any fundamental way.
The same considerations do not apply to disclosure, for which man-
datory provisions may be preferable. To be sure, our evidence in this
respect shows that the implementation of EU recommendations in the
Member States has been rather diverse. Disclosure of individual remu-
neration has increased remarkably over the last few years, while trans-
parency lags behind in the case of forward-looking policy, breakdown of
pay components, performance parameters for the variable component
and dynamics of stock-based compensation. No doubt the implemen-
tation of recommendations takes time, but the issue requires careful
monitoring, so as to better assess whether harmonisation of disclosure
might also be in order for non-financial firms.
The optimal degree of transparency about remuneration packages is
nonetheless debated. While disclosure may contribute to keeping man-
agerial rent-extraction under control, it could also determine a
‘ratcheting’ effect in firms where remuneration is below average. As a
result, additional disclosure could reduce the cross-sectional variance of
compensation, which is not necessarily a desirable outcome. Moreover,
remuneration disclosure is subject to intrinsic limitations, especially
where it forms the basis for a shareholder vote. Indeed, remuneration

packages are complex, and shareholders may lack the incentive and
expertise to analyse the relevant information and come to a correct
decision. Say-on-pay may be insufficient to control rent-extraction or,
worse, may favour herd behaviour (i.e. box-ticking and adherence to a
conventional standard model). The role of proxy advisers may be crucial
in this regard.
While it is difficult to say whether (and to what extent) managerial
remuneration is ‘excessive’ or ‘unduly complex’, the evidence produced
in this volume shows that the level and structure of CEO pay in non-
financial European firms have not changed much after the crisis. Of
course, this evidence has no clear implications about the presence of
rent-extraction by managers (which alternatively might be absent or not
have changed with respect to the situation before the crisis) or the
effectiveness of alternative governance arrangements. However, some
evidence shows that remuneration is related to firm fundamentals, such
as size, sector, growth opportunities and corporate results. This is con-
sistent with the hypothesis that the market for managerial services is, at
least to some extent, efficient.
7.3. Shareholder activism

It is uncertain whether regulation should promote shareholder
‘engagement’ with the firm and managerial accountability. The EU
Commission has already adopted a series of measures aimed at removing
impediments to the exercise of shareholders’ rights, thereby favouring
cross-border mobility of capital. Further measures are being considered.
Our evidence about shareholder activism in Europe is mixed. Activists
target firms that both underperform and are subject to governance
concerns. This suggests that, as in the US, shareholder activism may be
a useful disciplinary tool. However, shareholders, particularly in
Continental Europe, submit proposals much less frequently than in the
US. Furthermore, both the frequency and targets of activism differ
greatly across countries. Ownership structures and national corporate
law have an impact on activism. Shareholder proposals are not on the
rise and their success is limited across Europe irrespective of the issues
addressed. The votes cast in favour of shareholder proposals are on the
rise, which implies that shareholders’ ability to dissent is greater than
before (possibly an effect of the Shareholders’ Rights Directive).
However, the exercise of a ‘voice’ led to surprising results only in a
handful of ‘outrageous’ cases. The effectiveness of this type of

monitoring – particularly where ownership is concentrated – is open
to question.
It is therefore unclear whether further regulatory intervention is
needed, except for investor coordination and voting. In particular, the
EU rules on acting in concert may deserve clarification, so as to facilitate
engagement efforts, especially by institutional investors. Rules on insider
trading and market abuse might also be amended to facilitate proper
dialogue between companies and investors. There might also be some
scope for reducing the limits to cross-border voting by institutional
investors, even though it is unclear whether the procedural and infor-
mation costs of activism would be substantially reduced as a result.
Indeed, some limits to activism are beyond the reach of national and
EU regulators, while the transaction costs of activism are higher in the
case of cross-border investments. In general, the decision to engage in
activism should be left to individual investors, who will then proceed on
the basis of their own cost–benefit analysis.
Similar conclusions apply to activism in board elections. The available
evidence about the Italian investor-friendly voting system shows that the
existence and identity of active shareholders are associated with firm-
specific characteristics (mainly ownership structure and firm size), while
voting rules are comparatively less relevant. Institutional investors con-
centrate their efforts on a small number of blue chips, given transaction
costs and portfolio composition, and also the ‘political returns’ from
being active. Furthermore, they face some regulatory hurdles, which are
not easily overcome unless a favourable stance is taken by supervisory
authorities.
Activism in board elections takes place only in a minority of cases,
despite the list voting regime. This seems to indicate that shareholder
apathy is indeed rational and that shareholder-friendly rules generate a
modest incentive to be active. Board representation of minority investors
might simply not be a cost-effective monitoring instrument. As incen-
tives to activism depend on the institutional context, and this varies
across Member States, no clear case for EU harmonisation can be made.
Decisions in this field are best left to individual shareholders, who can
fully appreciate the costs and benefits of alternative strategies.
There is also growing pressure to enhance transparency about the level
and scope of asset managers’ engagement with investee companies, along
the lines of the UK Stewardship Code. Disclosure of engagement and
voting policies may, however, end up as a mere box-ticking exercise with
little value unless it is monitored by ultimate (individual or institutional)

investors. Moreover, increased transparency can create an artificial
demand for the services of proxy advisers, increasing the risk of
‘herding’ behaviour. Additional analyses are needed to address these
issues.
7.4. ‘Comply or explain’

‘Comply or explain’ is a core principle of European corporate govern-
ance, which was officially enacted through Directive 2006/46/EC man-
dating transparency as to the application of corporate governance codes.
This principle enjoys broad support in practice thanks to its flexibility
(Riskmetrics 2009). Codes of best practice allow individual firms and
countries to take advantage of previous experience and to improve
standards incrementally through a process of trial and error. The vol-
untary adoption of governance mechanisms diffuse in corporate practice
(such as board committees, senior independent director, separation
between chairman and CEO) has been impressive over the last decade.
However, flexibility also constitutes a weakness of ‘comply or explain’, as
it is often difficult to gauge the real conduct behind the words of a
governance statement. Furthermore, some statements are poorly drafted
and carry boilerplate explanations. Briefly, while the principle receives
broad support, its practical implementation is still far from perfect, and
improvements are no doubt possible.
First, codes usually include two layers of principles, and the ‘comply or
explain’ regime generally applies to one of them. Codes should distin-
guish more clearly between these two layers, reserving the ‘comply or
explain’ mechanism to high-level principles and to provisions that are
broadly recognised as ‘best practice’. Once this distinction is made,
deviations from the relevant provisions would, in principle, require a
specific explanation. However, the question arises of how to mandate
(and enforce) disclosure about conformity to a non-binding standard’.
Reputational mechanisms (relying on investor pressure and the media)
are increasingly perceived as insufficient.
The analysis conducted in this volume shows that it is not yet time for
further EU harmonisation. Different ownership and governance struc-
tures, as well as different legal regimes, counsel avoiding a uniform
approach. Rather, corporate governance commissions should better
explore how they can learn from each other and, possibly, align their
recommendations and terminology. At the same time, companies should
streamline their governance practices and disclosures, with the support

of European business associations. Only after a careful preparatory work
could reasonable high-level principles be developed at the European
level. However, national standard setters should probably remain free
to adopt only those which fit best their context.
7.5. The Action Plan

When this volume was almost ready for publication, the European
Commission disclosed its Action Plan on European company law and
corporate governance, reflecting the outcomes of the 2012 public con-
sultation.36 Some of the proposals set out in the Action Plan are directly
relevant for the topics addressed in this volume and deserve brief com-
ment in this introductory chapter.
Board structure will not be targeted by specific regulation. The
Commission acknowledges the coexistence of different board models,
deeply rooted in national legal systems (and possibly linked with differ-
ent ownership structures) and does not pursue further harmonisation.
This is consistent with the results of our analysis. However, we regard the
Commission’s proposals on board composition as problematic. The
Commission will act in order to enhance diversity (in addition to having
proposed a directive on gender diversity). No doubt, introducing dis-
closure requirements relative to firms’ board diversity policy is a form of
light-touch regulation. Nonetheless, we see a similar move as weakly
grounded, since strong evidence that board diversity is suboptimal is
currently lacking.37 In addition, this move is not supported by the results
of the consultation on the 2011 Green Paper, the responses to which
were almost equally divided between those favouring and those opposing
specific measures in this regard. Furthermore, disclosure requirements
could pave the way to substantive regulation if the former were found
insufficient to attain the stated regulatory objective. From a similar
perspective, disclosure is problematic to the extent that it is used to
36
Communication from the Commission to the European Parliament, the Council, the
European Economic and Social Committee and the Committee of the Regions, Action
Plan: European Company law and corporate governance – a modern legal framework for
more engaged shareholders and sustainable companies, COM(2012) 740/2.
37
In this regard, gender diversity is different, since regulatory proposals are clearly
stakeholder-oriented. However, even in this case, the Commission has completely over-
looked the practical issues potentially associated with the implementation of quotas.
attain indirectly a given governance structure on which no consensus

presently exists.
A substantial part of the Action Plan’s proposals aim at enhancing the
engagement of shareholders. First, the Commission plans to strengthen
the transparency rules for institutional investors. While this develop-
ment is welcome in principle, small investors usually lack either the
competence or incentive to monitor the behaviour of investment man-
agers. Therefore, initiatives in this direction may practically translate
into box-ticking exercises or determine herding behaviour. We also
suggest caution in devising the proposed regulation of proxy advisers,
so as to avoid mistakes similar to those made in the past vis-à-vis rating
agencies. Regulation should improve the transparency and limit the
conflicts of interest of proxy advisers, but avoid creating perverse incen-
tives to the use of their services (such as attaching legal consequences to
the same, thereby protecting the business model of a conflicted partici-
pant in an oligopolistic market).
Our analysis shows that shareholder activism carries transaction costs,
as well as benefits. Therefore, shareholders should ultimately decide on
activism, as also suggested by the Action Plan, which does not mandate
engagement with listed companies. Rather, the Commission’s strategy
focuses on two goals. The first is to remove some of the regulatory
obstacles to shareholder engagement. The Commission plans to work
with national authorities and ESMA to increase legal certainty on the
relationship between investor cooperation on corporate governance
issues and the rules on acting in concert. The second goal is to encourage
shareholder engagement on specific issues, with respect to which outside
monitoring of managerial and/or board actions looks particularly useful
and cost-effective. The Commission proposes to enhance oversight of
directors’ remuneration through harmonised disclosure and voting on
the firms’ remuneration policy and remuneration report, and to promote
shareholder oversight of significant transactions with related parties. On
the other hand, the Commission has abstained from proposing rules on
minority representation in corporate bodies.
The Commission’s approach to activism is in accord with our policy
conclusions. However, increased activism may still raise some concerns,
especially in the area of ‘say-on-pay’. On one side, the ‘mandatory
shareholder vote’ contemplated by the Action Plan should also provide
sufficient flexibility to accommodate non-binding regimes, which have
not been proven to be dysfunctional. On the other, remuneration pack-
ages are intrinsically complex and hard to evaluate. Moreover, the sheer
number of investee firms is a formidable obstacle to specific analysis by

institutional investors. As a result, investment managers wanting to
comply with the new requirements may simply outsource the whole
process to proxy advisers, who may come to dominate the same from
their oligopolistic position in the relevant market.
Finally, corporate governance codes based on the comply-or-explain
approach have, once more, substantially passed the Commission’s scru-
tiny and will not be targeted by new regulation. However, the Action
Plan remarks that the explanations provided by companies are often still
insufficient, even though some national self-regulatory bodies try to
improve the quality of explanations. The Commission wishes to encour-
age the exchange of best practices developed in different Member States
and will take a further initiative – possibly in the form of a
recommendation – to improve the quality of corporate governance
reports. This development is also consistent with our analysis.
8. Concluding remarks
This volume analyses a number of topics concerning the role of boards
and shareholders in the corporate governance of European listed firms.
The evidence provided in the following chapters challenges the conven-
tional wisdom that corporate governance arrangements in European
firms are systematically dysfunctional and have contributed to the
financial turmoil. Even though our volume does not specifically target
financial institutions, a growing body of evidence indicates that, when
looking for the ultimate cause of the financial crisis, lack of proper
regulation and supervision is a more likely candidate than flawed cor-
porate governance.
We analyse four main topics in the corporate governance of European
listed firms: board structure/composition and its interaction with owner-
ship structure, board remuneration, shareholder activism and corporate
governance disclosure based on the ‘comply-or-explain’ approach. For each
of them, we provide new evidence which allows us to derive specific
implications relevant for the policy debate both at Member State and EU
level. Basically, we show that proposals aimed at increasing disclosure and
accountability are generally well-grounded, particularly in the areas of
remuneration and of compliance with corporate governance codes.
However, we suggest caution with respect to proposals targeting specific
governance arrangements, as they may lead to unintended consequences.
Whilst the European Commission has – so far – refrained from adopting an
excessively intrusive stance, further analysis would, in any case, be needed

before adopting harmonisation measures in the fields of board composition
and shareholder activism.
References
Adams, R. B. 2009. ‘Governance and the Financial Crisis’, ECGI. Finance Working
Paper 248/2009.
Adams, R. B. and D. Ferreira 2009. ‘Women in the boardroom and their impact on
governance and performance’, Journal of Financial Economics 94: 291–309.
Adams, R. B., B. E. Hermalin and M. S. Weisbach 2010. ‘The role of boards of
directors in corporate governance: A conceptual framework and survey’,
Journal of Economic Literature 48: 58–107.
Adams, R. B. and H. Mehran 2003. ‘Is corporate governance different for bank
holding companies?’, Economic Policy Review 9: 123–42.
Admati, A. R., P. M. DeMarzo, M. F. Hellwig and P. Pfleiderer 2012. ‘Debt
Overhang and Capital Regulation’, Rock Center for Corporate
Governance at Stanford University, Working Paper No. 114.
Ahern, K. R. and A. K. Dittmar 2012. ‘The changing role of the boards: The impact
on firm valuation of mandated female board representation’, The Quarterly
Journal of Economics 127: 137–97.
Andres, C. 2008. ‘Large shareholders and firm performance: An empirical exami-
nation of founding-family ownership’, Journal of Corporate Finance 14:
431–45.
Armour, J. and B. Cheffins 2009. ‘The Rise and Fall (?) of Shareholder Activism by
Hedge Funds’, ECGI Law Working Paper 136/2009.
Armour, J., S. Deakin, P. Lele and M. M. Siems 2009a. ‘How do Legal Systems
Evolve? Evidence from a Cross-country Comparison of Shareholder,
Creditor and Worker Protection’, ECGI Law Working Paper 129/2009.
Armour, J., H. B. Hansmann and R. H. Kraakman 2009b. ‘Agency problems and
legal strategies’, in Kraakman et al., pp. 35–54.
Armour, J. and W. G. Ringe 2011. ‘European company law 1999–2010:
Renaissance and crisis’, Common Market Law Review 48: 125–74.
Bainbridge, S. M. 2002. ‘Why a board? Group decisionmaking in corporate
governance’, Vanderbilt Law Review 55: 1–55.
2012. Corporate Governance after the Financial Crisis, Oxford University
Press.
Barca, F. and M. Becht (eds.) 2001. The Control of Corporate Europe, Oxford
University Press.
Barnea, A., R. A. Haugen and L. W. Senbet 1981. ‘Market imperfections, agency
problems, and capital structure: A review’, Financial Management 10: 7–22.
Barontini, R. and L. Caprio 2006. ‘The effect of family control on firm value and
performance: Evidence from Continental Europe’, European Financial
Management 12: 689–723.
Basel Committee 2010. Principles for Enhancing Corporate Governance, Basel
Committee on Banking Supervision.
Bebchuk, L. A. 2005. ‘The case for increasing shareholder power’, Harvard Law
Review 118: 835–914.
Bebchuk, L. A. and J. Fried 2004. Pay without Performance: The Unfulfilled
Promise of Executive Compensation, Harvard University Press,
Cambridge, MA and London.
Bebchuk, L. A., J. Fried and D. Walker 2002. ‘Managerial power and rent extrac-
tion in the design of executive compensation’, University of Chicago Law
Review 69: 751–846.
Bebchuk, L. A. and M. J. Roe 1999. ‘A theory of path-dependence in corporate
ownership and governance’, Stanford Law Review 52: 127–70.
Bebchuk, L. A. and H. Spamann 2010. ‘Regulating bankers’ pay’, Georgetown Law
Journal 98: 247–87.
Becht, M., P. Bolton and A. Röell 2002. ‘Corporate Governance and Control’,
ECGI Finance Working Paper 02/2002.
2012. ‘Why bank governance is different’, Oxford Review of Economic Policy 27:
437–63.
Becht, M., J. Franks, C. Mayer and S. Rossi 2009. ‘Returns to shareholder activism:
Evidence from a clinical study of the Hermes UK focus fund’, Review of
Financial Studies 22: 3093–129.
Becht, M. and C. Mayer 2001. ‘Introduction’ in Barca and Becht (eds.), pp. 1–45.
Beltratti, A. and R. M. Stulz 2012. ‘The credit crisis around the globe: Why did
some banks perform better?’, Journal of Financial Economics 105: 1–17.
Bhagat, S., P. Bolton and R. Romano 2008. ‘The promise and perils of corporate
governance indices,’ Columbia Law Review 108: 1803–82.
Black, B. 1990. ‘Shareholder passivity reexamined’, Michigan Law Review 89:
520–608.
Blair, M. M. 1995. Ownership and Control: Rethinking Corporate Governance for
the Twenty-first Century, The Brookings Institution, Washington DC.
and L. Stout 2001. ‘Trust, trustworthiness, and the behavioral foundations of
corporate law’, University of Pennsylvania Law Review 149: 1735–810.
Braithwaite, J. 2008. Regulatory Capitalism: How it Works, Ideas for making it
Work Better, Edward Elgar, Cheltenham, UK; Northampton, MA.
Brandeis, L. 1914. Other People’s Money and How the Bankers Use It, Frederick
A. Stokes Publishers, New York.
Bruno, S. and E. Ruggiero 2011. Public Companies and the Role of Shareholders:
National Models towards Global Integration, Kluwer Law International,
Alphen aan den Rijn, the Netherlands.
Carter, D. A., B. J. Simkins and W. G. Simpson 2003. ‘Corporate governance,
board diversity, and firm value’, The Financial Review 38: 33–53.
Cheffins, B. R. 2001. ‘Does law matter? The separation of ownership and control in
the United Kingdom’, Journal of Legal Studies 30: 459–84.
2009. ‘Did corporate governance “fail” during the 2008 stock market melt-
down? The case of the S&P 500’, Business Lawyer 65: 1–65.
Claessens, S., S. Djankov and L. H. P. Lang 2000. ‘The separation of ownership and
control in East Asian corporations’, Journal of Financial Economics 58: 81–112.
Clarke, T. and J. F. Chanlat 2009, European Corporate Governance: Readings and
Perspectives, Routledge, New York.
Coffee Jr. J. C. 1989. ‘The mandatory/enabling balance in corporate law: an essay
on the judicial role’, Columbia Law Review 89: 1618–91.
2001. ‘The rise of dispersed ownership: The roles of law and the state in the
separation of ownership and control’, Yale Law Journal 111: 1–82.
2002. ‘Understanding Enron: “It’s the gatekeepers, stupid”’, Business Lawyer
57: 1403–20.
2005. ‘A theory of corporate scandals: Why the USA and Europe differ’, Oxford
Review of Economic Policy 21: 198–211.
Conyon, M. and D. Leech 1994. ‘Top pay, company performance and corporate
governance’, Oxford Bulletin of Economics and Statistics 56: 229–47.
Cools, S. 2005. ‘The real difference in corporate law between the United States and
continental Europe: Distribution of powers’, Delaware Journal of Corporate
Law 30: 697–766.
Corrigan, E. G. 1982. ‘Are banks special?’, in The Federal Reserve Bank of
Minneapolis: 1982 Annual Report, available at www.minneapolisfed.org.
2000. ‘Are banks special? A revisitation’, The Region, available at www.minne-
apolisfed.org.
Davies, P., K. J. Hopt, R. Nowak and G. van Solinge 2012. ‘Boards in Law and
Practice: A Cross-country Analysis in Europe’, Forum Europaeum
Corporate Boards (FECB), mimeo.
Dyck, A. and L. Zingales 2002. ‘Private Benefits of Control: An International
Comparison’, NBER Working Paper 8711.
Easterbrook, F. H. and D. L. Fischel 1991. The Economic Structure of Corporate
Law, Harvard University Press, Cambridge, MA and London.
Ellul, A. and V. Yerramilli 2012. ‘Stronger risk controls, lower risk: Evidence from
US bank holding companies’, Journal of Finance, forthcoming, available at
www.ssrn.com.
Enriques, L., H. B. Hansmann and R. H. Kraakman 2009. ‘The basic governance
structure: The interests of shareholders as a class’, in Kraakman et al.,
pp. 55–88.
Enriques, L. and P. Volpin 2007. ‘Corporate governance reforms in continental
Europe’, Journal of Economic Perspectives 21: 117–40.
European Company Law Experts (ECLE) 2011. ‘Response to the European
Commission’s Green Paper ‘The EU Corporate Governance Framework’,
available at www.ssrn.com.
Faccio, M. and L. H. P. Lang 2002. ‘The ultimate ownership of Western European
corporations’, Journal of Financial Economics 65: 365–95.
Fahlenbrach, R. and R. M. Stulz 2011, ‘Bank CEO incentives and the credit crisis’,
Journal of Financial Economics 99: 11–26.
Fama, E. F. 1980. ‘Agency problems and the theory of the firm’, Journal of Political
Economy 88: 288–307.
and M. C. Jensen 1983. ‘Separation of ownership and control’, Journal of Law
and Economics 26: 301–25.
Ferrarini, G. 2012. ‘Bankers’ compensation and prudential supervision’, in Hill
and Thomas (eds.)
Ferrarini, G. and P. Giudici 2006. ‘Financial scandals and the role of private
enforcement: the Parmalat case’, in Armour, J. and J. A. McCahery (eds.),
After Enron, Hart Publishers, Oxford, pp. 159–213.
Ferrarini, G., K. J. Hopt, J. Winter and E. Wymeersch (eds.) 2004. Reforming
Company and Takeover Law in Europe, Oxford University Press.
Ferrarini, G. and G. Miller 2010. ‘A simple theory of takeover regulation in
the United States and Europe’, Cornell International Law Journal 42:
301–34.
Ferrarini, G. and N. Moloney 2004. ‘Executive remuneration and corporate
governance in the EU: Convergence, divergence, and reform perspectives’,
European Company and Financial Law Review 1: 251–339.
Ferrarini, G. and N. Moloney 2005. ‘Executive remuneration in the EU: The
context for reform’, Oxford Review of Economic Policy 21: 304–23.
Ferrarini, G., N. Moloney and M. C. Ungureanu 2010. ‘Executive remuneration in
crisis: A critical assessment of reforms in Europe’, Journal of Corporate Law
Studies 10: 73–118.
Ferrarini, G. and M. C. Ungureanu 2010. ‘Executive pay at ailing banks and
beyond: A European perspective’, Capital Markets Law Journal 5: 197–217.
2011. ‘Economics, politics, and the international Principles for Sound
Compensation Practices: An analysis of executive pay at European banks’,
Vanderbilt Law Review 64: 431–502.
Financial Stability Board (FSB) 2010. Thematic Review on Compensation: Peer
Review Report, available at www.financialstabilityboard.org.
Franks, J. and C. Mayer 1996. ‘Hostile takeovers and the correction of managerial
failure’, Journal of Financial Economics 40: 163–81.
Franks, J., C. Mayer, P. Volpin and H. F. Wagner 2012. ‘The life cycle of family
ownership: A comparative study of France, Germany, Italy and the U.K.’,
Review of Financial Studies 25: 1675–712.
Garrido, J. and A. Rojo 2003. ‘Institutional investors and corporate governance:
Solution or problem?’, in Hopt, K. J. and E. Wymeersch (eds.). Capital
Markets and Company Law, Oxford University Press, pp. 427–48.
Gillan, S. L. and L. T. Starks 1998. ‘A survey of shareholder activism: Motivation
and empirical evidence’, Contemporary Finance Digest 2: 10–34.
2007. ‘The evolution of shareholder activism in the United States’, Journal of
Applied Corporate Finance 19: 55–73.
Gilson, R. J. 2006. ‘Controlling shareholders and corporate governance: compli-
cating the comparative taxonomy’, Harvard Law Review 119: 1642–79.
Gilson, R. J. and J. N. Gordon 2011. ‘Capital Markets, Efficient Risk Bearing and
Corporate Governance: The Agency Costs of Agency Capitalism’, prelimi-
nary draft January 2011, downloadable at www.law.columbia.edu.
Gilson, R. J. and R. H. Kraakman 2003. ‘The mechanisms of market efficiency twenty
years later: The hindsight bias’, The Journal of Corporation Law 28: 715–42.
Gordon, J. N. 1989. ‘The mandatory structure of corporate law’, Columbia Law
Review 89: 1549–98.
2002. ‘What Enron means for the management and control of the modern
business corporation: Some initial reflections’, University of Chicago Law
Review 69: 1233–50.
2007. ‘The rise of independent directors in the United States, 1950–2005: Of
shareholder value and stock market prices’, Stanford Law Review 59:
1465–568.
2008. ‘Proxy contests in an era of increasing shareholder power: Forget issuer
proxy access and focus on e-proxy’, Vanderbilt Law Review 61: 475–96.
and M. J. Roe (eds.) 2004. Convergence and Persistence in Corporate
Governance, Cambridge University Press.
Hall, B. and J. Liebman 1997. ‘Are CEOs Really Paid like Bureaucrats?’, NBER
Working Paper 6213.
Hansmann, H. B. and R. H. Kraakman 2001. ‘The end of history for corporate
law’, Georgetown Law Journal 89: 439–68.
Harris, M. and A. Raviv 2008. ‘Control of Corporate Decisions: Shareholders vs.
Management’, Working Paper, Center for Research in Security Prices.
Hermalin, B. E., and M. S. Weisbach 2003. ‘Boards of directors as an endogenously
determined institution: A survey of the economic literature’, Federal
Reserve Board of New York Economic Policy Review 9: 7–26.
Hill, J. and R. Thomas (eds.) 2012. Research Handbook on Executive Pay, Edward
Elgar Publishing, Cheltenham.
Hill, J. and C. M. Yablon 2002, ‘Corporate governance and executive remuner-
ation: Rediscovering managerial positional conflict’, University of New
South Wales Law Journal 25: 294–319.
Holmstrom, B. and S. Kaplan 2003. ‘The state of US corporate governance: What’s
right and what’s wrong?’, Journal of Applied Corporate Finance 15: 8–20.
Hopt, K. J. 2011. ‘Comparative corporate governance: The state of the art and
international regulation’, American Journal of Comparative Law 59: 1–74.
Jagannathan, R. and S. B. Srinivasan 2000. ‘Does Product Market Competition
Reduce Agency Costs?’, NBER Working Paper 7480.
Jensen, M. C. 1988. ‘Takeovers: Their causes and consequences’, Journal of
Economic Perspectives 2: 21–48.
and W. H. Meckling 1976. ‘Theory of the firm: Managerial behavior, agency
costs and ownership structure’, Journal of Financial Economics 3: 305–60.
Jensen, M. C. and K. Murphy 1990. ‘Performance pay and top management
incentives’, Journal of Political Economy 98: 225–64.
2004. ‘Remuneration: Where we’ve Been, How we Got to Here, What are the
Problems, and How to Fix Them’, ECGI Finance Working Paper 44/2004.
Kahan, M. and E. B. Rock 2007. ‘Hedge funds in corporate governance and
corporate control’, University of Pennsylvania Law Review 155: 1021–94.
2011. ‘The insignificance of proxy access’, Virginia Law Review 97: 1347–434.
Kaplow, L. 1992. ‘Rules versus standards: An economic analysis’, Duke Law
Journal 42: 557–629.
Kirkpatrick, G. 2009. ‘The corporate governance lessons from the financial crisis’,
OECD Journal Financial Market Trends 2009/1: 61–87.
Kraakman, R. 2004. ‘Disclosure and corporate governance: an overview essay’, in
Ferrarini et al. (eds.), pp. 97–126.
J. Armour, P. Davies, L. Enriques, H. B. Hansmann, G. Hertig, K. J. Hopt,
H. Kanda and E. B. Rock 2009. The Anatomy of Corporate Law: A
Comparative and Functional Approach, 2nd edn., Oxford University Press.
Laeven, L. and R. Levine 2009. ‘Bank governance, regulation and risk taking’,
La Porta, R., F. Lopez-de-Silanes and A. Shleifer 1999. ‘Corporate ownership
around the world’, Journal of Finance 54: 471–517.
and R. W. Vishny 1997. ‘Legal determinants of outside finance’, Journal of
Finance 52: 1131–50.
1998. ‘Law and finance’, Journal of Political Economy 106: 1113–55.
2000. ‘Investor protection and corporate governance’, Journal of Financial
Economics 58: 3–27.
Levine, R. 2004. ‘The Corporate Governance of Banks: A Concise Discussion of
Concepts and Evidence’, World Bank Policy Research Group, Working
Paper No. 3404, available at www.ssrn.com.
Levitt, A. 2005. ‘Corporate culture and the problem of executive compensation’,
Journal of Applied Corporate Finance 17: 749–54.
Loewenstein, M. J. 1996. ‘Reflections on executive compensation and a modest proposal
for (further) reform’, Southern Methodist University Law Review 50: 201–23.
Macey, J. P. and G. P. Miller 1992. ‘Double liability of bank shareholders: History
and implications’, Wake Forest Law Review 27: 31–62.
Macey, J. and M. O‘Hara 2003. ‘The corporate governance of banks’, Economic
Policy Review 9: 91–107.
Mahoney, P. G. 1995. ‘Mandatory disclosure as a solution to agency problems’,
University of Chicago Law Review 62: 1047–112.
Manne, H. G. 1965. ‘Mergers and the market for corporate control’, Journal of
Political Economy 75: 110–26.
Masulis, R. W., C. Wang and F. Xie 2007. ‘Corporate governance and acquirer
returns’, Journal of Finance 62: 1851–89.
Maury, B. 2006. ‘Family ownership and firm performance: Empirical evidence from
Western European corporations’, Journal of Corporate Finance 12: 321–41.
Mayer, C. 2013. Firm Commitment, Oxford University Press, forthcoming.
McCahery, J. A., P. Moerland, T. Raaijmakers and L. Renneboog (eds.) 2002,
Corporate Governance Regimes: Convergence and Diversity, Oxford
University Press.
Miller, G. P. 2004. ‘Catastrophic financial failures: Enron and more’, Cornell Law
Review 89: 423–55.
Morck, R. K. and L. Steier 2005. ‘The global history of corporate governance: An
introduction’, in Morck, R. K. (ed.). A Global History of Corporate
Governance, National Bureau of Economic Research and University of
Chicago Press, pp. 1–64.
Mülbert, P. O. 2010. ‘Corporate Governance of Banks after the Financial Crisis –
Theory, Evidence, Reforms’, ECGI Law Working Paper 130/2009 (new
version: April 2010).
Murphy, K. J. and M. C. Jensen 2011. ‘CEO Bonus Plans: And How to Fix Them’,
Harvard Business School NOM Unit Working Paper 12–022.
Polo, A. 2007. ‘Corporate Governance of Banks: The Current State of the Debate’,
Munich Personal RePEc Archive Paper 2325.
Posner, R. 2009. ‘Are American CEOs overpaid, and, if so, what if anything should
be done about it?’, Duke Law Journal 58: 1013–47.
Pozen, R. C. 2003. ‘Institutional Perspectives on Shareholder Nominations of
Corporate Directors’, Harvard Law School John M. Olin Center for Law,
Economics and Business Discussion Paper 429.
Prevost, A. K., R. P. Rao, and M. A. Williams 2009. ‘Labor Unions as Shareholder
Activists: Champions or Detractors?’, Working Paper, Ohio University.
Ribstein, L. 2003. ‘Bubble laws’, Houston Law Review 40: 77–97.
Riskmetrics 2009. Study on monitoring and enforcement practices in corporate
governance in the Member States, available at http://ec.europa.eu/internal_
market/company/ecgforum/studies_en.htm.
Rock, E. B., P. Davies, H. Kanda, and R. Kraakman 2009. ‘Fundamental changes’,
in Kraakman (et al.), pp. 183–224.
Rock, E. B. and M. L. Wachter 2001. ‘Islands of conscious power: Law, norms, and
the self-governing corporation’, University of Pennsylvania Law Review
149: 1619–700.
Roe, M. J. 1994. Strong Managers Weak Owners, Princeton University Press,
Princeton and Oxford.
2002. ‘Corporate law’s limits’, Journal of Legal Studies 31: 233–71.
2005. ‘The institutions of corporate governance’, in Ménard, C. and
M. M. Shirley (eds.). Handbook of New Institutional Economics, Springer,
Dordrecht, pp. 371–400.
Scharfstein, D. S. 1988. ‘The disciplinary role of takeovers’, Review of Economic
Studies 55: 185–99.
Schmidt, R. H. 2004. ‘Corporate governance in Germany: An economic
perspective’, in Krahnen, J. P. and R. H. Schmidt (eds.) The German
Financial System, Oxford University Press, pp. 386–424.
Senior Supervisors Group, ‘Observations on Risk Management Practices during
the Recent Market Turbulence’, Report, 6 March 2008.
Shleifer, A. and L. H. Summers 1988. ‘Breach of trust in hostile takeovers’, in
Auerbach, A. J. (ed.), Corporate Takeovers: Causes and Consequences,
University of Chicago Press. pp. 33–68.
Shleifer, A. and R. W. Vishny 1997. ‘A survey of corporate governance’, Journal of
Finance 52: 737–83.
Skeel, D. A., Jr. 2011. ‘Inside-out corporate governance’, Journal of Corporation
Law 37: 147–201.
Sorkin, A. R. 2009. Too Big to Fail: The Inside Story of how Wall Street and
Washington fought to save the Financial System and Themselves, Viking,
New York.
Sraer, D. and D. Thesmar 2007. ‘Performance and behavior of family firms:
Evidence from the French stock market’, Journal of the European
Economic Association 5: 709–51.
Stulz, R. M. 2008. ‘Risk management failures: What are they and when do they
happen?’, Journal of Applied Corporate Finance 20: 39–48
Tirole, J. 2006. The Theory of Corporate Finance, Princeton University Press,
Princeton and Oxford.
Weil Gotshal & Manges 2002. Comparative Study of Corporate Governance Codes
relevant to the European Union and its Member States – Final Report and
Annexes I–III available at http://ec.europa.eu/internal_market/company/
otherdocs/index_en.htm.
Williamson, O. 1985. The Economic Institutions of Capitalism, The Free Press,
New York.
1988. ‘Corporate finance and corporate governance’, Journal of Finance 43: 567–91.
1996. The Mechanisms of Governance, Oxford University Press.
2008. ‘Corporate boards of directors: in theory and in practice’, Journal of Law,
Economics & Organization 24: 247–72.
Woidtke, T. 2002. ‘Agents watching agents? Evidence from pension fund owner-
ship and firm value’, Journal of Financial Economics 63: 99–131.
2
European corporate governance codes

and their effectiveness
eddy wymeersch
1. Implementation and enforcement of corporate

governance codes
The purpose of this chapter is to investigate the instruments adopted to
support the implementation of the corporate governance codes in the
European financial markets. It is based on comparative research of
the practices in a selected number of Western European jurisdictions.
The issue of the implementation of the corporate governance codes
has received ample attention, both in studies ordered by the EU
Commission1 and in academic research in Europe and around the
world2 (Bajpai et al. 2005; Millstein et al. 2005; Wymeersch 2005;
Mohd and Nariman 2007; Pietrancosta 2011). The formal degree of
implementation, namely, the often high numbers of companies that in
their corporate governance statement claimed to have implemented the
applicable governance code, is not the direct subject of this investigation,
1
Comparative Study Of Corporate Governance Codes Relevant to the European Union And
Its Member States, January 2002, research undertaken by Weil Gotschall and Manges,
LLP, European Association of Securities Dealers and European Corporate Governance
Network, January 2002, available at http://ec.europa.eu/internal_market/company/docs/
corpgov/corp-gov-codes-rpt-part1_en.pdf; see also Berglöf and Claessen (2004) pointing
to the complementarity of public and private tools of enforcement, stating that: ‘political
economy constraints resulting from the intermingling of business and politics, however,
often prevent improvements in the general enforcement environment.’
2
See also for examples of the worldwide interest in effectiveness of corporate governance
provisions in the Malaysia Code, Corporate Governance issues and enforcement activ-
ities of the Malaysian Corporate Regulators, February 2007, calling for stronger regu-
latory involvement in the absence of shareholders’ action, www.clta.edu.au/professional/
papers/conference2007/2007AS_CGIEAMCR.pdf; cf. the Nigerian code. A similar trend
can be seen in Nigeria: Odidison Omankhanlen, ‘CBN urges SEC to enforce corporate
governance code, November 2011, available at www.tribune.com.ng/index.php/money-
market/31895-cbn-urges-sec-to-enforce-corporate-governance code.
67
68 eddy wymeersch
but rather, the question as to whether attention is paid to non-

implementation, or to nominal, implementation including formal
explanations. A different but related question concerns the verification
by national bodies of whether spurious explanations are further inves-
tigated and whether corrective action is requested or imposed. On the
basis of this comparative overview, some conclusions can be drawn
pointing towards improvements of implementation techniques, or
even changes in the overall framework.
2. Relationship of the corporate governance codes with the

legal environment
By way of introductory remark, attention should be paid to the nature of
corporate governance codes and the framework within which they have
been developed. There are considerable differences in this respect, from a
purely self-regulatory instrument without any monitoring of its imple-
mentation, to a regulation of public character with regulatory monitor-
ing. These differences are deeply embedded in the economic and social
framework within which these codes have been developed.
Implementation of corporate governance rules in Europe is very much
linked to the existence of the corporate governance codes that have now
been adopted in all European jurisdictions. They are based mainly on
self-regulation, whereby the internal bodies of the company are expected
to ensure that the company is run and that its internal bodies act in
accordance not only with the rules of company law, but also with the
governance code or its internal rules of organisation. The action is based
on ex-ante implementation by the company itself, along with some ex-
post pressure exercised by internal and external influences or decisions:
e.g. shareholders in the Annual General Meeting (AGM) or as activist
investors, pressure groups, the press, corporate governance commis-
sions, securities regulators, to name but a few.
The relationship between company law and regulation and the legal
environment has been changing recently under the pressure of the
financial crisis, whereby a certain number of topics have been transferred
from a self-regulatory code to a strictly legally binding provision.
Particularly in the field of remuneration, the unwillingness voluntarily
to upgrade, adopt and effectively implement the code provisions has led
to legislative action in several states. This issue illustrates the oft-
mentioned tension between the codes and the law, and the fear that
effectiveness of corporate governance codes 69
over time the law would absorb much of the codes’ substance, which is
one of the drivers to improve on the content of the codes.
The meaning of a corporate governance code has to be read differently
depending on the applicable legal framework within which these codes
have to be placed: as the laws are quite – and increasingly – different, the
role of codes, as a complementary source of conduct rules, varies sub-
stantially. The monitoring effort will also vary, but not necessarily
proportionately. Linked to the insertion of the codes in the legal frame-
work is the question whether the law takes into account the existence and
the provisions of the code, and whether legal remedies may be attached
to breaches of the codes. Here, again, the situation is quite diverse, at
least, as far as case law is concerned.
In any case, one should mention that the overall corporate governance
system cannot be judged on the mere provisions of the codes, and that
usually very substantial governance conduct rules are laid down in the
provisions of the Companies Act, or other legislation (especially finan-
cial regulation). The present analysis is limited to the governance codes.
Adoption of a corporate governance code has become mandatory on
the basis of Article 46a of the Fourth Directive on Company Law that in
its amended reading states:
A company whose securities are admitted to trading on a regulated

market within the meaning of Article 4(1), point (14) of Directive
2004/39/EC of the European Parliament and of the Council of 21 April
2004 on markets in financial instruments (1) shall include a corporate
governance statement in its annual report. That statement shall be
included as a specific section of the annual report and shall contain at
least the following information:
(a) a reference to:
(i) the corporate governance code to which the company is subject,
and/or
(ii) the corporate governance code which the company may have
voluntarily decided to apply, and/or
(iii) all relevant information about the corporate governance prac-
tices applied beyond the requirements under national law.
Where points (i) and (ii) apply, the company shall also
indicate where the relevant texts are publicly available; where
point (iii) applies, the company shall make its corporate gov-
ernance practices publicly available;
(b) to the extent to which a company, in accordance with national law,
departs from a corporate governance code referred to under points
(a)(i) or (ii), an explanation by the company as to which parts of the
70 eddy wymeersch
corporate governance code it departs from and the reasons for doing
so. Where the company has decided not to apply any provisions of a
corporate governance code referred to under points (a)(i) or (ii), it
shall explain its reasons for doing so.
Most of the Member States had already developed a national corporate

governance code before the transposition of this provision. As provided
in the Directive, in some Member States companies are allowed to
choose a code other than the national one. Depending on the national
law, the companies publish a corporate governance statement in their
annual report, or in a separate brochure. As part of the annual report, the
statement may, according to national law, be subject to a review by the
auditor (Bratton 2003; Bedard and Johnstone 2004; Coffee 2005; Wong
2005). This would normally not be the case for statements that are
published separately (e.g. a corporate governance ‘charter’, as required
in some Member States).
The Directive refers to the ‘comply or explain’ approach, requesting
identification of the parts of the code where the company departs from it
and, more importantly, the reasons for doing so. In principle, it does not
request companies to detail the measures they have adopted to conform
to the code, meaning that they can simply state that the code is fully
implemented (see the practice in Germany). In practice, most companies
elaborate on their implementation of the respective code provisions,
even when they conform with the applicable code.
3. Implementation of corporate governance codes

Before analysing enforcement of corporate governance codes, one
should have a more precise view about how implementation takes
place within the companies and how companies report it. As the codes
deal with a wide range of subjects, some of the information is fairly
traditional, historical or standardised, and may remain the same over
several years. However, other information is more sensitive and needs to
be reviewed on an annual basis; this revision would normally be pre-
pared by the company secretary and reviewed by the chairman of the
board. As part of the annual report, the statement will be approved by
the board and submitted to the general meeting. The formal adoption of
the reference to the corporate governance code and the approval of the
corporate governance statement by the board are important aspects of
the overall regime, raising the question as to whether these decisions aim
merely at disclosure, or could lead to binding the company to the
positions taken by the board. Case law has not been very supportive of
that line of reasoning and generally has not held that the code is binding
or that third parties could derive rights from it. The case in which the
general meeting would adopt – and not merely take note of – the
corporate governance statement has not been tried in practice; its impact
would be all the more significant, as it might affect the relationship
between the AGM, the shareholders and the board.
4. Measuring implementation of corporate governance codes

Much attention has been paid to measuring the extent to which the
provisions of the codes have been adopted – i.e. a formal reference is
made to the applicable code – and implemented – i.e. the code’s provi-
sions are both applied or derogated from – as, in the latter case, an
explanation is due. Elaborate statistics have been drawn up leading to the
analysis of how many of the provisions have been applied, and in the case
of derogations, how many explanations have been given. These statistics
give an indication of the overall implementation and further specify to
what extent specific provisions have been implemented. The statistical
tables also give a good view of the way governance issues are dealt with in
the different jurisdictions. However, in some states, alternative tables
have been published indicating a noteworthy lower degree of implemen-
tation. This difference may point to a difference in perception, linked to
the position of the body that makes the evaluation, e.g. the assessment by
the corporate governance commission or by an investor protection
association.3
As a large number of the codes’ provisions really do not stir much
debate, the degree of implementation would better be measured by
reference to the more controversial provisions, most conspicuously
those on remuneration. In the latter case, the degree of implementation
has generally been considerably lower, leading in some states to legis-
lative intervention. Obviously the self-regulatory instruments have not
been able to ensure effectiveness of these provisions.
Implementation also raises the more important question as to whether
the statement is a merely formal disclosure, or whether it corresponds to
3
See in the Netherlands the figures published by VEB (the Dutch Investors Association) in
Effect, 2009, n. 26 42, available at www.veb.net; also VEB Effect, 2009, n. 26 42, nt. 4; cf.
Portugal, Relatório anual sobre o governo das sociedades cotadas em Portugal 2009,
Tabela IX, available at www.cmvm.pt.
72 eddy wymeersch
the actual conduct of the company. This remark addresses both cases
where the company asserts full implementation and when it gives an
explanation for derogatory conduct. Most statistical overviews do not
probe very deeply into the matter, as researchers do not have investiga-
tory powers, nor do they engage with the company’s management in
order to verify the information. Hence, there are some inherent limita-
tions as to the verification of the veracity of the information disclosed.
Another handicap also mentioned by some governance studies is the
case of meaningless explanations,4 whether boilerplate, information
identical from year-to-year, or explanations that refer to specific circum-
stances that are, however, not further elaborated upon. Information of
that type should be discarded as a valid ‘explanation’. No cases have been
reported where corporate governance statements have been deliberately
false or misleading, but the hypothesis should not be excluded. In a
broader context, false statements, on corporate governance or on any
other subject, may lead to civil or criminal liability, depending on the
national legislation. No cases specifically relating to corporate govern-
ance items have been found, and the causality requirement would prob-
ably bar any civil liability.
5. Drivers for implementation

The implementation of the corporate governance codes, most of which
are largely voluntary, takes place under internal and external pressure.
Company boards normally feel strongly implicated in reporting their
activity and that of the company, as it makes their action more credible
and contributes to building confidence in the board and in the
company’s business. Boards have become more conscious of the import-
ance of their role in this respect, organising specialised governance
committees, being very elaborate on some aspects of the code, proceed-
ing to self-assessment, and in some cases, external assessment. In
Europe, only a handful of companies still seem to resist paying any
attention to the governance matter. It is quite striking that adoption of
the codes also takes place in companies with concentrated ownership.
The shareholders are the first addressees of these reports and deter-
mine their position – in votes, in trading policy – on the basis of the
performance of the company, including its governance. In some juris-
dictions, the auditors verify data in the reports. The investors, acting in
4
According to one observer, this relates to about 30 per cent of the statement.
the AGM, request good governance and focus on perceived weaknesses.

Institutional investors, whose large blocks are often untradeable, have no
choice but to enter into a more continuous dialogue. Activist investors
pay attention to governance issues and have requested changes in gov-
ernance practices. Very recently, investors have exercised pressure on
management’s remuneration within the non-binding ‘say on pay’ regime
(Thomas et al. 2012; Cheffins and Thomas 2001).5
Some of the pressure originates from outside the company. The
markets are important levers for requesting ‘better’ governance from
companies and indirectly have been able to urge some important
changes. The press and specialised media act as a conveyor belt for
information to the market participants. Some firms publish corporate
governance ratings, pretending that better ratings contribute to higher
returns, a statement that enjoys much support in the advisory world, but
for which hard empirical evidence seems rather controversial.
Last, but not least, is the action developed by governance commissions
who, apart from drawing up the governance codes, in several states are
also involved in monitoring, and/enforcing the corporate governance
provisions. As illustrated in the overview below, there is a wide diversity
of systems, structure and practices with respect to the way in which these
commissions deal with their relationship to listed companies to which
the codes are applicable.
6. The scope
The corporate governance codes generally apply only to listed compan-
ies, defined as listed on a stock exchange, or on a multilateral trading
facility. Most codes mention that their provisions may also serve as a
source of inspiration for unlisted companies.
The legal domicile of the company is generally used as the connecting
factor, but some codes also take into consideration the place of trading.
The European Corporate Governance Forum has issued a statement
relating to the case where different codes would be applicable.6
5
See also Financial Stability Board, Implementing the FSB Principles for Sound
Compensation Practices and their Implementation Standards, 13 June 2012 and the FSF
Principles for Sound Compensation Practices, 2 April 2009.
6
See European Corporate Governance Forum, Statement of the European Corporate
Governance Forum on Cross-border issues of Corporate Governance Codes (2009), avail-
able at www.ec.europa.eu.
74 eddy wymeersch
The discussion about corporate governance codes has been largely

superseded in the banking sector, where specific recommendations and,
in the future, hard supervisory law provisions,7 will govern the internal
governance of the banks (Hopt and Wohlmannstetter 2011).
7. Comparative country analysis

This analysis is based on the description of the corporate governance
codes and their implementation practice in a selected number of
European jurisdictions. It allows us to situate the codes on the back-
ground of the social and economic context in which they have been
developed and reflects the concepts adopted by the respective business
leaders. As will be further illustrated, there are significant differences
between the codes and the context in which they function, which there-
fore makes a comparison difficult (Coombes and Wong 2004). Therefore
it would be superficial to make a horizontal comparison between the
different codes. The method followed here is a bottom-up comparison,
country by country, as this allows us to position each code in its overall
context, dealing, for example, with the nature of the code, the composi-
tion of the monitoring committees, the follow-up methods and the
structure of the securities market, etc. The description is essentially
based on publicly accessible information.
The countries compared have been selected on the basis of their prom-
inent position in the corporate governance debate, and the accessibility of
the materials in the original language. Moreover, they represent several
governance models, i.e. one-or two-tier boards, concentrated or dispersed
ownership, self- or public regulation, and similar differences. This selection
also reflects the bias – and limitations – of the author, and does not indicate
that other jurisdictions do not have a valid governance practice. The
countries selected are:
* Austria;
* Belgium;
* Denmark;
* France;
* Germany;
* Italy;
7
See CRD IV, Directive 2013/36 of 20 June 2013, OJ 27 June 2013, L176/338. These
provisions have been severely criticised by Winter (2012a).
* Luxembourg;
* the Netherlands;
* Portugal;
* Spain;
* Sweden;
* Switzerland;
* UK.
7.1. Austria
Austria adopted a corporate governance code in 2002, which was drawn
up by the Austrian Corporate Governance Commission.8 The commis-
sion was composed of a wide group of representatives of different stake-
holders in the governance field.9 Among them were two representatives
of the financial regulator, FMA, and one representative of the govern-
ment in charge of the capital market.10 The code is updated every year,
most recently in January 2012. The listing rules of the Vienna Stock
Exchange oblige listed firms to adopt the code as a condition for access to
the first segment of the market.11 On the basis of the Enterprise Law,12
listed companies13 are required to publish a corporate governance state-
ment in which they designate a code that is generally applicable in
Austria or in the market of listing, or to state the reasons for not applying
any code. The law does not require companies to elaborate on the
provisions of the code with which they comply, only to state the reasons
for non-compliance with certain provisions. Information about compo-
sition of the supervisory and management board and their subcommit-
tees and the way they function is mandatory. Also mandatory is a
8
Österreichischen Arbeitskreises für Corporate Governance, available at www.corporate-
governance.at. The code was drafted on the basis of proposals by the Institute of
Austrian Auditors (IWP) and the Austrian Association for Financial Analysis and
Asset Management (ÖVFA).
9
Academics, auditors, two members of the financial regulator FMA, one of the ministry,
and further representatives of the investor associations, the stock exchange, listed
companies and practising lawyers.
10
‘Regierungsbeauftragter für den Kapitalmarkt’, see www.wienerborse.at/beginner/lexi
con/18/876.
11
Companies have to subscribe to the code by a ‘Verpflichtungserklärung’.
12
Art. 243b Unternehmensgesetzbuch.
13
Reference is made to listing of shares, but also to companies that only have other
securities listed on the Stock exchange or on an MTF.
76 eddy wymeersch
statement regarding the measures adopted to support the presence of

women in these different boards.14
The code consists of a very detailed questionnaire (76 questions) based
on a ‘comply or explain’ technique. It is divided into three sections: L for
the legal provisions, C for those that apply on a ‘comply or explain’ basis
and R for the recommendations. Apart from the general provisions
included in most codes, the code pays specific attention to the position
of the shareholders, issues of conflict of interest and the position and role
of the auditors.
No mention has been found concerning monitoring of the individual
corporate governance statements.
7.2. Belgium
The Belgian corporate governance code was originally developed by the
Corporate Governance Commission, as a non-governmental initiative,
on the basis of an agreement between the principal employers’ associa-
tion, the Brussels Stock Exchange, and the securities regulator, which
was reflected in its original composition. Since then its composition has
been broadened to include the persons active in the Institute of Auditors,
the Federation of Pension Funds and the Federation of Investment
Clubs.15 Since its original version of 2004, the code was updated in
2009, which is now the version that has been officially designated16 as
the code applicable to all listed companies pursuant to the Companies
Act.17 The code applies to companies with shares listed on a stock
exchange, but is recommended as a reference for other companies. It is
essentially based on a ‘comply or explain’ approach; however, since its
original adoption an increasing number of provisions that were part of
the code have been introduced into the law. Regarding certain aspects,
the code calls for stricter requirements than those required by the law,
14
Art. 243b, § 2 Unternehmensgesetzbuch.
15
It is composed of 23 members, several from industry, some academics, and persons from
different professions.
16
Royal Decree, 6 June 2010 designates the Belgian Corporate Governance Code,
December 2009, as the code of reference; available at www.corporategovernancecom
mittee.be.
17
Article 96 § 2 Companies Act, also referring to a series of additional information items.
The same provision enables a governance code to be designated by Royal Decree of 6
June 2010.
also clarifying the substance of several of the requirements laid down in

the law.
A substantial body of corporate governance rules is now laid down in the
Companies Act, as amended by the 2010 ‘Corporate Governance law’18 and
other legal acts.19 The law requires a substantial number of disclosures that
are relevant from the corporate governance point of view, referring not only
to the applicable code, but also to practices that go beyond the code require-
ments,20 making these disclosures mandatory.
This parallel set of rules has created some confusion, as both call for a
corporate governance statement, but with a somewhat different con-
tent.21 Of course, only the Act is enforceable in court. The securities
regulator, the FSMA, also acts to enforce the legal provisions within the
context of its vetting of prospectuses, or reviewing some of the annual
reports. It has, for example, held company directors to good governance
practice, i.e. arguing that independent directors should not receive
income-related remuneration (e.g. share options). Obviously this action
has not been successful.
The monitoring of the implementation of the code takes place, accord-
ing to the Preamble to the code, in:
a combined monitoring system that relies on the board, the company’s
shareholders, the statutory auditor and the Banking, Finance and
Insurance Commission (CBFA, now FSMA), as well as other possible
mechanisms.
The code refers to the corporate governance charter that is posted on the
company’s website and contains the main elements of the company’s
policies in this respect, and the corporate governance statement that
is part of the annual report, and contains, apart from a reference to
18
Law of 6 April 2010 ‘for the strengthening of good governance of listed companies and
independent public sector companies, and modifying the regime of the banking and
financial sector’, or ‘Wet tot versterking van het deugdelijk bestuur bij de genoteerde
vennootschappen en de autonome overheidsbedrijven en tot wijziging van de regeling
inzake het beroepsverbod in de banken financiële sector’, B.S., 23 April 2010.
19
See the Act on Gender Diversity L. 28 July 2011, BS 31 August 2011, and art. 96 § 2
Companies Act; also in the financial sector, the provisions dealing with banks’ gover-
nance and the implementing circulars, see art. 20 et seq. L. 22 March 1993 on the legal
status and supervision of credit institutions, B.S., 19 April 1993.
20
Article 96, §2, 1st al., Companies Act.
21
In fact, the ‘Declaration of Good Governance’ provided for in the law regroups some
information items for which disclosure was already mandatory, but that has now been
regrouped. Both documents must contain a remuneration report to be adopted by the
companies, but that imposed by the law imposes some additional disclosures.
78 eddy wymeersch
the applicable code, a more factual analysis of the actual corporate

governance practices. It is the latter that is also addressed in the above
mentioned Companies Act provisions.
Statistical monitoring reports based on the ‘corporate governance
charter’ and ‘corporate governance statements’ have been published by
a consortium of VBO, the employers’ organisation, and Guberna, the
Institute of Directors, relating to the 20 companies that are part of the
most important market index, the BEL 20. Grant Thornton has also
published an overview of the corporate governance practices in Belgium,
leading to comparable findings.22
The FSMA has published analytical reports on compliance with the
code, covering all companies that are listed on the exchange. The latest
report was published in September 2011,23 relating to the 2010 corporate
governance statements. From the comparison of these reports it appears
that the degree of compliance is significantly higher for the ‘BEL 20’
companies – approaching 100 per cent – while in the wider group of
listed companies and depending on the specific items, compliance is
substantially lower, notwithstanding significant improvements.24
In its 2011 report, the FSMA formulated a considerable number of
recommendations to increase the level of compliance, particularly in the
following fields: remuneration – now largely covered by the Companies
Act – board evaluation, internal controls and risk management.25 It was
found that the ‘explain’ approach was not always followed with sufficient
strictness, in some cases being completely ignored. If certain provisions
of the code are considered non-applicable, companies should state this
explicitly, giving the relevant explanations, as otherwise it gives the
impression that the company has complied. Remuneration continues
to be a point of interest, and disclosure could be improved with respect to
pensions, as this data was generally missing.
From both monitoring documents it appears that the follow-up prac-
tices are analysed from a more formal point of view, essentially
22
Grant Thornton, ‘Corporate Governance Review 2011: Listed companies make progress
in applying corporate governance regulation’, available at www.grantthornton.be; see
for comparison, Heidrick and Struggles, Challenging Board Performance, European
Corporate Governance Report 2011, available at www.heidrick.com.
23
FSMA, ‘Les premières déclarations de gouvernement d’entreprise: étude de suivi de
l’Etude n° 38’, Etudes et documents nr 40, available at www.corporategovernancecom
mittee.be.
24
Improvements particularly in the fields of risk management and internal controls.
25
Elements that are now part of the CG law of 6 April 2010, nt. 21, see art. 96 § 2, 3rd,
Companies Code.
evaluating whether the necessary explanations have been given, yet

rarely criticising their substance. In some cases the FSMA reminds
companies of their obligations on the basis of the new law.
The Corporate Governance Commission has also published a guide-
line for a meaningful ‘explain’ approach.26
The auditors are required to certify whether the annual report con-
tains the disclosures that have been mandated by the law and correspond
with the data in the annual accounts as have been certified by them.27
They do not report on corporate governance issues in general.
Institutional investors who participate in listed companies rarely
exercise significant corporate governance action (Van der Elst 2010;
2012). An attempt by an investor protection association to undertake
activism through a specialised investment fund that owned shares in all
Belgian listed companies has proved unsuccessful and was abandoned.
Belgian case law on the corporate governance code is limited to one
decision, Fortis.28 Fortis was a bi-national company, in the sense that the
Fortis share consisted of two shares (‘stapled shares’), one of the Belgian
parent and one of the Dutch parent. In principle, both companies would
have to concur. According to Dutch law, certain important decisions
have to be submitted to the AGM of the Dutch company, while no
similar provision exists on the Belgian side. In order to bridge this
difference, the Fortis articles of incorporation contained a provision
according to which the board would decide in accordance with the
Fortis Governance Statement. This statement contained a provision
providing for a requirement similar to that applicable under Dutch
law. As a consequence, for identical matters a decision by the Belgian
AGM would be required. In summary proceedings, the Court of Appeal
of Brussels overruled the first instance judge and held that the transfer of
the bank activities to a third party without the agreement of the share-
holders was apparently ‘seriously illegal’ and could have been declared
void. Therefore the court decided to suspend it, recognising that the
statement has a certain legal force. However, the argument was not
pursued in the case on the merits.
26
Corporate Governance Commission, ‘The Corporate Governance Commission helps
Companies to Draw Up a Meaningful “Explanation”’, Press release, 14 February 2012.
27
Art. 144, § 6 Companies Act.
28
See Cass., 19 February 2010, Revue pratique des sociétés, 2009, nr. 7009, 421 and Court
of Appeal, Brussels, 12 December 2008, Revue pratique des sociétés, 2009, nr. 7010, 432
and the comments by De Cordt (2009).
80 eddy wymeersch
7.3. Denmark
Denmark had first adopted corporate governance recommendations in
December 2001, and the last revision to date was in 2010. The legal basis
of the code is section 107b of the Financial Statements Act, declaring that
the corporate governance statement will be part of the management review
in the annual report. The code was drafted by the Committee for Corporate
Governance,29 who recommended the code for adoption by the board of
Nasdaq OMX, which implements the code in its listing rules. This commit-
tee was originally composed of four prominent Danish personalities, while
the present committee comprises 9 independent persons (Andersen 2004).
The code is referred to as containing good practice provisions and is based
on the ‘comply or explain’ approach.30
The code pays ample attention to the relations of the company with its
shareholders, especially the institutional shareholders, referring to the
notion of ‘active ownership’ as mentioned in the EU recommendation of
30 April 2009.31 It refers to the concept of wider stakeholdership.
Strikingly, the code contains a provision on board neutrality in takeover
cases32 and other provisions dealing with the board’s obligation. The
code calls for a comprehensive corporate governance statement that
takes a position on each of its items, and is part of the management
report in the company’s annual report. Publication on the company’s
website is a valid alternative. There is no updating during the year.
There is no monitoring of the code, except that the exchange assesses
whether the explanations are understandable. But it is clearly stated that
the exchange ‘does not intend to assess whether the content of an
explanation is good or bad’.33
7.4. France
The monitoring of corporate governance rules in France should be
looked at from multiple viewpoints; apart from the elaborate rules of
company law, the monitoring action by the securities regulator34 and the
29
‘Komiteen for god Selskabsledelse’.
30
Section 107b and Rule 4.3 of the Rules for Issuers of Shares of Nasdaq OMX Copenhagen
1–7–2010.
31
Recommendation 2009/385, Preamble 10.
32
See the provisions in the Danish Companies Act.
33
www.corporategovernance.dk.
34
On the basis of art. L.621–18–3 of the Code monétaire et financier, the AMF has
requested companies to publish an annual report on corporate governance.
two main ‘codes of conduct’, one drawn up by the employers’ organisa-

tions, a second by the asset management bodies, must both be integrated
in the overall view. The securities market supervisor AMF (Autorité des
Marchés Financiers) plays a very strong role, essentially at the regulatory
and advisory level. The AMF has been mandated by law to evaluate
corporate governance in general, irrespective of the source on which the
provision or practice is based. The AMF sees its role as ‘supportive of
good corporate governance practices with a view of stimulating their
adoption by issuing recommendations and analysis for further
development’.35 On this basis, it has published 8 annual reports. In its
most recent report, it assesses the compliance with the legal require-
ments and the way companies have dealt with the disclosure called for in
the self-regulatory instruments. From this integrated approach, one can
derive that, according to the AMF, the system has to be looked at as a
whole, and that the self-regulatory provisions are partly in addition to,
partly an extension of the legal requirements.
Several important corporate governance provisions have been intro-
duced in the Companies Law, and in these last few years in the Code
monétaire et financier, making the system based largely on hard law. The
attention given to corporate governance issues has increasingly become a
matter of application of the law and regulations. In its annual reports on
governance, the AMF analyses both the implementation of the govern-
ance provisions and the evolution of practices by the listed companies,
on the basis of which it criticises existing practices or formulates
‘recommendations’ which, without being legally binding, are expected
to be followed. The AMF’s annual report mentions cases in which the
recommendations were not followed,36 each year repeating the recom-
mendations that were made in the previous report.37 The pressure for
adoption of the recommendations is therefore not negligible. In addi-
tion, some recommendations propose changes, some of which go beyond
the existing obligations.38 Research studies on specific items, for exam-
ple, on audit committees, or the functioning of the general meeting, have
35
“Faire état des bonnes pratiques des entreprises en matière de gouvernance et d’en
favoriser le développement à travers la formulation de recommendations et de pistes de
reflexion’ (AMF, Rapport sur le gouvernement d’entreprise et la rémunération des
dirigeants, December 2010, p.15) available at www.amf-france.org/documents/general/
10249_1.pdf.
36
AMF 2011 Report, nt. 40 reports on self-evaluation of the board, and on the requirement
to submit to the board the acceptance of board positions in other companies.
37
AMF 2011 Report, nt. 40. 38 Ibid.
82 eddy wymeersch
been undertaken by or under the auspices of the AMF. The latter work-
stream extends the perspective to the role of the shareholders, although
those were usually not addressed in these reports.
The AMF has been invited to give its opinion on matters of remuner-
ation, on the role of the audit committee, the internal controls and risk
management. The initiatives taken by the AMF to urge companies to
strengthen and report on their risk policies39 and the way these must be
dealt with in financial reporting and in the annual accounts should be
mentioned. However, the AMF does not seem to play an important role
in the field of enforcing corporate governance rules.
In February 2012, the AMF published a report on the functioning of
the general meeting of shareholders, focusing on the interaction between
shareholders and issuers, the exercise of the voting rights, the function-
ing of the meeting with special attention for the bureau and the rules
relating to the agreements with conflicting interests, the so-called
‘conventions réglementées’, i.e. the agreements between related parties.40
The report contains conclusions on a certain number of changes in the
Companies Law, in internal practice rules, or in rules concerning the
auditors intervening in some of these procedures.
There are several soft law instruments relating to corporate govern-
ance in France: the main codes are the AFEP-Medef code, applicable to
the largest market capitalisations, and the Middlenext code, for medium
and small listed companies. In addition, one should mention the prop-
ositions of the Institut français des administrateurs. The AFG, or
Association française de gestion financière, has issued recommendations
addressed to the asset managers active in the field of investment funds.
The French corporate governance code (AFEP-Medef) is followed
by almost all French companies trading on the French official market.41
The code has been developed by two associations – the AFEP42 and
the Medef43 – and was republished in a consolidated version in
December 2008 to allow for the incorporation of the remuneration
39
AMF, ‘Recommendation de l’AMF sur les facteurs de risque’, 29 October 2009.
40
AMF 2012,‘Report of the Working Group on General Meetings of Shareholders of Listed
Companies’, available at www.amf-france.org/documents/general/10334_1.pdf.
41
For the list, see 2010 Report AFEP-Medef, indicating that one French company did not
apply the code (annex 2).
42
Association française des enterprises privées groups all major listed French companies.
It was created in 1982. See www.journaldunet.com/economie/enquete/afep/afep.shtml.
43
Mouvement des entreprises de France is the largest employers’ association, with 700,000
members, 50 per cent of which are SMEs.
rules, representing about half of the code’s provisions, the other half
being dedicated to the composition and functioning of the board of
directors. There is no mention of the relationship with shareholders,
with statutory auditors, or with other stakeholders. The AFEP-Medef is
the usual reference code called for by the law.44
The code is based on a ‘comply or explain’ approach. In order to ensure
its application, the two associations declare that they will analyse the
information published by the companies that are part of the SBF 120, the
index of the Société des bourses françaises. They add that if they determine
that one of its recommendations is not applied, and this without sufficient
explanation, they will submit the issue to the leadership of that company.45
The findings from this action are published in an annual report, of which
three have now been released.46 The said reports analyse the different
recommendations, giving statistical data about the options chosen by the
companies, for example on the structure of the board, the number of
directors, the number of directorships occupied by a director, or about
gender diversity, providing an interesting image of the top French corporate
world. The report also reproduces explanatory statements of companies
that did not comply with the code’s recommendations, giving an overview
of the diversity of arguments used for diverging from the code. However,
the report does not comment on the explanations given. Moreover, there is
no evidence of follow-up action by the said two associations as far as non -
compliance by these companies is concerned.
A third interesting player in the French corporate governance debate is
the shareholders, especially the institutional shareholders, acting through
the AFG, regrouping the collective and individual asset managers. The AFG
has published a set of Recommendations on Corporate Governance47
essentially dealing with the participation of asset managers in the general
meetings of listed companies and indirectly addressing corporate
44
Loi n°2008–649 du 3 juillet 2008 portant diverses dispositions d’adaptation du droit des
sociétés au droit communautaire modifiant les articles L.225–37 et L.225–68 du code de
commerce: ‘Lorsqu’une société se réfère volontairement à un code de gouvernement
d’entreprise élaboré par les organisations représentatives des entreprises, le rapport
prévu au présent article précise également les dispositions qui ont été écartées et les
raisons pour lesquelles elles l’ont été.’
45
www.code-afep-medef.com/la-mise-en-œuvre-des-preconisations.html.
46
See 2e Rapport annuel sur le code AFEP-Medef, application du code consolidé de
gouvernement d’entreprise des sociétés cotées par les sociétés de l’indice SBF 120,
exercice 2009, November 2010.
47
AFG, Recommendations on Corporate Governance, January 2012, at www.afg.asso.fr/
index.php?option=com_content&view=article&id=98&Itemid=87&lang=en.
84 eddy wymeersch
governance issues that are submitted to a shareholder vote, such as the

organisation of the board of directors and its responsibilities, their remu-
neration and particular resolutions in the AGM such as those dealing with
anti-takeover measures. The indirect legal basis is a legal requirement48
according to which asset managers should exercise the voting rights
attached to their portfolio shares and, if they do not, should explain their
position. The AMF’s réglement général obliges asset managers to publish
their voting policies, the votes cast and the reasons for negative votes or
abstentions.49 The AFG follows up the agendas for forthcoming general
meetings along with its comments for its members (‘alerts’), expressing
criticism with respect to the motions proposed, or indicating how they
should be evaluated on the basis of its recommendations.
In a 2012 publication, the AFG analyses the results of the votes cast in
the general meetings and identifies some interesting trends (Pardo and
Valli 2012). It thus reported on an increase in asset managers’ partici-
pation in more than 80 per cent of the cases referring to AFG recom-
mendations and alerts, while asset managers frequently engage with
listed companies ahead of the AGM directly, or with the assistance of
the AFG. The number of negative votes is quite considerable,50 especially
addressing dilutive capital transactions and director appointments.
These AFG ‘monitoring alerts’51 deserve special mention: the AFG regu-
larly publishes comments on forthcoming AGMs, especially highlighting
48
Loi securité financière 2003, art. 533–22.
49
AMF, Réglement general, art. 314–100 et seq.; it is on this basis that AFG publishes a
‘bilan des votes’, see www.afg.asso.fr/index.php?option=com_content&view=article&
id=106&Itemid=152&lang=fr.
50
In at least 80 per cent of French companies, the asset managers reportedly voted against
at least one motion. The AFG publishes the following list of reasons for casting a
negative vote: (1) dilutive equity financing transactions: capital increases without pre-
emptive rights, capital increases with preference periods, debt issuance, etc. (29 per
cent); (2) appointment of members to boards of directors or supervisory boards:
percentage of inside directors, directorship appointments, etc. (23 per cent); (3) equity
financing transactions considered to be anti-takeover measures: issuance of ‘poison pill’
warrants, share buybacks, etc. (14 per cent); (4) management and employee share-
holding schemes: grants of bonus shares and stock options, executive remuneration,
etc. (13 per cent); (5) approval of regulated agreements (11 per cent); (6) appointment
and remuneration of statutory auditors (4 per cent); (7) changes to constitutional
documents that impact negatively on shareholders’ rights: multiple voting rights and
limitations, amendments to articles of association, etc. (3 per cent); and (8) approval of
financial statements and allocation of net income (3 per cent).
51
SBF 120 alerts emanating from a ‘cellule de veille’: see ‘Programme de veille 2012 de
gouvernement d’entreprise sur les sociétés du SBF120’, available at www.afg.asso.fr/
index.php?option=com_docman&task=cat_view&gid=499&Itemid=151&lang=en.
proposals that run against their recommendations, or raise other governance

issues. It is impossible to give an overview of the numerous alerts published
each year.52 For example, one alert relates to the appointment of an
‘independent’ director proposed for election although he holds a 9.9 per
cent stake in the company, or to an anti-takeover device, under the form of
double voting rights in a company where the functions of chairman and CEO
are not separated, while there is only one independent director.53 These alerts
indicate to the AFG members and to the public where corporate governance
issues lie and how they relate to the AFG recommendations, but without
expressing a formal direction for voting. The alerts merely recall the legal
obligation for French asset managers to exercise their voting rights. In fact,
the alerts implicitly indicate where the proposals diverge from normal
governance practice.
Outside the strict framework of the exercise of voting rights, the AFG
mentions the frequent instances in which asset managers contact issuers
whether informing them about their general voting policies and govern-
ance standards, or their reasons for opposing certain resolutions, while
on the other hand, companies sometimes consult asset managers – or the
AFG – about controversial resolutions.
7.5. Germany
The German Corporate Governance code (Deutscher Corporate
Governance Kodex) dating from February 2002, was developed by the
German Corporate Governance Commission and is published in the
Official Gazette. This Commission was installed in 2001 by the government
(to be more precise, the Ministry of Justice), which appoints its members.54
Membership is composed mainly of academics, business leaders and rep-
resentatives of the stock exchange, and the asset management or investor
protection associations. According to its charter, the purpose of the code is
to strengthen the confidence of German and international investors in
German business by making the German business organisation more trans-
parent and understandable, especially as to the two-tier board structure. It
also aimed at introducing more flexibility in German company regulation,
as the law itself is mandatory.55
52
Ten cases in 2011, and going back to 2001.
53
Circulaire N° 4 concernant Derichebourg (2012) can be cited as an example.
54
It is officially designated as the ‘Regierungskommission’. 55 See § 23(4) AktG.
86 eddy wymeersch
The code is based on the concept underlying German corporate law that
companies are run with a view of their continuity and the creation of added
value on a sustainable basis. This concept is part of the market economy
where the interests of the enterprise and its continuity takes precedence over
the interest of the shareholders. The code has been revised several times;
most recently amendments have been proposed to its 2012 version. The
code is based on a three-pronged series of provisions:
* recommendations that are binding on a comply or explain basis;56
* suggestions57 that are not binding and may be left aside without
disclosure; and
* legal provisions that are per definition binding, and are reproduced
for reasons of clarity.
The use of the code has been made obligatory by § 161 of the Companies
Act (AktG) which implements Article 46a of the European IVth
Directive, as amended. It applies to companies that are either listed or
have other securities – such as bonds – traded on regulated markets,
including multilateral trading facilities (MTFs). The said legal basis
requires companies to deliver a compliance declaration
(Entsprechenserklärung) that is published in the online version of the
Official Gazette. Companies that fully comply with the code can merely
state that they comply in full;58 other companies will have to explain with
respect to which provisions they do not comply, sometimes limiting
themselves to stating that they will not apply the provisions, or that the
provision is not adapted to the company’s situation. The statements
relate to the past, and to the intention of the company for the near
future, the latter not being binding.
The Commission does not engage in monitoring tasks, other than
updating the code. The implementation is followed up by the Berlin
Center of Corporate Governance, led by Professor Axel von Werder,
who has published yearly analytical reports since 2003 (Von Werder and
56
Identified by using the term ‘shall’.
57
Identified by using the term ‘should’ or ‘can’.
58
See, e.g., the full compliance declaration by BMW, avalaible at www.bmwgroup.com/
bmwgroup_prod/d/0_0_www_bmwgroup_com/investor_relations/fakten_zum_unter
nehmen/Entsprechenserklaerung_2011_DE.pdf or partial compliance by BASF, www.
basf.com/group/corporate/de/investor-relations/corporate-governance/index, indicating
that it complies almost fully with the non-mandatory provisions or ‘suggestions’; but
compare Deutsche Bank www.deutschebank.de/ir/de/download/Entsprechenserklaerung_
25_Okt_2011.pdf, mentioning that it will maintain its internal approach to conflicts of
interest notwithstanding a decision of the Frankfurt Court.
Talaulicar 2009).59 In particular, the 2012 report deals with the assess-
ment of the code by the leaders of German listed companies on the basis
of a survey of almost all companies (Von Werder and Bartz 2012). The
survey indicates that the code is generally considered as ‘negative’,60
while there are some items where the persons surveyed considered it
particularly weak (provisions dealing with the cooperation between
Vorstand and Aufsichtsrat).
There are a number of investor associations active in the field of
protection of shareholders’ rights,61 however, it appears that these do
not play a prominent activist role.62 The Vereinigung Institutionelle
Privatanleger, or VIP (Association of Institutional Shareholders), sup-
ports ethical and Environmental, Social and Governance (ESG) objec-
tives and has recently taken an activist position in a prominent case.63
Some associations represent mainly the listed companies;64 others are
mostly concerned with specific ESG issues65 or aim at improving the
functioning of the supervisory boards.66
The perspective of the state, and in particular, state funds as a major
shareholder in nationalised banks, is new (Hopt 2009). It is also worth
mentioning that several other corporate governance codes have been
59
See the reports 2003 to 2012 on www.bccg.tu-berlin.de/main/publikationen.htm. They
have also been published in Der Betrieb.
60
The ‘comply or explain’ concept, or ‘Regulierungskonzept’ received a rather negative assess-
ment: ‘eine eindeutig negative Haltung zur Funktionalität des Kodexregimes’, Survey, Der
Betrieb, 2009, 872; from the industry side, too, questions have been raised about the
usefulness of the code, as opposed to legal obligations: E. Voscherau, ‘Anforderungen an
Aufsichtsratsmitglieder’, Deutsches Aktieninstitut am 26 Oktober 2010; Lufthansa
Aufsichtsratschef Jürgen Weber ‘Perspektiven der Corporate Governance in der nächsten
Dekade’ 64. Dt. Betriebswirtschafter-Tag, 29 September 2010.
61
Deutsche Schutzvereinigung für Wertpapierbesitz; Schutzgemeinschaft der
Kapitalanleger.
62
See Corporate Governance-Kodex für Asset Management-Gesellschaften, 27 April 2005,
www.dvfa.de/files/home/application/pdf/Kodex_CorpGov_AssetMmt.pdf. The code
requires these companies to cast their vote; it refers to its ‘comply or explain’ basis.
63
According to its website: ‘A number of investors, including the U.K.’s Hermes and
German shareowner association VIP have filed a no-confidence motion against the
Deutsche Bank Supervisory Board. Investor complaints include dissatisfaction over
the board’s succession planning for CEO Josef Ackermann as well as misaligned
executive pay and a poor sustainability strategy’. See Reuters, 24 April 2012.
64
Deutsches Aktieninstitut, mainly representing the German listed companies; available at
www.dai.de/internet/dai/dai-2-0.nsf/dai_startup_e.htm.
65
Dachverband der Kritischen Aktionärinnen und Aktionäre e.V., www.kritischeaktio
naere.de/presse.html.
66
Vereinigung der Aufsichtsräte in Deutschland e.V. (VARD).
88 eddy wymeersch
developed, for example, for family firms67 or for firms in which the
German state participates.68 However, it has not been possible to assess
their role.
The legal function of the code has been discussed in legal writings and
has led to a number of court decisions (Gebhardt 2012). One of these
relates to the validity of decisions of company bodies that run against one
of the recommendations of the code. Generally speaking, one could state
that liability would attach to untrue or incomplete statements, but not to
the provisions of the codes as such. Companies should ensure that their
governance statements are always up to date and justify their investors’
reliance. Business leaders have shown real concern about this aspect of
their liability (Von Werder and Bartz 2012).
A first Supreme Court decision69 on the legal position of the German
corporate governance code relates to Kirsch v. Deutsche Bank, where the
latter and its chairman were sued on the basis of a public declaration of
the chairman of the bank about its solvency. The bank had accepted
responsibility for its chair’s statement, but had not declared this fact in
its corporate governance statement, nor the way the boards had dealt
with it, although they were clearly obliged to do so on the basis of the
code’s conflicts of interest provisions. The Court decided that this
omission was legally relevant, being an untrue statement about a sig-
nificant item. It would make the decision of the general meeting con-
cerning the discharge of liability voidable. The code is to be considered
the expression of a general legal rule: although departures from it are
allowed, in the present case, these are departures not from a specific rule,
but from a general principle or norm that is expressed in the code’s
provision (Lutter 2011). The statement is comparable to a prospectus for
issuing securities, where there is reasonable expectation that its content
is true, and that the expectations raised will be founded.
The Court of Appeal of Munich70 upheld the possibility of having the
decision of the AGM set aside for violation of a code provision relating to
67
Governance Kodex für Familienunternehmen – Leitlinien für die verantwortungsvolle
Führung von Familienunternehmen – (version 19 June 2010), available at www.intes-
online.de/UserFiles/File/GovernanceKodexDeutsch.pdf.
68
Public Corporate Governance Kodex des Bundes (Public Kodex), Principles of Good
Corporate Governance for Indirect or Direct Holdings of the Federation, available at
bundesfinanzministerium.de/nn_39010/DE/Wirtschaft__und__Verwaltung/Bundesliege
nschaften__und__Bundesbeteiligungen/Public__corporate__governance__Kodex/Anlag
ePCGKengl,templateId=raw,property=publicationFile.pdf. These principles are declared
to be applicable from 1 July 2009. They do not apply to listed companies.
69
BGH, 16 February 2009, II ZR 185/07. 70 OLG Munich, 6 August 2008, 7 U 5628/07.
the age limit of members of the Supervisory Board. The nullity of the
company’s decision – not decided for factual reasons – was justified by
the fact that the company bodies had not adapted the governance state-
ment, although the company had committed to do so, and shareholders
would have been entitled to rely on it, implying that the statement had to
be adapted for changes intervening during the year. But the decision
could not be based on the formal provisions of the law sanctioning its
violation of the company’s charter, as the code is not a legal instrument,
and not being enacted by government, consists merely of conduct of
business rules.
7.6. Italy
The corporate governance code (Codice di Autodisciplina)71 has been
developed by a Committee for Corporate Governance, composed of
major business leaders, and supported by the main business organisa-
tions,72 including the associations for institutional investors and Borsa
Italiana S.p.A. The committee is part of the organisation of the Italian
Stock Exchange, which follows up the application of the code indicating,
where necessary, possible improvements. The code can be viewed as an
instrument for preparing companies for a stock exchange listing. Its
latest version dates from December 2011.
The law provides that listed companies must publish in their manage-
ment report a ‘report on corporate governance and ownership structure’,
the content of which is determined in the law itself.73 In the same report
companies are required to give information with respect to the ‘adoption
of a corporate governance code of conduct issued by regulated stock
exchange companies or trade associations’.74 As a consequence, the
Codice di Autodisciplina is now adopted by almost all listed Italian
companies.75 This statement is subject to a ‘comply or explain’ regime,
as the company will have to give reasons for not adopting specific
71
Corporate Governance Committee, Corporate Governance Code, December 2011 (latest
version).
72
ABI, ANIA, Assonime, Confindustria and Assogestioni.
73
Article 123bis TUF, or Testo unico finanziaro. 74 Article 123bis(2) TUF.
75
In fact, 95 per cent of listed companies, but 13 have expressly stated that they do not to
adhere to the code, but give information about their own system of governance: see
Assonime, Noti e Studi, 2012, § 2.1. In addition, 31 other companies have announced
that they would not apply one or several of the code’s provisions, especially those dealing
with the independence of board members.
90 eddy wymeersch
provisions of the code. The practices followed by the company ‘over and
above’ the legal requirements also have to be stated.76
The provisions of the code are divided into principles and criteria
of application77 that are binding on the companies that profess to
respect the code on a ‘comply or explain’ basis. The code also provides
comments, consisting of ‘suggestions’ that can be disregarded without
explanation.78 Companies are expected to publish a corporate govern-
ance statement, stating how the principles and criteria have been
applied, or the reasons for not applying them. The committee declares
that it will ‘monitor’ the implementation of the code.79
Starting in 2001, Assonime (since 2004 with Emittenti Titoli) pub-
lishes a detailed report each year containing data and analysis regarding
the compliance by Italian listed companies with the Corporate
Governance Code.80 These annual reports provide an in-depth analysis
on their compliance with the code’s most significant recommendations,
but also discuss several recent regulatory issues, or take a position on
questions of interpretation. The Assonime reports offer a valuable source
of information and insight into the Italian governance system. These
reports do not publish names of, nor can Assonime engage with, com-
panies that have not implemented the code.
Despite the fact that the corporate governance statements and the remu-
neration reports are regulatory information,81 Consob, the market regula-
tor, is mainly involved in enacting regulatory statements, but apparently not
in the implementation of the code. Up to now the Corporate Governance
Committee has not functioned as an enforcer of the code either. Recently
the committee stated its intention to evaluate whether to reinforce the
‘comply or explain’ mechanism and its monitoring activity.
The code essentially pays ample attention to board issues, while data
about the ownership structure also receive much attention (Bianchi et al.
2011). The code has been amended several times: in 2011 to introduce
76
Article 123bis(2) TUF.
77
Criteri applicativi, explained as the recommended behaviour necessary for achieving the
objectives of the code’s principles.
78
They serve to illustrate the meaning of the principles and ‘criteri applicativi’ and some
ways for achieving the stated objectives.
79
The expression ‘monitor’ has been used in the introduction to the 2011 revision of the
code.
80
See the latest report, ‘Corporate Governance in Italy: Compliance with the CG Code and
Related Party Transactions (2011)’, available at www.assonime.it.
81
Article 113ter TUF, referring to the information viewed in Chapters I and II, Sections 1,
I-bis, and V-bis of the same Title.
the rules on remuneration, while its December 2011 version incorporates

different changes that have appeared over time and takes account of the
needs of smaller listed companies. The new code has been streamlined
and, at the same time, strengthened, with a view to increasing the
effectiveness of the recommendations, thereby taking into account the
most recent national and international best practices, notably with
respect to the central position of the board of directors, including of its
‘independent directors’, the role of board committees and the strength-
ening of the internal control system.
In particular, the code focuses on the role and composition of the
board and its internal committees, the independent directors, the remu-
neration regime and internal control and risk management. The content
and transmission of information to Consob takes place partly in accord-
ance with standardised tables, established in accordance with Consob’s
instructions.82 Pursuant to a law of 2011, Consob has published guide-
lines ensuring gender diversity in Italian listed companies.83 A 2011 law
has further limited the number of directorships an individual director
can exercise in financial institutions.84 The law provides that asset
management companies must cast the votes for their portfolio compan-
ies.85 Academic research has been undertaken on the role and influence
of specific groups of institutional investors (Bianchi and Enriques 2001;
Barucci and Cecacci 2005; Barucci and Falini 2005; Bianca 2008).
In an academic research study, the compliance issue has been investi-
gated by Bianchi et al. (2011), concluding that there is a lower degree of
compliance than officially stated. The latest Assonime and Emittenti Titoli
analysis shows that the quality and quantity of disclosure are generally good,
and that in recent years, transparency about the reasons for not adopting
the code has improved. However, business leaders have called for inde-
pendent monitoring, without specifying how this is to be achieved.
82
Consob, Regolamenti emittenti, Art. 100 – Composizione degli organi di amministra-
zione e controllo, direttore generale, pursuant L. 12 June 2011 nr. 120.
83
Article. 1 of Law number 120 of 12 July 2011 has revised the Consolidated Law on
Finance (Art 147 and 148b) requiring the introduction of statutory provisions that can
be reserved for the less-represented gender in the relevant bodies to a share of one-third
of the board of directors.
84
Law No. 214/2011, entitled ‘Protection of competition and personal cross-shareholdings
in credit and financial markets,’ bans executives from holding a board seat in more than
one financial institution operating in the same sector or market.
85
Article. subs. 2, TUF (Consolidated Law on Finance or Legislative Decree No. 58 of 24
February 1998; Consolidated Law on Finance pursuant to Articles 8 and 21 of Law no. 52
of 6 February 1996).
92 eddy wymeersch
7.7. Luxembourg
Luxembourg adopted its corporate governance recommendations in
2006,86 as part of the listing conditions imposed by the Luxembourg
Stock Exchange, the latter conditions being approved by ministerial
decree. The code was drawn up by a committee composed of members
of listed companies, financial intermediaries and representatives of the
exchange, with academic support from the Luxembourg law faculty. It
applies to Luxembourg listed companies, while special attention is paid
to companies with multiple listings that can freely follow stricter foreign
conditions. Companies are expected to publish a ‘Governance Charter’
on their website and a ‘Governance Statement’ in their annual report.
The code is quite an elaborate document, composed of principles that
must be applied, recommendations that call for a ‘comply or explain’
approach, and non-binding guidelines. The Stock Exchange ensures the
adoption of the recommendation in the framework of its external checks
on disclosures provided in the listing conditions, while substantive
follow-up is referred to as the task of the shareholders. Mandatory
disclosures fall under the monitoring of the financial supervisor, the
CSSF, or ‘Commission de surveillance du secteur financier’.
7.8. The Netherlands

Corporate governance and the implementation of the applicable provi-
sions, especially of the codes, has received ample attention in the
Netherlands (McCahery and Vermeulen 2009).
7.8.1 The corporate governance code

The Dutch corporate governance practice is based on elaborated legal
provisions, laid down in Book 2 of the Civil Code, and in a detailed
corporate governance code, which built on previous similar documents
known as the Tabaksblat Code, although previous documents have also
raised an early interest in the matter.87 The present code dates from 2008
and is based on a two-fold layer distinguishing between principles and
practice notes. The code has been designated by the ministry as the
86
The official name is Les dix Principes de gouvernance d’entreprise de la Bourse de
Luxembourg, 2nd rev. edn 2009.
87
This was the so-called Peeters Recommendations 1997, see www.ecgi.org/codes/docu
ments/nl-peters_report.pdf.
applicable code, as referred to in article 391(4) of the Companies Act,88

according to which it will decide on the ‘actual value and usability’ of the
code.89 The code is based on a ‘comply or explain’ approach. The self-
regulatory provisions of the code are not subject to enforcement by the
regulators, i.e. both the market regulator and the prudential regulator.
But both are confronted with governance rules laid down in the respect-
ive EU prudential or market directives.
Corporate governance practice is well documented in the elaborate
reports of the Monitoring Commission Corporate Governance, which
has published its third assessment in this field. This commission was set
up on 2 July 2009 by decision of the Cabinet90 and is composed of seven
members – business leaders, practising lawyers and academics, assisted
by a secretariat composed of persons originating from the two ministries
involved (finance and economy). The task of the Corporate Governance
Commission has been described as ‘drawing up an inventory, on an
annual basis, about how and to what extent the provisions of the code
have been implemented, and the identification of the gaps or impreci-
sions in the code; moreover the Commission will keep itself informed
about the international developments in the field of corporate gover-
nance and this from a perspective of the convergence of the codes’.
The commission proceeds to a detailed review of the corporate gov-
ernance statements published by the Dutch listed companies. This ana-
lysis gives rise to an elaborate number of findings, for example, with
respect to the overall compliance rate,91 the progress realised since the
previous report and specific comments whereby compliance is noted, in
addition to insufficient compliance or lack of compliance.
The commission has refined the notion of ‘comply or explain’, distin-
guishing what is considered to refer to strict application of the code’s
requirements from compliance, referring both to application of the provi-
sion, as well as non-application with reasoned explanation. The monitoring
reports identify the state of application, and mention the most significant
provisions where explanations are given. The selection of the provisions to
be commented upon change over time,92 depending on the commission’s
attention to a specific item.
88
Besluit, 20 maart 2009, Stb. 2009, 154. 89 ‘actualiteit en bruikbaarheid’
90
Besluit van 6 december 2004, gepubliceerd in Staatscourant nr. 241 van 14 december
2004.
91
See Monitoring Commission Corporate Governance, Report 2011, p. 24 e.s.
92
For the list see ibid., p. 19.
94 eddy wymeersch
The monitoring report identifies the cases of non-compliance, espe-

cially when no explanation is given, or when the explanation is insuffi-
cient. If the company states that the provision is not applied due to
reasons specific to the firm, or that the provision will be decided on a
case-by-case basis, this is considered non-compliance. The commission
also points to the danger of standardised explanations that companies
sometimes reproduce from each other,93 a bad habit, as the explanation
should be firm specific. Also frequent are the transitory derogations,
which are acceptable, although not for more than one year. Explanations
should be explicit, e.g., with respect to board evaluation, where the
method of evaluation and the outcomes must be described. In some
cases the commission proceeds to an interpretation of the code’s
provision.94
Beyond these surveys the commission has developed techniques to
incentivise companies to make progress in adopting the code, for exam-
ple, by holding application meetings95 with groups of directors, top
managers and shareholders, calling attention to specific subjects and
exchanging experiences. The report does not mention whether individ-
ual cases, especially shortcomings, are discussed in these meetings;
however, due to their composition, this seems rather unlikely. Another
type of meeting is those organised with the employers’ associations, the
main unions, the Association of Securities Issuers, the Association of
Investors and of Accountants, and the two supervisory authorities.
Engagement with the Parliament’s Finance Committee was announced
for 2012.
Under the heading of the AGM the commission has analysed the
position of the institutional investors. The Dutch corporate governance
code states that institutional investors should publish their voting poli-
cies, the implementation of these policies (on an annual basis) and how
they have voted (on a quarterly basis). Eumedion’s ‘Best practice 7’
requests its participants to use their voting rights in a well-considered
manner and in line with their voting policies. See the next section.
93
Ibid., p. 12 et seq.
94
E.g., is the payment of an exit premium in the case of a voluntary exit by a director a case
of where the exit payment is not acceptable according to the Commission?: Monitoring
Commission Corporate Governance, Report 2011, p. 11, or the holding period for a
director’s restricted shares, allowing for an exception for sales serving to financing the
upfront tax burden, ibid., p. 14?
95
‘Nalevingsbijeenkomsten’ or application meetings.
7.8.2 Other corporate governance recommendations

7.8.2.1. Institutional investors Other organisations have published
corporate governance recommendations addressing issues from their
specific point of view. Eumedion, which is the Dutch professional organ-
isation of the large institutional investors, mainly pension funds and
asset managers, has published ‘Best Practices for Engaged Share
Ownership’,96 which request members of Eumedion and other interested
shareholders to engage actively with the investee companies and report
on an ‘apply or explain’ basis.97 The Eumedion secretariat monitors
these management efforts on the basis of published annual reports of
the member organisations and reports back about it in a report to the
Eumedion board, parts of which are published.
As part of the guidance provided in the Best Practices document, a list
of instruments have been identified reflecting a set of escalating steps in
cases where differences of opinion have not been bridged between the
institutional investor and the management of the investee. The list
reproduced hereunder can be considered the standard set of instruments
for investors to ensure that boards take appropriate account of their
points of view.
Elements for a policy of this kind may include:
– writing a letter to the management and/or supervisory board in which
the matters of concern are explained;
– holding additional meetings with the management and/or supervisory
board, specifically to discuss matters of concern;
– holding meetings with other stakeholders, such as other shareholders,
banks, creditors, the works council and non-governmental organisa-
tions (NGOs);
– expressing concerns in a shareholders’ meeting;
– issuing a public statement;
– intervening jointly with other institutional investors on specific issues;
– requesting that certain subjects be placed on the agenda for the
shareholders’ meeting or asking that an extraordinary shareholders’
meeting be convened;
96
The Best Practices were adopted on 30 June 2011, available at www.eumedion.nl/en/
public/knowledgenetwork/best-practices/best_practices-engaged-share-ownership.pdf.
97
See § 1.3 of the Best Practices, nt. 8.
96 eddy wymeersch
– submitting one or more nominations for the appointment of a mem-

ber of the management board and/or supervisory director as
appropriate;
– taking legal action, when appropriate, such as initiating inquiry pro-
ceedings at the Enterprise Chamber of the Amsterdam Court of
Appeal;
– selling the shares.
Eumedion also publishes position papers about specific issues; often
recommendations are attached to these analyses. Several of these state-
ments deal with remuneration issues, dividend policy and the acquisition
of own shares98 and usually go beyond the requirements of the general
corporate governance code. Thus, the latest Principles on remunera-
tion99 stated that not only must the remuneration policy be approved by
the AGM, but that the remuneration report, as drawn up by the super-
visory board, and containing the individualised implementation of its
general remuneration policy, should be submitted to the AGM for a
‘vote’, whereby this probably should not be read as a formal approval. In
case the supervisory board would not do so, the shareholders could
express their views, and specifically their dissatisfaction, in the context
of other votes, especially as part of their decisions to grant discharge of
liability to the members of the supervisory board.
Some institutional investors have created separate investment funds,
where securities are located that form the basis for a more active
approach to engagement and which can then be held for the longer
term. Depending on their organisation, these separate funds may also
help to resolve the conflicted situation, for example, related to receiving
price-sensitive information.
7.8.2.2. Public investors The Association of Securities Investors

(Vereniging van Effectenbezitters, or VEB) is an 86-year-old association
regrouping individual and corporate investors through investment clubs
(352) and engaging in the defence of their rights. The VEB attends about
150 general meetings yearly, and engages in activist positions, the most
visible part of which is the numerous lawsuits for mismanagement,
98
Recommendation 2009, ‘Aanbevelingen inzake de machtiging tot inkoop van eigen
aandelen en inzake de verantwoording over het dividendbeleid’, available at www.
eumedion.nl.
99
‘uitgangspunten’, or assumptions.
market manipulation and the publication of misleading information.

These highly visible suits have led to several of the leading decisions in
the fields of investor protection and of corporate governance, as will be
explained further. The VEB has a direct communication investors serv-
ice with helpline, publishes regular information on investment funds,
along with a rating of these funds, and has engaged in public action
against sales fees. It has published a takeover code, containing some
high-level principles100 and corporate governance ratings, but the latter
practice seems to have been discontinued.
7.8.2.3. Banks As a consequence of the banking crisis, in 2010 a

voluntary code was drawn up by the Dutch Bankers’ Association
(NVB) as the follow-up to the elaborate report of the Advisory
Committee on the Future of Banks.101 The code has been subscribed
by the Netherlands licensed banks and aims at restoring public con-
fidence in the banks after the crisis.102 It contains several provisions
about corporate governance, especially on composition and function of
the supervisory board and the management board of banks, stressing the
importance of risk management and the ethical conduct of banks. It
requires banks to be ‘managed carefully considering the interests of all of
the parties involved in the bank, such as the bank’s clients, its share-
holders and its employees’.
The code is based on a ‘comply or explain’ approach and is followed
up by an independent Monitoring Commission Banks, appointed by the
NVB, in consultation with the Ministry of Finance. The first report of the
Monitoring Commission identifies the progress in adopting the princi-
ples made, especially by the largest banks. The report is essentially
descriptive and does not identify issues relevant to a specific bank. It
openly mentions doubt about the self-regulatory approach, but consid-
ers that a conversion into hard law would be ‘inopportune and
premature’.103
100
VEB Annual Report 2010, p. 15.
101
See Adviescommissie Toekomst Banken: ‘Naar Herstel van Vertrouwen’, 2009, available
at www.nvb.nl/publicaties/090407-web_rapport-adviescommissie_toekomst_banken_def.
pdfen. www.nvb.nl/index.php?p=290335. The code is dated 9 September 2010.
102
It was mentioned that this initiative was adopted to avoid more intrusive government
regulation.
103
Monitoring Commission Code Banks, Rapportage Implementatie Code Banken,
December 2011, p. 8, available at: www.nvb.nl/code-banken/rapportage_implementa-
tie_codebanken_dec2011.pdf.
98 eddy wymeersch
7.8.2.4. Other governance codes There are several other fields where
voluntary governance codes have been developed, largely inspired by the
ideas and principles in the main code for listed companies.104
7.8.3. Dutch case law

Several significant decisions have been rendered by Dutch courts,
including the Supreme Court, referring to corporate governance issues.
Most of the cases were first brought before the Enterprise Chamber
(Ondernemingskamer), a specialised chamber of the Amsterdam
Court of Appeal, that has been charged by the Companies Act to
deal with disputes between boards, shareholders and company
employees. The Chamber has the right to determine enquiries in
company matters and adopts the measures that are necessary to
remedy the conclusions from these findings – essentially whether
mismanagement has occured. The powers of the Chamber are consid-
erable and include the power to annul or suspend the decisions of any
corporate body to replace, suspend, or even dismiss members of the
supervisory board or the board of directors; to temporarily change the
articles of association; to transfer shares to an administrator; and, if
necessary, even to wind up the company.105 Several decisions of the
Enterprise Chamber have dealt with corporate governance issues, the
most recent referring to the principles of governance or to the code
itself. These decisions are often quite complex; therefore, the following
summary is limited to the statements of the courts that relate directly
to corporate governance.
Several decisions relating to the relationship between the board –
usually the supervisory board, common to the large Dutch listed
companies – and the shareholders, usually contesting decisions of the
board that were considered contrary to their interests, or for which
they deemed not to have been adequately informed, or for not having
submitted their proposal for approval to the AGM. Most of the time the
Enterprise Chamber held in favour of these claimants, but the decision
was frequently overturned by the Supreme Court. In one of its leading
decisions, the Court held that neither in company law, nor the changes to
the company law then under consideration, nor in the generally
104
See, e.g., for the hospital sector: Zorgbrede governance code, available at www.branche
organisatieszorg.nl/governancecode_.
105
Article 2:356 Civil Code.
accepted views on corporate governance,106 could support be found for

holding that the board must submit a proposal to the AGM for a private
bid on a substantial part of the company’s activities.107
In the ABNAmro case concerning the sale of US-based LaSalle Bank,
shareholders contested the sale of the bank by the sole decision of the
board, and not of the AGM. The Supreme Court – reversing the
Enterprise Chamber’s decision – decided that sufficient consultation
had taken place and that the interests of all parties concerned had been
sufficiently taken into account, in conformity with the preamble to the
Tabaksblat Code. According to the Court, the latter expresses the com-
monly accepted view in Dutch law, and corresponds to the core company
law notions of ‘reasonableness and justice’ applicable to all company
decision making (art. 2.9 Civ. code), while the decision further met the
criteria for the correct implementation of their duties by each director.108
Neither on the basis of the law, nor of the articles of incorporation,
does the AGM have an approval right or is the board obliged to consult
the shareholders. According to the Supreme Court, the commonly
accepted legal opinion as expressed in the corporate governance code
does not lead to the conclusion that the board of a company is obliged to
submit for shareholder approval or consultation a decision that is within
the competency of the board on the mere ground that shareholders have
an interest in selling their shares at the highest price.109 In another
passage, the Court held that the code’s provision on the relationship
between the Supervisory Board and the shareholders does not imply that
the board should justify its decisions on the matter at hand.110
In the Versatel case, decisions of a court-appointed provisional admin-
istrator were contested; the Supreme Court held that the powers of the
administrator flow from the provisions of the law on the powers of the
directors and from the provisions of the corporate governance code.111
The ASMI112 decision concerned the action of activist investors who
wanted the company to change its strategy and spin off some of its
106
The decision refers to the 1997 statement of the Peeters Commission, www.commis
siecorporategovernance.nl/Commissie%20Peters, on which the company had com-
mented in its annual report: ‘Aanbevelingen voor goed bestuur, adequaat toezicht en
het afleggen van verantwoording’ in the report, ‘Corporate Governance in Nederland’
1997.
107
Hoge Raad (HR), 21 February 2003, Hollandsche Beton Group, decision AF1486,
§ 6.4.2.
108
HR, 9 July 2010, ABNAmro decision 09/04465 and 09/04512, § 4.4.2.
109
HR, July 2010, ABNAmro decision 09/04465 and 09/04512. 110 Ibid., § 4.5.1.
111
HR, 14 September 2007, Versatel § 4.3. 112 HR, 9 September 2010, ‘Asmi’.
100 eddy wymeersch
activities. These shareholders had not succeeded in gaining direct influence,

as the company had adopted the so-called ‘continuity’ model whereby
preference shares were held by a foundation and the directors were
appointed on the binding proposition of the Supervisory Board,
a proposition that could only be set aside by a two-thirds majority.
The Enterprise Chamber considered that this governance structure was
outdated and defensive and prevented shareholders from exercising
any influence on its strategic decisions in the company due to its ‘closed
position’. The Chamber ordered an inquiry on the protective construction.
On appeal, the Court annulled the decision of the Enterprise
Chamber. It held that the Chamber’s analysis did not correspond to
Dutch law, where the board is required to serve ‘the interests of the
company and of the enterprise it runs’,113 and involves the interest of all
stakeholders, inter alia, the shareholders. This view conforms to the
provisions of the Tabaksblat Code, whish ‘expresses the commonly
accepted legal opinion in the Netherlands and reflects the above
mentioned legal concepts of “reasonableness and justice”’. It is up to
the board to decide on the company’s strategy and on how far it involves
consultations with the shareholders.114 The arguments of the opposing
shareholders were held to have been rejected on valid reasons. The
Court did not find that certain provisions of the Code had not been
met; instead, it took account of the fact that the company had promised
to live up to the governance code in the future.
In the case of Begeman,115 a small company heading for market exit,
the company had stated in its annual report that it would probably not
meet the code’s requirements – among others, its provisions on conflicts
of interest – and therefore had not published a governance statement in
its annual report. Nevertheless, the Enterprise Chamber held that it
should have applied the code’s provisions on conflicted directors, raising
the double question about voluntary adherence to the code and its
effects, even in the absence of adherence. The court deemed that the
rules on conflicts of interest should nevertheless apply.
This short overview can be summarised as follows: corporate govern-
ance is often referred to in Dutch case law and is held to express some of
the basic principles of Dutch company law. Violations of the codes as the
basis for legal action have been attempted, but the attempts have not
been successful before the Supreme Court, not because the code was
113
Article 2: Burgerlijk Wetboek. 114 HR, 9 September 2010, Asmi § 4.4.1.
115
Enterprise Chamber, 28 December 2006, Begeman, § 3.7.
rejected as a legal instrument – several times it was held to express the

concepts underlying Dutch company law – but because the factual
situation did reveal a violation of Dutch law, and that was sufficient for
the Court. Logically the code’s provisions are sometimes cited in support
of the analysis of Dutch company law in general. Although a comparison
is difficult to make, one could put the code at the same level as case law,
not legally binding, but a useful – and authoritative116 – source of
information on the substance of the law.
7.9. Portugal
In Portugal, the Corporate Governance Recommendations are adopted
and implemented by the securities regulator, the CMVM.117 As they
relate closely to some legal provisions, a consolidated document has been
published indicating item by item the legal requirements and the addi-
tional recommendations of the CMVM.118 Although based on the advice
of representatives of the business community, the code is essentially a
document generated and monitored by the CMVM. The ‘comply or
explain’ basis does not prevent it from being a statement of a public
authority: the obligation to state the applicable regime and the relatively
high level of generality of the code point in the same direction. The
adoption of the code119 – which purportedly contains recommendations,
not formal legal obligations120– was made mandatory in 2001, obliging
companies, on a comply or explain basis, to express themselves on the
state of compliance and the means put at work. The most recent require-
ment to date is formulated in the CMVM regulation 1–2010, requiring
companies to implement the code or, in specific circumstances, a similar
116
But this aspect does not appear from the decisions.
117
The last update dates from 2010: CMVM Corporate Governance Code 2010
(Recommendations). Commercial organisations publish in-depth investigations with
critical comments, but always on a no-name basis.
For the text, see www.cmvm.pt/EN/Recomendacao/Recomendacoes/Documents/
2010consol.Corporate%20Governance%20Recommendations.2010.bbmm.pdfwww.
cmvm.pt/EN/.
118
Consolidation of the Legal Framework and Corporate Governance Code, www.cmvm.
pt/EN/Recomendacao/Recomendacoes/Documents/20122010.Cons.MM.BB.Cons%20
Fontes%20Norm%20%20e%20CGS%202010%20trad%20inglês.pdf.
119
Issuers may comply with a different corporate governance code instead. However, since
there is no other Portuguese corporate governance code, the CMVM Code has been the
only one adopted.
120
The preamble refers to ‘recommendations’ but most of the code’s obligations are
formulated in the ‘shall’ mode.
102 eddy wymeersch
code, to state which recommendations have and have not been adopted and,
if it is the case, to explain the reasons for the non-adoption of some of the
recommendations. The model and the data to be included in the govern-
ance report are detailed in the elaborate annex to the regulation, from which
companies can depart on the condition that they state their reasons and
publish other relevant remarks. The CMVM verifies the effective compli-
ance with the recommendations and the quality of the explanations given. If
the company fails to report compliance with the code and/or to explain the
reasons for not complying with some of the recommendations, the CMVM
has the power to apply administrative sanctions to the company. In at least
one case it has also imposed a fine in this context.
The verification process is divided into two parts: the first consists of a
check of completeness of the disclosures in accordance with the legal
requirements, leading to corrections or completion of information. At
this stage, the CMVM verifies if every listed companies has (i) adopted
one (the) corporate governance code, (ii) stated its compliance or non-
compliance with each recommendation thereto, and (iii) explained the
reason for not complying with some recommendations. In the second
stage, a more in-depth analysis is undertaken and discussed with the
companies during a hearing about differences between the CMVM’s and
the company’s reading. Further individual meetings with the represen-
tatives of the company concerned may take place in order to ‘convince’
them about adherence to the recommendations. Compliance is finally
assessed at the end of this process, aiming at urging companies to
provide adequate and coherent explanations for not following the
recommendation.
The CMVM reports in detail on the outcomes of this process, holding
a press conference where information is given about the most and the
least compliant companies, also highlighting the better explanations.
The most significant enforcement instrument, however, is the
CMVM’s annual report which publishes detailed nominative statements
about the state of compliance.
The CMVM has closely studied the corporate governance publications
and practices of the 49 Portuguese listed companies for several years.
The last detailed overview of its corporate governance analysis deals with
the statements for 2009,121 illustrating the considerable efforts in terms
of staff and time that are invested in this matter. The overview publishes
121
Relatório anual sobre o governo das sociedades cotadas em Portugal 2009, available at
www.cmvm.pt/CMVM/Estudos/Pages/20110519a.aspx.
its assessments in a nominative way, indicating for each of the listed

companies whether and to what extent the recommendations have been
met. Rankings and score lists are established indicating the assessment
gap, or the differences between the – usually more positive – assessment
of each item by the company,122 and the separate assessment by the
CMVM, distinguishing between the ‘essential recommendations’ and
the ‘other recommendations’ and adding a column for a ‘synthetic
indicator’ for total compliance. Aspects considered essential, such as
the functioning of the general meeting, the position of the sharehold-
ers123 and the recommendations on the boards are further explored,
while several other, more sensitive recommendations, such as the rec-
ommendation on remuneration, are individually analysed. All of this
information is given per company, indicating in suggestive colours
which company has been deficient in a specific class of recommenda-
tions. The table relating to ‘remunerations’ particularly shows much
more ‘black’ than the tables for the other recommendations, indicating
a poor degree of implementation.
On the assessment of the ‘comply or explain’ practice, the report
identifies the main items of concern: anti-takeover protections,124 pro-
visions on the designation, evaluation and dismissal of the auditor,
remuneration rules and the provisions relating to alignment of
shareholders’ and directors’ interests. All of these are fields in which,
according to the report, the compliance rate is significantly lower than
the average compliance. Although the report extensively uses the ‘name
and shame’ instrument by identifying non-compliant companies, the
judgment is sometimes a balanced one: e.g. in the case of Portugal
Telecom, where the company had refused to abandon its anti-takeover
defences, the draftsperson of the report stated that he saw no arguments
for condemning the company’s position. Non-compliance leads to indi-
vidual meetings with the companies concerned in order to ‘convince’
them about adherence to the recommendations.
The Portuguese regime is specific in the sense that it comes close to a full
regulatory regime, although still based on a ‘comply or explain’ technique. It
is unclear to what extent the CMVM imposes its views and whether the
122
Often due to a more optimistic reading of the Recommendation, according to the
CMVM.
123
E.g., the recommendation in favour of ‘one share, one vote’.
124
As the Recommendations contain a statement in favour of ‘one share, one vote’,
remarks are addressed to limits on voting rights, protective charter provisions, or
quorum requirements.
104 eddy wymeersch
‘explain’ view, provided it is adequately motivated, prevails. Since its latest

2009 report, the CMVM has stepped up its monitoring efforts, proceeding
to a more thorough analysis of the level and quality of explanations.
7.10. Spain
On the basis of the Securities and Markets law of 1988,125 a decree126 has
delegated to the CNMV, the Spanish securities regulator, the power to
define the content and form of the annual report, including the corporate
governance statements. The decree contains a fairly elaborate list of
items to be included in the CNMV’s implementation document, the
‘Unified Code’.127 On the basis of this decree, it is responsible for
drawing up not only the code, starting from the two previous codes,128
but also for exercising surveillance on its application.129 In July 2005, a
Special Working Group was designated to assist the CNMV in drafting
the code, in close consultation with the private industry and the
Ministries of the Economy and of Justice and the Central Bank. The
Unified Code was adopted by the CNMV on 19 May 2006, and adapted
to include remuneration provisions in 2009. In its annual reports for
2009 and 2010, the CNMV explains in detail the action it has developed
to ensure effectiveness of the Unified Code, and other provisions affect-
ing company life. The code is based on a ‘comply or explain’ method,
against the background of the applicable legal provisions, among them
the accounting rules that call for close scrutiny from the CNMV.
According to the code:
It will be left to shareholders, investors and the markets in general to

evaluate the explanations companies give of their degree of compliance
with Code recommendations.
125
Article 116 of the Ley 24/1988, de 28 de julio, del Mercado de Valores, modified by Ley
26/2003, de 17 de julio.
126
Ordinance ECO/3722/2003, of 26 December 2003, ‘sobre el informe anual de gobierno
corporativo y otros instrumentos de información de las sociedades anónimas cotizadas
y otras entidades’, noticias.juridicas.com/base_datos/Privado/o3722–2003-eco.htm.
127
Special Working Group, Unified Code on Good Corporate Governance, January 2006.
128
‘La Comisión Nacional del Mercado de Valores queda habilitada para dictar las
disposiciones necesarias para desarrollar, en el ejercicio de las competencias que le
son propias, lo dispuesto en la presente Orden’.
129
Se faculta a la Comisión Nacional del Mercado de Valores para determinar las especi-
ficaciones técnicas y jurídicas, y la información que las sociedades anónimas cotizadas
han de incluir en la página web, con arreglo a lo establecido en el presente apartado
Cuarto de esta Orden.
In other words, the extent of compliance or the quality of explanations

will not give rise to any action by the CNMV, as this would directly
invalidate the voluntary nature of the code. This affirmation is under-
stood to be without prejudice to the monitoring powers assigned to the
CNMV with regard to the Annual Corporate Governance Report of
listed companies in article 116 of the Securities Market Law and related
Order.130 This statement underlines the double nature of the Spanish
code: it is not a purely self-regulatory document, but has been elaborated
by the CNMV and is applicable on the basis of a legal provision.
Moreover, the CNMV closely monitors the way the code is applied, as
will be illustrated below.
The CNMV has developed an elaborate practice relating to imple-
mentation of the corporate governance rules, as laid down in the code.
This action is based on its general competence to verify the disclosures in
the annual reports made by listed companies. It goes along with an
equally important action in the field of accounting by listed companies.
Both are commented upon in some detail in the CNMV’s annual reports.
In its first annual report since the entry into force of the new regime
(2009),131 the CNMV noted in 53.2 per cent of the cases factual non-
compliance, a mere mention of the existence of a deviation without
explaining the reason, or general disagreement with the recommenda-
tion. On the basis of this finding, the Commission has sent out deficiency
letters that have led to changes in the disclosures, to rectifications and
expansion of previous disclosures. Special attention was drawn to the
qualification of independent directors, on which a table of possible
grounds for lack of independence is established, and to related party
transactions.
The annual report for 2010 shows a significant increase in terms of
reporting of supervisory action. In the context of its oversight on finan-
cial reporting, this report calls for special attention to issues of internal
controls and risk-management systems, as mandated by a change in the
law.132 Already in previous years, the CNMV strongly focused on
accounting issues, especially on the basis of auditors’ reports. The over-
view gives a detailed analysis of the number of cases where qualified
130
Whereby the CNMV may order companies to make good any omissions of false or
misleading data.
131
CNMV Informe Annual de Gobierno Corporativo de las companies del IBEX 35, 2009,
available at www.cnmv.es/DocPortal/Publicaciones/Informes/IAGC_IBEX_09.pdf.
132
Implementation of the Law of 4 March 2011, L1/2011.
106 eddy wymeersch
reports were delivered and the types of shortcomings that were identified
by the auditors. It also proceeds to a ‘substantive review’ of a number of
accounts of companies, selected on a risk and random basis and resulting
in ‘deficiency letters’ asking for additional information on accounting
policies and information breakdowns.133 Worth mentioning is the sig-
nificant number of letters relating to accounting policies, specifically on
valuation, related party transactions, impairments, etc. On that basis, the
CNMV can require additional information, reconciliations, corrections
and, in material cases, entire restatements.
In the corporate governance field, apart from statistical information
on general compliance with the different provisions of the Unified
Code,134 the CNMV’s report gives statistics on board composition,
remuneration, general meetings, etc.135 With respect to the ‘comply or
explain’ principle, the CNMV investigates cases where compliance was
deficient, too generic or redundant, requesting further information or
clarifications. These may result in further information, amendments,
new information or additional explanations in ‘explain’ cases. Some of
this information is included in the centralised company information
database, organised by the CNMV.
Special action relates to independence criteria as laid down in the Unified
Code: in case of doubt, ‘deficiency’ letters are sent for clarification or
modifications. The type of violations are reported on. For example. in
2010 the number of cases where independent directors had ‘significant
business relationships’ was quite substantial, but as far as one can derive
from public documents, this did not lead to any additional information, nor
to corrective action.
The report contains detailed tables on related party transactions,136
stating the amounts involved and the variations vis-à-vis a previous
period. Here again, the role of the CNMV is to ensure transparency.
133
See CNMV, Annual Report 2009, Deficiency Letter on Independent Directors,
p. 140.
134
For 2009, the 2010 Annual Report found full compliance in 77 per cent of the cases with
the recommendations and 10 per cent partial compliance, especially on remuneration.
135
Detailed information can be found in a separate publication, published annually:
Informe de Gobierno Corporativo de las entidades emisoras de valores admitidos a
negociación en mercado secundarios oficiales, last issue 2012, available at www.cnmv.
es/portal/Publicaciones/PublicacionesGN.aspx?id=21.
136
Subdivided in transactions with significant shareholders, persons or companies belong-
ing to the group, directors and executives and other related parties. The table also
includes regular flows, e.g. due to provisions of goods or services, dividends, licence
agreements and similar items.
There is no indication that these disclosures have led to action by the

CNMV such as criticising certain intra-company transactions or trans-
fers to shareholders.
7.11. Sweden
The Swedish Corporate Governance Code was first adopted in 2005, and
updated in 2010.137 The code is drafted by the Swedish Corporate
Governance Board, which is part of the ‘Association of Generally
Accepted Principles in the Securities Market’, a body composed of
members of the Swedish private corporate sector organised among the
ten leading business associations. The Association itself is composed of
three self-regulatory bodies: the Swedish Securities Council, created in
2005, in charge of overseeing self regulation in the securities market and
formulating ‘Good practice in the Swedish securities market’, the
Swedish Financial Reporting Board, and the Corporate Governance
Board. Respecting these good practices is part of the listing agreement.138
The Council – and not the Board – gives opinions on issues of inter-
pretation of the code.
The code is a fully self-regulatory body of rules, applicable to all
companies that have their shares or depositary receipts listed on one of
the Swedish regulated markets.139 It applies in addition to the
Companies Act, containing an increasing number of formerly code
provisions (Unger 2006). The Swedish Annual Accounts Act requires
companies with their shares, warrants or bonds listed on a regulated
market to publish a corporate governance report.140 Except for a limited
number of provisions in the Annual Accounts Act, the content require-
ments are laid down in the code.
The requirement to adopt the code is laid down in the stock exchange
rules, and the exchange verifies whether the code is applied (Von
Haartman 2010).141 It could take disciplinary action in the case of a
137
The latest version to date: ‘The Swedish Corporate Governance Code’, 2010, available at
www.corporategovernanceboard.se.
138
See Nasdaq OMX Stockholm AB’s and Nordic Growth Market NGM AB’s respective
rulebook for issuers.
139
Nasdaq OMX Stockholm and NGM Equity.
140
Annual Accounts Act 1995:1554, chapter 6, ss. 6–9 and chapter 7 s. 31.
141
Fifty per cent of the companies follow the code without variation; another 40 per cent
with one explanation, the remainder with more than one.
108 eddy wymeersch
serious breach which the company was unwilling to correct,142 but

usually a dialogue will suffice. In principle, the board states that it is up
to the markets and the investors to judge the quality of the information,
while the exchange verifies whether the information is such that readers
understand the reasons for non-compliance and what alternative solu-
tions have been put forward.
The Board publishes an annual report, in which it states its general
approach to governance, expressing its firm belief in the Swedish com-
pany model with dominant shareholders. It publishes detailed data
about the way the code has been applied in the most recent year,
comparing it with data from the previous years. The code being based
on ‘comply or explain’, the board refers to the possibility of explaining
about alternative solutions. Alternatives or derogations are often justi-
fied with a reference to the Swedish company model. The analysis is
statistical and does not mention individual companies but, instead,
contains relevant data, for example, on the frequency with which audi-
tors review the corporate governance statements. Every second year, the
board’s annual reports also contain a ‘corporate governance barometer’
that reflects opinions on corporate governance issues on the basis of a
survey of the general public and of company leaders and investors,
sometimes illustrating specific trends in their perception of corporate
governance practices or proposals.
The board’s annual report also publishes opinions on a wide range of
company policy issues, e.g. dealing with the recent EU Commission
proposals, on auditor appointment (by the state), rotation, independent
directors, etc. According to the relevant statement, the board confirmed
that it had given a negative submission to the Commission’s consultation
on corporate governance.
7.12. Switzerland
Apart from the fairly elaborate provisions of the Companies Act, the
Swiss corporate governance rules are based on two sets of provisions: the
self-regulatory code elaborated by Economiesuisse, the Swiss Federation
of Business Associations143 and the instructions of the Swiss stock
142
One case has been reported.
143
Among these the Bankers’ Association, the Institute of Certified Accountants and Tax
Consultants, the Insurance Association, the chemical industry.
exchange,144 acting as a delegated body for the implementation of the

listing and disclosure rules.145 The corporate governance code, or
‘recommendations’, was originally drawn up in 2002 by an expert com-
mittee, and updated in 2007, with respect to remuneration matters,
which is still the most elaborate section of the code.146 The supporting
associations are reported to have a wide freedom to emphasise specific
aspects or depart from the code where necessary. At the level of these
associations, the code does not refer to the ‘comply or explain’ approach.
The main instrument of the Exchange for regulating the issuers’
activity are the Listing Rules,147 which form the basis of the different
disclosure obligations to which listed companies are subject. The
exchange, especially its listing department called ‘Swiss Exchange
Regulation’ , acts as an ‘independent regulatory body’, granting admis-
sion to the exchange and monitoring the implementation of the Rules
within the framework of its duty to organise the market. The information
is produced under the issuer’s responsibility, and technically submitted
to the exchange by the sponsoring securities dealer.148 The said depart-
ment is assisted by two expert panels, one dealing with the developments
in the fields of company reporting, the other more particularly with the
application of IFRS (the Financial Reporting Expert Advisory Panel and
the Specialist Pool for IFRS Issues).149
The listing Rules (art. 49), contain a reference to the Directive on
Corporate Governance,150 a statement drawn up by the Exchange’s
144
This analysis relates to SIX, the Zurich exchange; there is also an exchange in Berne,
specialising in SMEs.
145
Based on Art. 8 of the Federal Act on Stock Exchanges and Securities Trading (SESTA),
the Regulatory Board shall decide on the admission of equity securities to trading in the
SIX Swiss Exchange-Sponsored Segment and shall supervise compliance with the
requirements of these rules during the process of admission to trading.
146
The latest version to date: “Swiss Code of Best Practice for Corporate Governance”,
2007, available at www.economiesuisse.ch.
147
www.six-exchange-regulation.com/admission_manual/03_01-LR_en.pdf, adopted on
the basis of Art. 8 of SESTA. Other bodies of rules also play a role in corporate
governance matters, such as the Directive on Ad hoc Publicity and the Directive on
Disclosure of Management Transactions.
148
See Rules for the Admission of Equity Securities to Trading in the SIX Exchange-Sponsored
Segment, www.six-exchange-regulation.com/admission_manual/05_01-RSS_en.pdf; art. 24
stated that:’ ‘These Rules were approved by FINMA, the Federal Financial Market
Supervisory Authority on 23 April 2009 and enter into force on 1 July 2009.’
149
www.six-swiss-exchange.com/media_releases/online/media_release_201012081530_en.pdf.
150
Directive on Information relating to Corporate Governance (Directive on Corporate
Governance, DCG), 29 Oct. 2008, www.six-exchange-regulation.com/admission_ma-
nual/06_15-DCG/en/index.html.
110 eddy wymeersch
listing department. In its annex the directive contains a list of disclosure

items that must be included in the annual report, and this on a ‘comply
or explain’ basis, requiring companies to disclose their business practices
or to explain where they depart from the code. In the case of non-
compliance, information will have to contain ‘an individual, substanti-
ated justification for each instance of such non-disclosure’. Particular
attention is paid to remuneration issues, and to the presentation of the
structure of ownership. Strikingly, the directive contains no explicit
reference to the Recommendations of Economiesuisse151 as it deals
only with the disclosure requirements, while the substance is left to the
latter’s code of conduct on the basis of a division of tasks agreed in 2002
(Kunz 2010).
As part of its overall duty to ensure adequate information to the market,
the listing department of the Swiss Stock Exchange plays a significant role in
ensuring compliance of the disclosure provisions. It examines annual
reports, mostly by random-sampling, and comments on a number of
specific items.152 Following this review, each issuer will receive a comment
letter, unless a preliminary investigation is required. The exchange regularly
publishes statements about the investigations and about breaches of the
rules, mentioning the name of the company involved and the nature of the
violation. It may address reprimands to the company and also impose fines
for breaches.153 The sanctions are imposed by an internal Sanction
Commission, deciding on a proposal from the Listing and Enforcement
department.154 The financial supervisor, FINMA, is obviously not involved
in securing compliance with the listing requirements or the directive, except
in cases of price manipulation.
Shareholder associations are also active, for example, in ESG.155
151
This and the initial report are only mentioned among the other sources of information.
152
www.six-exchange-regulation.com/admission_manual/09_04_03-SER201103_en.pdf.
153
Article 61, 1 of the listing rules provides for the following instruments: the reprimand: a
fine of up to CHF 1m (for negligence) or CHF 10m (if deliberate); suspension of
trading: delisting or reallocation to a different regulatory standard: exclusion from
further listings: withdrawal of recognition. For applications, see Bergbahnen
Engelberg-Trübsee-Titlis AG, where a fine was imposed for not informing the exchange
about management transactions; or, SIX Swiss Exchange fines Altin Ltd, a fine for
breaching the ad hoc disclosure obligations of the DCG (26 January 2012; fine of CHF
100,000); investigation against Dufry Ltd of 19 January 2012, for not disclosing
management transactions.
154
See art. 9, ‘Sesta’ Federal Act on Stock Exchanges and Securities Trading, 24 March
1995; see for further details, Luechinger, S., Updates on Issuer Regulation, SIX
Exchange Regulation, 24 November 2011.
155
ACTARES, Aktionärinnen für nachhaltiges Wirtschaften, Schweiz.
7.13. UK
The UK Governance Code in its version of June 2010 is the successor of
several other leading self-regulatory instruments156 that have shaped
governance in the UK and in many other countries. The code is widely
followed by listed companies of all sizes. It is applicable to the companies
with a UK Premium listing157 of equity shares, whether they have been
incorporated in the UK or abroad. The rules of the UK Listing Authority,
part of the Financial Services Authority (FSA), now Financial Conduct
Authority (FCA), requires the application of the code, in addition to
several other more detailed disclosure provisions.158
In July 2010, in light of the diminishing position of UK institutional
investors as shareholders in UK companies,159 a ‘Stewardship Code’ was
adopted. The code states that it ‘aims to enhance the quality of engage-
ment between institutional investors and companies to help improve
long-term returns to shareholders and the efficient exercise of govern-
ance responsibilities by setting out good practice on engagement with
investee companies’. This code is principally applicable to the managers
of assets for institutional investors and more generally to all institutional
investors,160 whether UK domiciled or not. These parties should indicate
whether they have subscribed to this code and the FRC will subsequently
publish the list of the subscribers (Cronin and Mellor 2011).161
Both codes – Corporate Governance and Stewardship – are supported
by the FSA and based on a ‘comply or explain’ approach. Although
covering different fields, there is likely to be substantive interaction.
The Corporate Governance Code contains five main principles that
are mandatory,162 and forty-eight more detailed provisions that are
based on ‘comply or explain’. The overall implementation of the
156
The Combined Code was the immediate predecessor; the original 1992 Cadbury Code
stood as a model for most of the European governance codes.
157
This is the superequivalent regime under the listing rules, whereby conditions above the
EU Listing Directive (or the standard regime) apply.
158
See fsahandbook.info/FSA/html/handbook/LR/9/8, § LR 9.8.6, sub. 5.
159
See p. 9 of the FRC Developments on Corporate Governance 2011 announcing a study
identifying ownership of UK companies’ shares. On their holding in overseas compa-
nies, but then on a ‘best efforts basis’: see introduction to the code.
160
At the end of 2011, there were 234 signatories, among which were 175 asset managers,
48 asset owners (mainly pension schemes and investment trusts, of which many (31)
defined benefit schemes) and 12 service providers: FRC Developments on Corporate
Governance 2011, p. 20.
161
See also Financial Reporting Council, The UK Stewardship Code, July 2010.
162
Leadership, effectiveness, accountability, remuneration, relations with shareholders.
112 eddy wymeersch
Governance Code is monitored by the FRC, which publishes overviews

on the state of application, and the progress made, also giving indications
on objectives and policies. Specific topics like gender diversity, the use of
voting agencies and director rotation received special attention in the
2010 FRC overview. Full compliance is reported to have reached 50 per
cent for the FTSE 350 companies, while 90 per cent comply with all
but one or two of the provisions. Smaller companies also comply at
similar rates.163 The FRC also publishes guidance, for example on Board
Effectiveness, or on Audit Committees.164 A specialised panel within the
FRC, the Financial Reporting Review Panel (FRRP), on complaint,
screens the reporting by individual companies and publishes its decision,
but not its analysis.165 A newly created Financial Reporting Lab will
provide the opportunity for companies and investors to confront their
ideas and develop new reporting formats.
Commercial organisations publish in-depth investigations with crit-
ical comments, but always on a no-name basis.166
Arcot and Bruno (2006) investigated corporate governance disclo-
sures for the period 1998–2004 and found that on the one hand compa-
nies did not make very frequent use of explanations, and rather, were
inclined to box ticking, the relatively frequent absence of explanations
(17 per cent) being considered a signal of the disregard by shareholders
of the corporate governance matters. They also remarked that the quality
of the explanations was generally weak, and often remained the same
over the years. However, companies that did not publish explanations
were the ones that most considerably improved, once they decided to
comply.
The Stewardship Code, although also voluntary, is a different instru-
ment, being addressed to a different audience, with different obligations.
Up to now, the FRC has mainly launched a campaign for moving parties
163
See the Figures, p.11, FRC 2011, drawn from a Grant Thornton Study, www.grant-
thornton.co.uk/pdf/Corporate_Governance_Review_2011.pdf and Manifest Total
Remuneration Survey 2011.
164
March 2011.
165
Statement by the Financial Reporting Review Panel in respect of the report and accounts
of Rio Tinto Plc, 15 March 2011, where it was analysed whether the annual report
contained ‘a fair review of the company’s business and that the review required is a
balanced and comprehensive analysis of the development and performance of the
company’s business’.
166
A Grant Thornton Study (www.grant-thornton.co.uk/pdf/Corporate_Governance_
Review_2011.pdf); see also Heidrick and Struggles, European Corporate Governance
Report 2011, Challenging Board Performance.
to sign the code, the FSA having declared it mandatory in the sense that
‘managers should be required to disclose their commitment to the
Stewardship Code’.167 Signatories are expected to disclose, apart from
their commitment to the code, the way they have applied its principles,
and otherwise explain how they have taken the code’s obligations into
account. In its first report on the matter, the FRC identified four areas
where disclosure should be improved: conflicts of interest, strategy for
collective action, proxy voting agencies and accessibility to the stewardship
statements. If, in principle, the approach seems to be similar to the one
followed for the Governance Code, at the time of writing, it is still too early
to analyse the actual implementation and enforcement in more detail.
Individual FRC action against companies failing to implement the
Governance Code – or any of the other codes under the authority of the
FRC – has not been practised to date. However, there are some indica-
tions that the FRC may consider engaging more actively with deficient
practice, and that after having received a complaint about an individual
company, may consider engaging with that company.168
Several of the UK documents plead for maintenance of the self-
regulatory nature of these codes, reflecting some fear that ‘Brussels’
would impinge on this field.
Finally, to date there has been no judicial case law dealing with the
issue of codes on corporate governance. With respect to the legal status
of the code and its enforceability at law, one can only refer to an old case
relating to the Takeover Code, another self-regulatory instrument. The
court declared itself very reluctant to intervene on the substance of the
regulation.169
8. Preliminary findings
The overview of the different ways the national codes of conduct in the
field of corporate governance are being implemented and monitored
167
See Handbook Notice, www.fsa.gov.uk/pubs/handbook/hb_notice104.pdf asset (FSA
2010/57).
168
See FRPP Annual Report 2011, pp. 11–12. ‘In such circumstance, however, it would
need to be made clear that the judgment whether the governance arrangements adopted
by the company (as opposed to the description of those arrangements) were satisfactory
remained a matter for shareholders, not the Panel.’ See also FRC reform consultation §
5.9, ‘The intention would be to undertake supervisory inquiries to provide an under-
standing of the reasons for the collapse or near collapse of a public interest entity or
other issue affecting confidence in corporate governance and reporting.’
169
See R v. Panel on Takeovers and Mergers ex parte Datafin [1987] QB 815.
114 eddy wymeersch
should now be compared and evaluated on the background in which they

operate. This analysis may allow us to identify some general trends and
offer an insight in the way these corporate governance rules could be
made more effective.
8.1. The public or private character of the codes

In many jurisdictions the Corporate Governance codes are essentially
developed by the business firms and their associations. In some cases
there is no input from the public authorities or institutions, while in
others the input is very limited and takes the form of limited participa-
tion in the standard-setting body, or support in the standard-setting
process. Difficult to establish is the informal nodding by the public
institutions relating to the self-regulatory process, or their influence on
the appointment of members of the standard-setting body. Many codes
are private, but have an inkling of public interest, and therefore are not
indifferent to the public authorities that consider them as an alternative
to public regulation.
At the other end of the spectrum are the codes developed by the
securities regulator, as is the case in Portugal and Spain, where the
standard setting is ultimately the work of the regulator, while implemen-
tation is followed up and verified by the regulator as part of its monitor-
ing of the disclosure in the annual report. In some cases administrative
sanctions for violation of the code would be applicable, but obviously are
rarely imposed.
Between these two alternatives are several more nuanced forms. In the
UK the code is adopted by the FRC, a public sector regulator and
followed up by one of its subcommittees. But the code reflects strongly
the private sector mind and is the product of an extensive dialogue and
consultation of the private sector. Also, no enforcement or sanctioning
action has been undertaken, at least specifically on the basis of code
provisions. A different balance is found in France, where the AMF plays
a very visible role in the corporate governance debate and publishes a
report on the application of the code, although the latter is self-
regulatory.
8.2. The double-layered system

In a few jurisdictions, one sees an emerging trend to develop two layers
of recommendations, or codes: one addressed to the boards of listed
companies, and another addressed to their shareholders, especially the

institutional investors. As both levels are complementary, this approach
may lead to a more consolidated view of the developments relating to
governance issues.
Traditionally, most corporate governance codes were addressed to the
leaders of the companies and the issues they are confronted with:
composition, role of the chairperson, committees, remuneration, rela-
tions with management, etc. More rare are the codes where the role and
position of the shareholders are mentioned, and then only from the angle
of the board and the way it should deal with them (e.g. UK, Belgium).
The role of the shareholders, especially the institutional investors, has
recently been highlighted. Leading to the acknowledgement that their
monitoring action may be the other moving force in the governance
debate, authorities are increasingly relying on them, obliging them to
vote, or to engage in stewardship. The effectiveness of this action is
directly related to the structure of company ownership.
In countries where most of the publicly traded companies are domin-
ated by blockholders, or controlling shareholders, the corporate gover-
nance provisions usually reflect this reality. The way in which these
companies adapt to the corporate governance provisions is generally
very high (i.e. in the upper 90 per cents). The effectiveness of the code is
due to the acceptance by the directors – and usually also by the
shareholders – of the importance to be seen as adhering to the code, as
its provisions will generally not be in contradiction with their views. One
should also mention the role of the press, public opinion, and the
political world, especially as a consequence of their threat to adopt
hard law. All of these factors lead to high levels of acceptance of the
main provisions of the code. But on points that may potentially be
contrary to these shareholders’ interests, e.g. remuneration, anti-
takeover protection, or derogations from preferential subscription
rights, codes would be rather timid, if not silent.
This analysis is different with respect to countries where companies
are mainly characterised by dispersed share ownership, where the weak-
ness of collective action tools leads to negatively affecting the monitoring
role of shareholders on the boards. These companies are more exposed to
activist investors, building up significant stakes and putting pressure on
the board, sometimes leading to full takeovers. In these countries, insti-
tutional investors have organised themselves to weigh on companies’
decisions (UK, Netherlands). Comparable, but different, is the action
deployed by investor protection associations (Netherlands, Denmark).
116 eddy wymeersch
Third, in a separate class, are the attempts to mobilise the asset managers
to engage more actively with the companies in their portfolios (UK,
France). The action of these different groups of investors not only
addresses corporate governance issues, but views the entire range of
issues relating to the investee companies.
In studying the effectiveness of corporate governance codes, one
should also take into account this second layer of action, which is usually
not integrated in the codes, as the latter are mainly addressed to the
board and management. Different approaches will be commented on
later.
8.3. Comply or explain

Most of the self-regulatory codes discussed in the present Chapter are
based on a ‘comply or explain’ methodology, meaning that though they
may contain some binding principles, most of their recommendations
are not binding on substance, but allow the addressee to choose another
approach, in which case companies must provide appropriate explan-
ations.170 This does not mean that these codes are non-binding: the
national provisions adopted in implementation of the Fourth Company
Law Directive state that listed companies must designate a code that they
declare applicable to them (‘adoption of a code’). But pursuant to these
provisions, the application of the code is left to the company’s freedom. This
freedom is variable, as most codes contain different classes of provisions,
some of which are binding under the comply or explain regime, others are
not binding in the sense that companies may, but are not obliged to deal
with them, or do not have to state their reasons for not applying these
provisions.
The notion of ‘comply or explain’ is somewhat ambiguous and has
stirred some debate with respect to its place in the overall legal system
(Poulle 2008; Couret 2010). According to some, companies that comply
do not have to give any explanation, thereby avoiding any risk of possible
deficiencies in the explanations, and the liability that may be attached.
They consider that only in the case of non-compliance are explanations
due. Another group states – rightly – that the rule is: ‘explain if and how
you comply and explain if you do not comply, why and what your
170
See Statement of the European Corporate Governance Forum on the comply-or-explain
principle, 22 February 2006, available at htpp://ec.europa.eu/internal_market/com
pany/docs/ecgforum/ecgf-comply-explain_en.pdf.
alternative is’. Companies should state their understanding about their

governance model, and explain why they have chosen a specific formula,
which in their view may be better than any one proposed in a code. The
ultimate purpose is to inform the markets about how the company is
governed, and what its views are about the topics dealt with in the code.
Too many companies consider this as a ‘compliance’ exercise, mainly by
way of box ticking.
This technique leaves a great freedom to companies, and deliberately so:
in several jurisdictions the opinion lives that the codes allow companies to
structure their governance the way they see best, and would not curtail their
freedom to look for other, more effective governance techniques. After all,
pillars of today’s governance as independent directors, non-executive chair-
men, audit committees, lead directors, and so on, are due to governance
practice, but later picked up by regulation. This large freedom constitutes
one of the weaknesses of the code system: on the one hand it is very difficult
to gauge precisely what conduct lies behind the words in the governance
statement, and the degree of reliability of the statements; on the other, it is
well known that some statements are far from perfect.
What is a ‘proper explanation’? The published reports frequently
mention irrelevant, boilerplate explanations, carried over from year to
year, practice that should certainly be refused as an ‘explanation’. Several
jurisdictions have published guidelines about the appropriate character
of an explanation (Belgium, Netherlands, UK). An explanation should
be considered sufficient if it allows the normal reader to understand
which way the company is dealing with the specific issue, and why it is
doing so. Giving the rationale for the conduct, or how the derogatory
consequences have been mitigated, are part of a valid explanation. Mere
reference to tradition, to internal agreements, even to charter provisions,
are not convincing. Temporary derogations should be identified as such,
indicating the time period for which they will apply. The statistical data
therefore have to be accepted with caution. But what about reliability of
explanations? There have been cases of misleading explanations and
some suspected to have been false. Only in a few jurisdictions is there
any monitoring of the meaningfulness of the statements. The question
has been raised as to who is responsible internally for the statement: does
the board approve the statement, and what is the involvement of the
chair? Or is all this left to the corporate secretary or an assistant as a
necessary but not very meaningful exercise? In certain matters, how can
the explanation (e.g. about internal processes) be verified by an external
observer?
118 eddy wymeersch
8.4. How to measure effectiveness

The effective implementation of hard law provisions is very often not
directly measurable and can only be determined by a detailed observa-
tion or analysis of the sanctioning regime. This is rarely undertaken
in the traditional company law fields. On the contrary, regarding the
implementation of the governance codes provisions in all jurisdictions
compared, extensive statistical reports have been published, essentially
illustrating a strong level of adoption of the code.
Measuring effectiveness is a difficult exercise: it can be undertaken on
the basis of the disclosures, presupposing that these reflect reality. More
ambitious is the verification in a survey of the opinion of the different
business leaders involved. A third method is external monitoring, as is
undertaken in Portugal and Spain on the basis of the quality of the
disclosures. Each of these methods has it advantages, but also its
shortcomings.
There are some questions with respect to these statistical data.
Generally, they measure the overall adoption of the code (usually in
the upper 90 per cents), however, the individual items usually obtain
much lower scores. A large part of the detailed statement are not very
controversial: they reflect usual practice, and are carried over from year
to year. The statistics should focus on the other provisions, as there the
implementation is much lower, as was repeatedly evidenced by the data
about remuneration. It also appears that the statistical figures published
by monitoring commissions are considerably more optimistic than those
drawn up by investor associations, or even by regulators.171
Specifically in Germany, von Werder and Bartz (2012) have inves-
tigated the effectiveness of the 90 recommendations of the German code
on the basis of a survey with chairmen of the Vorstand and Aufsichtsrat.
In their view the conclusions are more optimistic than one might have
expected, but are still evidence of the weakness of the approach. They
depict an overall Codexmood (‘Kodex Klima’) that is moderately posi-
tive in the eyes of the interviewed, but is clearly more negative on specific
points such as the cooperation between the Vorstand and the
Aufsichtsrat. Their effectiveness analysis is based on internal informa-
tion, leading one to wonder how investors can assess governance on the
basis of the statements in the governance reports.
171
See the VEB Effect, 2009, nr. 26, p. 42, nt. 4; cf. Portugal, nt.137.
8.5. The codes versus hard law

In all jurisdictions compared there is, of course, a more-or-less detailed
set of hard law rules applicable to companies on the basis of the
Companies Act, accompanied by more-or-less elaborate rules stemming
from financial regulation. Corporate governance rules supplement these
legal provisions in the fields that are difficult to capture in hard, legally
sanctioned regulations, where no ready made solutions should apply.
This function is undoubtedly useful, more so as it allows for sufficient
freedom for companies to frame their own organisation and conduct
rules, flexibility that is more limited in law-based, even default rules. It
also strengthens the sense of ownership and responsibility of boards for
their own governance organisation.
The competition between the two sets of provisions is mentioned several
times in the national reports, but no further analysis is made. This menace
of a takeover of the soft law rules may be attributed to the lack of imple-
mentation of the corporate governance principles or provisions, the too
general character of some of these principles, making public authorities
distrustful of the outcomes, and at least for some of them, the bias in favour
of the company leaders’ positions. The process of ‘juridification’ is especially
visible in fields characterised by strong controversy (remuneration, gender
diversity) or slow progress (audit committees). The CRD IV contains
elaborate corporate governance provisions for credit institutions, where
the traditional voluntary provisions are considered to have been too weak
in light of the financial crisis. These provisions will become binding legal
rules, from which no derogation will be allowed, leading to quite an
important change in the system (Barret 2012; Winter 2012a). Here, too,
the informal, soft law-based action – often with the support of the
supervisors – has obviously not been considered sufficient to avoid formal
legislation to intervene.
In my view, the acceptance of the self-regulatory or soft law approach
to corporate governance hinges on the development of a stronger system
of monitoring and implementation and, if needed, some form of external
enforcement. Stronger implementation and credible enforcement are
essential to avoid corporate governance principles being further crystal-
lised in formal state legislation.
8.6. Drafting the codes

Part of the credibility of the corporate governance codes depends on
their draftsmen. In most jurisdictions, especially where the codes
120 eddy wymeersch
originated from the stock exchange area, leading business people took
the initiative, while the drafting took place under their guidance by their
assistants, usually with some flavour from academia, and in some cases, a
link to the regulator as well. The ministries seem to have been involved in
some cases, but this influence is difficult to assess due to its pluriformity.
In any case, the codes mainly reflect the concerns of the business leaders,
and as a consequence essentially address the issues they are confronted
with within boards, with the management and in their relations to
shareholders or other stakeholders. This business bias probably may
explain the reduced trust of the political world. However, the corporate
governance codes should not be used as alternatives to government
regulation: they introduce additional guidance principally with respect
to the internal functioning of companies, and cannot be used to pursue
public interest policies.
Although presenting some distinct differences, the national codes by
and large all reflect the same approach and express the same concerns.
Originally the Cadbury Code 1992 stood as their model, but since then
national diversity has taken hold. The drafting process has become more
refined over time, with public consultations on the basis of a proposal or
an exposure draft, feedback statements responding to the consultation,
stating reasons for the adopted solutions. A cost–benefit analysis has
rarely been found, and may sometimes be welcome. In practice, drafts-
men usually take inspiration from the codes of the neighbouring states,
and the experiences in other jurisdictions. But efforts could be under-
taken to better familiarise draftsmen with these evolutions, including the
case laws that have been rendered in other jurisdictions. Some coordin-
ating action undertaken by the FRC deserves support.
Access to the codes is greatly facilitated by their posting on national
websites and for the complete worldwide collection on the website of the
European Corporate Governance Institute (ECGI).
8.7. Observing the adoption and the application of the code

‘Adoption’ of the code is used here in the sense that companies acknow-
ledge that a code is applicable. This usually also means that the company
applies the code. In some jurisdictions (Denmark, Austria), no further
statement is necessary if all provisions of the code are complied with. In
others, the code calls for a description as to how the company complies
with the different provisions of the code, including the provisions which
it does not apply. This distinction is important in the description of the
monitoring of the codes: some monitoring only addresses ‘adoption’,

other includes ‘application’.
The follow-up of the corporate governance codes takes different
forms. In certain states, it is considered that both adoption and applica-
tion is a matter for the shareholders and the markets. Hence no specific
efforts are undertaken, although initiatives are mentioned to stir interest
for the corporate governance theme in general, and for the code in
particular.
In most EU countries, there is a body – a commission or other body –
systematically analysing the different corporate governance statements
and registering the responses in a survey, mainly for statistical purposes.
In this case it is the adoption that is measured. But as the analysis goes to
the level of the individual provisions of the code, the survey yields some
insight in the way companies deal with the different issues in the code.
This body is usually also in charge of the original drafting of the code;
it will also follow up on the code’s regular updating, whether pursuant to
its original mandate or on a self-appointed basis. The identity of these
observation bodies – here referred to as ‘governance commissions’ –
differs considerably: in several states, these ‘corporate governance
commissions’ are the reflection of the draftsmen of the codes and there-
fore originate mostly from the local business circles or the stock
exchanges. In others the distance from the business world is greater,
and some members are selected from academia, or even from the public
bodies, including ministries (Austria). The latter may also have a say in
the appointment of the members of these bodies (Netherlands,
Germany). Even in these cases, the codes remain essentially a self-
regulatory instrument, but with external monitoring as to adoption.
Intermediate cases are frequently found: the public sector securities
regulator may be actively involved, whether by publicly taking a position
on the way the code is being implemented, suggesting initiatives for
improving the content of the disclosures, defining what is a ‘proper
explanation’, or even offering interpretation of the code. In these cases,
there is a subtle transition from a purely self-regulatory instrument, to an
intermediate form going in the direction of regulation.
The UK position is special: the corporate code is part of the codes and
standards adopted by the FRC, ‘an independent regulator responsible for
promoting high quality corporate governance and reporting to foster
investment’. The FRC is not a government body in the traditional sense
of the word, but accomplishes tasks of public interest, some of which
have been delegated on the basis of different pieces of legislation.
122 eddy wymeersch
At the other end of the spectrum, one finds the Portuguese and
Spanish models, where the codes are ‘self-regulatory’ by name, but
have been drawn up by or in close coordination with the securities
commission, and are verified by the latter on the basis that the govern-
ance statements are part of the public disclosure made by listed com-
panies and hence subject to verification like any other public
information. These national bodies are more strongly involved in the
application of the code, analysing in depth the significance of the
explanations given. Adoption of the codes is mostly actively pursued in
surveys, whether undertaken by the governance commissions, by inde-
pendent third parties (academics, accounting firms), by the securities
commissions, whether alongside their action on applications (Portugal,
Spain)172 or not (Belgium, to some extent France).
There is no general tradition for the governance commissions to deal
with the application of the codes and verify the quality of the disclosures
and of the explanations, as distinct from the formal implementation.
Although references are made to this type of more intrusive monitoring,
few are the states where it is effectively undertaken. It would imply at
least that the governance commission is able to address itself to the
company’s top bodies and analyse the motives for not applying the
codes’ provisions. The absence of an appropriate monitoring technique
and investigative powers may explain why most monitoring is limited to
statistical observations and general analysis.
This alternative approach, however, is pursued in some jurisdictions,
most prominently in Spain and Portugal, where corporate governance
statements are actively analysed and negative findings discussed with the
company. A similar type of monitoring is found is Switzerland. Changes
in the disclosures are requested, and in the case of refusal would give rise
to disciplinary action. It would seem that the UK is also considering a
change in that direction.
By way of conclusion, the European jurisdictions compared present a
wide scale of answers to the question how the corporate governance
code’s implementation can be monitored. These differences reflect
fundamentally different legal traditions, and different business and
political environments. Therefore it will be difficult, if not impossible
to prescribe a single pattern for the implementation of the code through-
out the EU. This does not prevent some minimum level of monitoring
being pursued.
172
Where the verification has been internalised in the securities regulator.
8.8. Disclosing names

If the implementation of a corporate governance code has been insuffi-
cient, one might expect the monitoring commission, after having
respected due process and discussed the matter with the appropriate
bodies within the company, to disclose the names of the companies that
have resisted its recommendations. This would warn the markets about
deviations from the code, and also work to shame the unwilling company
(‘name and shame’).
This is not the usual approach of the monitoring bodies. One pre-
sumes that legal reasons – the rules on libel and slander – prevent them
from being outspoken about the kind of violation, and the identity of the
perpetrator. Moreover, the process of establishing breaches of the code
would also have to be clearly worked out, as publication will legally be
considered a sanction in some jurisdictions at least, triggering human
rights concerns. Public authorities would be reluctant to engage in this
type of action, as the code is a private statement that should not be
enforced by government action, unless authorised by the legislator.
Private bodies would not easily be mandated to engage in such disciplin-
ary action, although the listing conditions may confer this kind of power
(Switzerland, Luxembourg, Sweden, Denmark).
This modesty is not shared by the investor associations: the French
asset managers’ association, the AFG, publishes its ‘alerts’ by referring
to the agenda of the identified AGM and the items it criticises. The Dutch
association of institutional investors, Eumedion, screens the AGMs with
reference to the individual companies.173 In the Portuguese practice,
companies are mentioned by name, indicating, for example, their posi-
tion in the compliance scale. When action is undertaken, the outcome
of it is not, however, always mentioned. This is, of course, different from
the usually high-profile legal actions at the initiative of investor associ-
ations (especially in the Netherlands, France, and some in Germany
as well).
Based on techniques for comparing the quality of consumer products,
or financial services, one could also consider the publication of data
about critical governance points. This is attempted in the Netherlands,
173
See e.g. www.eumedion.nl/nl/public/kennisbank/ava-evaluaties/2011_ava_evaluatie.
pdf; see also the VEB Annual Report 2010, p. 13, according to which the Corporate
Governance Commission intends to name companies in breach of the code. VEB itself
publishes a list of purportedly independent directors that in its analysis are dependent;
see ibid., p. 17.
124 eddy wymeersch
where the VEB website contains data about the remuneration paid to the
CEOs of major Dutch firms.174
8.9. The general role of the shareholders in the codes’ effectiveness

The relatively recent involvement of the shareholders in corporate gov-
ernance regulation is obviously very different in countries with concen-
trated ownership versus those where shareholdings are widely spread.
However, one should warn against a simplistic view: in most European
countries the two models coexist, the dispersed model often applying to
the largest firms.
In companies with concentrated ownership, investors outside the
controlling group do not usually play an important role. Attendance at
the meeting by these investors is generally low, although institutionals,
acting through voting agents, may have swollen the numbers in the last
few years (Van der Elst 2012). The role the outside investors play is
generally limited, illustrated by the low percentage of negative votes or
abstentions,175 nonetheless they may weigh on the market price for the
company’s shares.
It is striking that, in both companies with concentrated and with
dispersed ownership, it is usual practice for boards to strive for adher-
ence and voluntary implementation of the corporate governance code,
and to avoid damage to reputation for deficient policies in this respect.
Both types of companies have been willing to make considerable pro-
gress in developing audit committees even before this was mandatory, to
develop adequate risk models and explain their risk-management tools,
and more recently to explain their business model and its viability. One
might refer to these examples as some form of ‘competition for
excellence’. More sticky has been the issue of remuneration, where
ultimately the legislator had to intervene (Lorsch and Simpson
2009).176 The same applies to board diversity.
174
On the website of VEB under the heading ‘bestuursvoorzitter’ with an indication of the
increase/decrease for the last year, and data about options and bonuses.
175
It is well known that these investors only cast a negative vote if no other way of
influencing the company’s decision has been successful.
176
Although in a certain number of cases investors have been able to refuse a proposed
remuneration plan: ‘Shell shareholders revolt on pay’, 19 May 2009, available at htpp://
news.bbc.co.uk/2/hi/business/8058103.stm. Recently, the ‘shareholder spring’ has wit-
nessed numerous cases of open criticism of remuneration, leading in some cases to
reduction of the amount, or even the departure of the CEO as his remuneration
increases notwithstanding the poor result of the company: ‘Moss, A. Aviva CEO, Resigns
Even in companies with concentrated ownership, this approach has

supported the standing of the company’s equity in the market, has
gained praise for the management, and last but not least, has supported
the market price of its equity to the benefit of both the blockholders and
the investing public.
The relevance of the corporate governance codes and their effect on share
price performance have been investigated by several American and
European researchers (Gompers et al. 2003). The opinions are divided.
Some papers draw attention to the need for stricter enforcement.
Whether institutional investors effectively influence general meetings is
also doubted. The following papers can usefully be consulted in this context.
Clacher et al. 2008 see an exceptionally strong link between corporate
governance in general and in performance, while MacNeil and Li Xiao
(2006) conclude that there is a strong correlation between performance
and non-compliance, meaning that shareholders of well-performing
firms tolerate non-compliance more easily. The conclusions of Shaukat
and Padgett (2005) point in the same direction: compliance matters, not
just as a box-ticking exercise, but as a trigger of real change in the
governance of large listed companies, for which shareholders are willing
to pay a premium. In Germany, some writers point to the beneficial role
of corporate governance codes: Zimmermann et al. (2004) concluded
that the degree of compliance with the code is value-relevant informa-
tion, attributing this to capital market incentives (or pressures) that
might lead to a broad adoption of the code’s recommendations, even
though the enforcement mechanisms connected to the code are relatively
weak. Drobetz et al. (2003) analysed differences in firm-specific corporate
governance and document a positive relationship between their Corporate
Governance Rating and firm value. More critical voices are heard from
Prigge (2010) pointing to limitations in the corporate governance empirical
analysis, while the role of the corporate governance code is doubted in
Nowak et al. (2006), who found that the code has little effect and state as one
of their conclusions that ‘further evidence to the hypothesis that self-
regulatory corporate governance . . .relying on mandatory disclosure with-
out independent monitoring and legal enforcement are ineffective and do
not positively influence shareholder value’. Van der Elst (2012) found that
the increased presence of institutional investors did not materially affect the
approval rates for the usual agenda items.
After Shareholder Revolt On Compensation’, 19 May 2012, available at 2012 www.huffing-

tonpost.com/2012/05/08/andrew-moss-aviva-ceo-resigns_n_1499209.html.
126 eddy wymeersch
In the companies with dispersed ownership, the corporate governance

rules have seen their full deployment. Institutional investors and hedge
funds have often taken an activist stand to convince the general meeting
of their point of view. In several cases battles were fought in public
opinion and sometimes successfully in the courts as well, most recently
and in some cases successfully, on the issue of compensation.177
In companies with a large, dispersed ownership structure, institu-
tional investors are usually fairly reluctant to undertake common action,
out of fear of being held to additional obligations, especially those
relating to ‘concert action’178 and to insider trading, or under the even
stricter rules of the US regulation FD on Fair Disclosure. Therefore they
usually prefer to remain passive, at most limiting themselves to the
exercise of their voting rights in cases where voting is mandatory.
Larger institutionals sometimes act individually and exercise pressure
on the companies’ governance (e.g. on appointments of directors and
their remuneration) and on strategy. In at least three jurisdictions, i.e.
France, the Netherlands and the UK, institutional investors visibly
exercise considerable influence on listed companies, especially with
respect to their governance. Methods are different: in the UK, an elab-
orate system, mainly based on the action of the asset managers, has been
put in place in the Stewardship Code. In the Netherlands, a powerful
association of institutional investors engages with the management,
while smaller investors act directly through an investor protection asso-
ciation, or pursue their objectives with public, and sometimes legal
action. In France, the two models are found, one being based on the
asset managers, the other represented by investor protection defend-
ers.179 Some of this action is deployed in highly visible public statements,
press interviews and legal action, while another part takes place in
discreet contacts and discussions. The effectiveness of these approaches
is difficult to measure, although the concern expressed in response to
activist (hedge) funds indicates that boards and management of target
companies are far from indifferent to the potential threat these investors
177
A prominent case is Storck, in which activist investors were able to rally a majority
around their proposition to split up the company, to the dismay of public opinion and
the press. See Court of Appeal Amsterdam (OK) 17 January 2007, LJN AZ6440, JOR
2007, 42 (Centaurus c.s /Storck).
178
Leading to additional disclosures under the Transparency Directive rules, or even to a
mandatory bid if the bid threshold is crossed.
179
Collete Neuville, founder of ADAM (Association de défense des actionnaires minori-
taires) is a well-known defender of shareholder rights in France.
represent to their position. However, it can safely be assumed that

corporate governance ideas are effectively relayed through organised
shareholders, whereby the invisible action – that has been mentioned
for some jurisdictions – is difficult to document, as investors prefer not
to influence the company’s policies by acting in the general meeting or in
public, which is considered a last resort. Normally, these large investors
will discuss the company’s policies including governance issues, with the
boards or with the management, but would be more reluctant to engage
with other investors out of fear of being held to the consequences of
concerted action. The Stewardship Code in the UK is partly based on this
assumption. Whether their opinion is taken into account often appears
to be doubtful. Therefore, in some instances, more forceful action has
been engaged: cases have been mentioned where several institutionals,
acting within their professional organisation, exercised collective pres-
sure to have a board member, or more likely a member of the manage-
ment, discreetly removed. But these are rather exceptional cases and only
known from hearsay. Further escalation of pressure takes place through
public statements – or even leaks – in the general meeting and legal
proceedings.
8.10. Tools for more effective monitoring by shareholders

The first main avenue to strengthen implementation of good governance
lies with the internal forces within the companies themselves, i.e. their
boards, especially the independent directors, shareholders and external
auditors (Weir et al. 2001). The board is the first in line to ensure good
governance, and should organise itself by attracting the necessary exper-
tise, create internal governance rules and procedures, institute govern-
ance committees, proceed to self-assessment and board evaluation,
externally if needed. The applicable laws have sharpened their role in
terms of liability.180
The auditors play a useful role in verifying the data as reflected in the
internal accounts, e.g. on remuneration (Hommelhoff and Mattheus
180
The legal liability of supervisory directors of German financial institutions (Aufsichtsrat)
has been sharpened by shifting the burden of proof to the directors and extending the
liability limitation period for listed credit institutions: ‘Gesetz zur Restrukturierung und
geordneten Abwicklung von Kreditinstituten, zur Errichtung eines Restrukturierungsfonds
für Kreditinstitute und zur Verlängerung der Verjährungsfrist der aktienrechtlichen
Organhaftung’ (Restrukturierungsgesetz, RStruktG).
128 eddy wymeersch
2003; Wohlmannstetter 2011). These different players could be called

‘internal’ monitors, as opposed to the ‘external’ ones to be discussed
below.
8.10.1. Making room for a more continuous dialogue

The shareholders are the ultimate beneficiaries of the corporate govern-
ance efforts undertaken by the companies and all other parties involved,
but the dialogue with them is often weak. Code provisions are adopted
without much consultation,181 and in companies the dialogue is limited
to casting one vote a year, usually through a voting agent.
Companies interact with shareholders once a year at the general
meeting: in many jurisdictions the AGM is usually a rather dull ritual,
although recent experiences with ‘say on pay’ may point to a reversal. In
other companies – well known in Germany – the AGM offers the scene
for the public expression of a very wide range of feelings and frustrations,
to the point that governance recommendations had to be adopted to
limit their length.182 What are the signals the companies receive from
their investors and how are these transmitted? In companies with con-
centrated ownership, the blockholder will indicate his views, leading
other shareholders to stand more or less aloof. From the company side,
signals are limited to the price evolution in the markets, and perhaps the
visit of one or two engaging shareholders. These are politely listened to,
or are received by the investor relations department. There would be
little real engagement, even less a sense of partnership. Companies
should be obliged to develop effective engagement procedures.
The essential elements would be that shareholders and investors
should be able to expose their concerns to the company management,
while the latter would be obliged to organise the necessary procedures for
allowing this dialogue. Such dialogue would contribute to avoiding
excesses or abrupt positions at the AGM, while withholding activist
investors from undertaking ill-informed actions. There are several
means for companies to open this more continuous dialogue, for exam-
ple, by the use of electronic means. This would modify the agency
relationship, interested shareholders being able to express their opinion
on a well-informed basis, and management having a feeling about
investor sentiment. The voting process as such would be the
181
See the German change of position in Von Werder and Bartz (2012), nt. 192.
182
See for Germany, where the Kodex recommends that a general meeting should not last
more than 4-to-6 hours (§ 2.2.4).
crystallisation of the relationship, not its exclusive and unique expres-

sion. It would also modify the role of the voting agencies, as they would
have to take into account the outcome of this dialogue of which they can
be an active part. Although this subject goes beyond the limits of tradi-
tional corporate governance concerns, one might explore to what extent
the objective could be included in the companies’ charter and further
organised on the basis of soft law standards.
8.10.2. Organising the role of institutional investors

Among the different classes of shareholders, the institutional investors
already play a prime role in the monitoring of governance and one should
further explore how this role can be made more effective. The obligation to
vote and report how the votes have been cast may be the least bad solution,
but is not a very convincing instrument, as institutionals will merely imple-
ment it without being really interested in or even aware of how the votes
have been cast. On the other hand, one cannot expect institutionals or asset
managers to follow up on each of the thousands of companies in their
portfolio. Feasibility and cost are the issues here (Van der Elst and
Vermeulen 2011).183 Selective but active engagement with investee compan-
ies seems a better alternative, provided the necessary guarantees are
introduced as to the negative consequences of this engagement, mainly
with respect to obtaining price-sensitive information that leads to block
trading or the application of the rules on concerted action. Solutions may be
found in internal procedures, similar to the Chinese walls that have been
adopted in investment firms.
The voting activity of institutional investors cannot be assessed with-
out considering the role of voting agents.184 For institutional investors
that hold hundreds, and in some cases, even thousands of lines of equity
investments, exercising their voting rights, as is now mandated in some
legislation, is practically impossible, particularly as most general meet-
ings take place within a timespan of about two months. In that sense,
voting agents offer an efficient solution to this challenge. However, the
intrinsic quality of this type of engagement should be questioned, as the
investor usually will not be able to make an individual determination
183
The authors see several cost elements that should be taken into account: (1) conven-
tionalism/micro-management, (2) management distraction, (3) risk aversion, and (4)
lack of transparency.
184
See the ESMA consultation ‘An Overview of the Proxy Advisory Industry.
Considerations on Possible Policy Options’, 22 March 2012; Fleischer (2011), propos-
ing to put proxy advisers under supervisory ‘Kuratel’, at 173.
130 eddy wymeersch
about how the vote should be cast taking into account the specific
circumstances of each investee. As a consequence, it will be the agent
determining how the principal will vote, based on the model schemes
that the voting agencies have developed. The argument that voting
agents discuss with their principals the way votes have to be cast seems
unlikely, due to the very considerable number of investees for which
votes have to be cast. By relying on the uniform voting instructions,
voting agencies acting for several institutionals amplify their impact,
increasing the risk of biased, or unfounded positions. Therefore, one may
wonder whether considered individualised voting should not be
preferred.
A more solid approach might be found in allowing institutionals – or
asset managers as their agents – to set up an entity,185 separate and
independent from their main portfolio, in which some shares could be
lodged of those companies in which stronger involvement would be
planned. This separate entity could act as the engaging shareholder,
with a clear mission to follow up on the affairs of the investee. The entity
would be forbidden from trading in the portfolio, which would eliminate
the conflicts of interest, and provided the legal prohibition on passing
inside information is adapted, most of the insider trading issue could be
eliminated. There should be an appropriate Chinese wall with the main
portfolio avoiding any suspicion of concerted action. Most importantly,
there should be no confusion in terms of objectives between the main
portfolio and the separate entity. The latter would be provided with a
clear separate governance, and a clear budget, on which the participants
in the main portfolio would have to decide. The present confusion about
who pays for governance activism would be eliminated.
8.10.3. Engagement and stewardship

How significant shareholders construe their relationship with the
investee companies presents a great individual diversity which is due
not only to differences in the percentages held – what is fairly obvious –
but also to the organisation of the controlling blocks and differences in
the legal and social environment in which these blocks are held. The
usual differentiation between companies with concentrated or dispersed
ownerships should be kept in mind. Companies with concentrated own-
ership have widely adhered to the corporate governance codes, although
these may have restricted the influence of their controlling shareholders.
185
Company, trust, foundation.
In companies with dispersed ownership, the governance matter is essen-

tially in the hands of the board, with little influence from the share-
holders and serves to rebalance the influence of the management. It is
essentially with the dispersed companies in mind that the proposals
about engagement and stewardship of the dispersed shareholders have
been developed; however, they deserve some further analysis in both
types of companies.
Winter (2012b) has recently proposed a three-layered form of involve-
ment by dispersed shareholders, which he called compliance, interven-
tion and engagement. He considers that much of the involvement of
institutional investors is in fact compliance with the applicable code or
instructions, by having adopted a mandatory voting policy leading to
thoughtless but mandatory exercise of voting rights, and entrusting a
voting agency with making the choices and casting the votes. As most
institutions have such large diversified portfolios, this is the only type of
involvement these investors can effectively practice. Intervention hap-
pens when shareholders enter into a dialogue with the management to
make it adjust its strategy or policy, often in order to increase share-
holder value or to support ESG objectives. The intervention of activist
shareholders belongs to this category, being characterised by the fact that
their action is usually a one-off. The third group, stewardship, refers to
the longer-term perspective of the engagement and the intention of the
investor to build a structural relationship. Therefore it presupposes a
long-term ownership, which is mostly incompatible with the regulation
applicable to some of these investors (particularly investment funds, held
to honour daily withdrawals). The relationship of these shareholders
bears some resemblance to that of the controlling shareholder, but
necessarily remains more high level, due to weak incentives and being
handicapped by the rules limiting access to confidential or price-
sensitive information and by the absence of effective instruments to
bend the company’s conduct. In the past, their action was limited to
recommendations or some often-discreet pressure, leading in a few cases
to changes in the board or in the management team. If they are able to
mobilise a sufficient number of shareholders behind their plan,186 these
institutionals may be able to trigger a bigger change, as has been illus-
trated by recent cases on ‘say on pay’. Decisions are than supported by a
vote in the AGM and reflect a long-term interest in the company, which
186
Even not a majority, as boards will be reluctant to oppose a large minority of
shareholders.
132 eddy wymeersch
is quite different from the actions undertaken by activist investment

funds who pursue short-term gains.
With respect to companies with concentrated ownership, the influ-
ence of the shareholders is more complex, covering a wide variety of
situations, with shareholders holding more than half of the votes, others
with a factual majority, while further down the majority may be based on
the alliance of a series of smaller blockholders. Control will often also be
based on control-enhancing mechanisms, among which the pyramids
occupy a central position. In front of this diversity, the way the views of
the control block is transmitted to the company will also vary signifi-
cantly: in the case of a majority owner, the company will likely be
informed about the owner’s view and at least will have sufficient regard
to it. The other types of controlling shareholders are likely to use more
indirect means, or safeguard support from other investors by following
policies agreeable to them (e.g. generous dividend distributions).
Common to all of these cases is that controlling power is usually asserted
by the possibility that the shareholder dictating or at least influencing the
appointment of himself or a number of allies on the board, but would
generally not extend to all decisions taken by the board. These share-
holders would normally only decide on significant aspects of company
life, especially the decisions that directly affect them, such as dividend
distributions, diluting share issues, changes of the articles of incorpora-
tion, business restructuring and, especially, mergers. Key appointments,
such as chairman and CEO, would also be at least pre-approved.
The possibility to practice ‘intervention’ by non-controlling share-
holders will depend on the coherence of the control block, but remains
possible even in non-majority-controlled companies. But even in fully
controlled companies, minority shareholders may claim for abuse of
majority powers which in some jurisdictions many lead to far-reaching
remedies. Activist shareholders have been able to mobilise the public
investors and impose drastic changes even in companies with block-
holders.187 In order to counter these attacks, companies have an interest
in presenting themselves as applying ‘good governance’, which will
effectively be achieved by pointing to the strict application of the govern-
ance code. The cited decisions of the Dutch Supreme Court could be an
example in this direction.
Stewardship is a more universal notion and could be applied in all
types of companies. It would, rather, function as a support to the
187
See the Storck case, n. 177.
incumbent board, but might conflict with the views of the controlling
shareholders, e.g. on their long-term development views. Up to now
there has been little experience with stewardship techniques in con-
trolled companies, where the matter is usually dealt with by the investor
relations department. Whether that will suffice in the future is question-
able, and boards may usefully look at better ways of communication.
Direct relations with controlling shareholders may also have to be
considered.
As to compliance, even if its overall usefulness can be doubted, it might
be worthwhile to restrict its mandatory use to specific items of the
agenda of the general meeting where the investors have a more direct
interest (e.g. anti-takeover devices).
8.11. External monitoring by the corporate

governance commissions
Several schemes have been developed to make the codes more effective,
and the acceptance of these schemes will depend on the legal, social and
political environment within which these codes have been adopted and
implemented. It is reasonable to admit that no one solution will fit the
whole European Union and beyond. Not only the content of the codes,
but also the methods of implementation have to take account of local
diversity, allowing for experimentation with a wide range of techniques,
some of which have been illustrated in the country overview.
A number of external drivers in support of effectiveness will be
mentioned only generally here, for lack of direct legal relevance. We
would mention the quite important influence of the press, public opin-
ion, the political world and the judiciary when they deal with code
provisions or concepts. Their effectiveness varies depending on the
individual issue (remuneration, gender diversity) and depends on the
existence of an alert press and public opinion. We will instead focus on
the structured external monitoring models, of which there are already
several in place.
8.11.1. Nature and role of the corporate governance

monitoring commissions
The strongest monitoring model is undoubtedly the Portuguese–Spanish
one, where the securities supervisor adopts the rules, assesses their
implementation and, if needed, takes enforcement action. In both cases
it is based on the inclusion of governance statements in the annual
134 eddy wymeersch
reports to which the securities supervisors extend their oversight in the

framework of the national legislation adopted pursuant to the imple-
mentation of the Transparency Directive.188 It is argued that, as this legal
basis applies throughout the Union, this supervisory regime should in
fact be applicable in the entire Union. However, the argument only
applies to a governance code that has a regulatory origin and when the
information to be disclosed is founded on a statutory provision, even if
the latter is of the ‘comply or explain’ nature. Therefore this model
cannot easily be exported to other jurisdictions, where governance
codes are the product of the private sector, where public authorities
might not be very willing to spend their time and means for enforcing
an instrument that originates outside the public sector and would oblige
them to impose objectives that are difficult to achieve, having been laid
down in a private instrument. In fact, the situation is more complex: in
Belgium and France the securities supervisors undertake elaborate scan-
ning of the governance practices, although the codes are undoubtedly of
a private nature. However, these regulators remain on the sideline as far
as non-regulatory standard setting and individual enforcement are con-
cerned, as this would otherwise change the corporate governance dia-
logue from a private, self-regulatory intervention to regulatory action.
On the other side, companies would strongly argue that a flexible,
preferably self-regulatory corporate governance model, would best
meet their needs. But it is undeniable that the trend goes in the other
direction, and that an increasing number of governance tools are likely to
become governed by official, or public regulation (e.g. on risk manage-
ment, diversity189).
Most corporate governance monitoring commissions have a private
law status, although with a varying degree of public involvement. These
bodies are very effective in the standard-setting process and in raising
awareness of governance issues. There is no particular call for adapting
their composition in depth: states that prefer a private body should be
able to continue to do so, similar to those relying on mixed formulas or
188
Transparency Directive 2004/109 of 15 December 2004. See art 4(2)(b) and art. 24(4)(h),
referring to the reporting framework. These provisions do not clearly oblige national
supervisors to verify the information in the annual reports.
189
The trend can be noticed in several EU states: Italy: Law 120 ‘Gender Balance on the
Boards of Listed Companies’ of 12 August 2011; Belgian Law on diversity, Act on
Gender Diversity L. 28 July 2011, nt. 22; and French act Loi relative à la représentation
équilibrée des femmes et des hommes au sein des conseils d’administration et de
surveillance et à l’égalité professionnelle, of 27 January 2011.
on a securities regulator. By allowing for this freedom of choice, private

companies will maintain ownership of the corporate governance project
and ensure its development, taking account of changing business prac-
tices. The presence of experienced business leaders will avoid code
provisions or decisions that might not reflect good business practice.
In some jurisdictions, no monitoring takes place: this is regrettable,
and should be remedied. The status of the monitoring body can be left
to a national decision, with a certain preference for private bodies
composed of experienced business leaders.
8.11.2. Monitoring tools

The external monitoring of actual governance practices essentially takes
place using statistical tools. Although these are useful, more attention
should be paid to the governance practices and explanations and to their
meaningfulness. This could be pursued by addressing specific practices
for all companies subject to the code, not in one particular, but across
several jurisdictions. A topical comparative insight is a powerful instru-
ment to identify weaknesses or recommend improvements, referring to
those who have already adopted the better discipline. Said comparative
action might lead to more de facto harmonisation and streamlining of
the disclosures, and contribute to the enhancement of governance prac-
tices as a consequence of ‘competition for excellence’. In the same vein,
more attention should be paid to the experiences in other jurisdictions,
not only with respect to the way the substantive issues are handled, but
also as to the means used for ensuring more effective implementation of
the codes.
The presumption that corporate governance reports reflect reality
should be subject to regular verification. Whether the auditor should
undertake this task – as is the case in some states – or whether it should
be undertaken in a direct dialogue with the companies, is open to debate.
But the knowledge that the disclosure may be subject to verification will
contribute to better practices and more meaningful disclosure. A risk-
based approach would be indicated here.
Corporate governance commissions should be able to engage with
companies and their stakeholders on specific subjects, e.g. topics that are
raised by investors that have been identified as outliers in the statistical
exercise, or more generally have received public attention. The corporate
governance commission should be entitled to interact with companies
with respect to their implementation of the code provisions and to
request explanations. In a high-level dialogue, the arguments for
136 eddy wymeersch
companies not applying certain provisions or not putting forward

appropriate arguments for departing from the code, should at least be
discussed. The discussion should not be bureaucratic or administrative,
but directly engaged with persons experienced and knowledgeable in
governance matters. On the other hand, companies should be obliged to
respond to the invitation and to answer the questions posed by the
monitoring commissions, particularly explaining the reasons for not
adopting the commission’s recommendations. But the last word remains
with the company.
If the foregoing remained ineffective, as the ultima ratio the corporate
governance commission should have the right to publish the company’s
name. In that case, it should be protected against legal action by the firm
criticised, except in the case of gross or wilful negligence. The protection
should extend to the rules on libel and slander. Before applying this instru-
ment, due process must be followed. As to whether other sanctions should
apply (such as civil liability or even professional disqualification), this is better
left to the applicable legal system.190 Said changes could be introduced on a
soft law basis, especially through contractual instruments191 or in the listing
conditions. The legal protection for the monitoring commissions might
require legislation. The concept of self-regulation and ‘comply or explain’
should be maintained, allowing for much desirable flexibility and avoiding
petrification. At the same time, the outcome should be credible and con-
tribute to healthier organisation and functioning of listed companies.
8.12. Is there a need for further harmonisation?

The issue of a Europe-wide harmonisation has been mentioned by
several national governance bodies. There is generally little love lost on
this idea. Therefore, it might be a valid alternative for governance
commissions to explore more in depth what they can learn from each
other and attempt to align their recommendations and move to more
common concepts. At the same time, as has been presented before, the
companies themselves could usefully strive to streamline their govern-
ance practices and disclosures. Voluntary initiatives based on a bottom-
up approach could be undertaken by the Europe-wide business
associations. This cooperative exercise has not been undertaken to date
190
So, e.g., in the Dutch system disqualification would belong to the measures that can be
ordered by the Enterprise Chamber (art. 2:356 Burgerlijk Wetboek).
191
Comparable to the code applicable to institutional investors in the UK.
and one may call on the national governance commissions to take

initiatives in this respect.
If none of these attempts were successful, it might be useful to develop
a series of high-level principles against which all national governance
codes could be measured, reflecting the common denominator among
the best practices as laid down in the codes today, but leaving the
national standard-setters free to adopt the principles or provisions that
are best adapted to their legal order.
The large diversity of company ownership and, hence, of governance
regimes in the EU, reflected in differences in the legal frameworks, are
strong arguments for avoiding a uniform approach to corporate governance
issues. This applies to both self-regulatory as well as legal interventions.
9. Recommendations
The chapter has identified a number of fields where the present status of
applying corporate governance principles could be improved. The fol-
lowing 10 recommendations are addressed to the national entities
responsible for developing, adopting and applying corporate governance
principles and codes and to the companies that apply these codes.
(1.) Corporate governance codes are useful instruments to deal with
governance issues. Their credibility will depend on the effective
application of the codes.
(2.) ‘Comply or explain’ is a sensible approach to corporate governance
issues: ‘comply’ should be understood as obliging companies to
extensively explain their governance model and related mecha-
nisms. ‘Explain’ should lead to proper, meaningful explanations,
especially in cases of non-compliance.
(3.) Companies should organise their contacts with investors on a more
frequent and intensive basis than merely at the AGM.
(4.) Institutional investors and asset managers should organise them-
selves to be able more actively to engage with investee companies,
avoiding restrictions in present regulations. The creation of a sep-
arate governance subsidiary with sufficient funding could usefully
contribute to that objective.
(5.) Regulations on concerted action and on insider trading should not
stand in the way of properly organised engagement efforts.
(6.) Companies should actively monitor their governance mechanism
internally.
138 eddy wymeersch
(7.) National corporate governance bodies should be installed in

charge of following up on developing and monitoring the applica-
tion of the codes. External monitoring should take place by senior
and experienced business persons.
(8.) Cross-border contacts, including monitoring on a cross-border
basis, can be a useful tool for developing a common benchmark
on certain governance subjects.
(9.) These corporate governance bodies should be entitled to engage in an
active dialogue with companies, allowing them to identify best prac-
tices, including with respect to the implementation of the codes
(10.) National corporate governance bodies should be entitled to publish the
names of companies with deficient corporate governance practices;
protection against liability and libel and slander should provided for.
References
Andersen, P. K. 2004. ‘Corporate governance in Denmark’, Stockholm Institute for
Scandinavian Law 45: 11–28.
Arcot, S. and V. Bruno 2006. ‘In Letter but not in Spirit: An Analysis of Corporate
Governance in the UK’, LSE Working Paper 031.
Bajpai, S. G. N., E. Berglöf and S. Claessens 2005. ‘Enforcement and Corporate
Governance: Three Views, Global Corporate Governance Forum’, Focus,
World Bank Series, available at www.ifc.org.
Barret, M. 2012. ‘The end of “comply or explain”? Corporate governance in the
United Kingdom and Ireland and the impact of CRD IV’, Journal of
International Banking Law and Regulation 27: 1, S. 4–12.
Barucci E. and S. Cecacci 2005. ‘On Ownership Structure, Investor Protection, and
Company Value in the Italian Financial Market’, Working Paper, available
at www1.mate.polimi.it/ingfin/document/barucci_ceccacci.pdf(sd).
Barucci, E. and J. Falini 2005. ‘Determinants of corporate governance in the Italian
financial market’, Economic Notes 34: 371–405.
Bebchuk, L. A. and H. Spamann 2010. ‘Regulating bankers’ pay’, Georgetown Law
Journal 98: 247–87.
Bedard, J. C. and K. M. Johnstone 2004. ‘Earnings manipulation risk, corporate
governance and auditors’ planning and pricing decisions’, The Accountant
Review 79: 277.
Berglöf, E. and S. Claessens 2004. ‘Enforcement and Corporate Governance’,
World Bank Policy Research Working Paper No. 3409.
Bianca, M. 2008. ‘The Evolution of Ownership and Control Structure in Italy in
the last 15 Years’ Banca d’Italia, Working Paper.
Bianchi, M. and L. Enriques 2001. ‘Corporate Governance in Italy after the 1998
Reform: What Role for Institutional Investors?’, Working Paper, available
at www.ssrn.com.
Bianchi, M., A. Ciavarella, V. Novembre and R. Signoretti 2011. ‘Comply or
explain: Investor protection through the Italian Corporate Governance
Code’, Journal of Applied Corporate Finance 23: 107–21.
Bratton, W. 2003. ‘Enron, Sarbanes-Oxley and accounting: Rules versus principles
versus rents’, Villanova Law Review 48: 1023–56.
Cheffins, B. and R. Thomas 2001. ‘Should shareholders have a greater say over
executive pay?: Learning from the US experience’, Journal of Corporate Law
Studies 1: 277–315.
Clacher, I., E. Doriye and D. Hillier 2008. ‘Does Corporate Governance Matter?
New Evidence from the United Kingdom’, Working Paper, available at
http://69.175.2.130/~finman/Turin/Papers/cdh011208.pdf.
Coffee, J. C., Jr. 2005. ‘A theory of corporate scandals: Why the USA and Europe
differ’, Oxford Review of Economic Policy 21: 198–211.
Coombes, P. and S. C. Y. Wong 2004. ‘Why codes of governance work’, The
McKinsey Quarterly 2: 48–53.
Couret, A. 2010. ‘The “comply or explain” principle, From a simple financial
markets regulation to a wide method of regulation’ Revue trimestrielle de
droit financier 4: 4–11.
Cronin, C. and J. Mellor 2011. An Investigation into Stewardship, Engagement
Between Investors and Public Companies: Impediments and their Resolution,
Foundation for Governance Research and Eucation (FGRE), available at
www.foundationgre.com.
De Cordt, Y. 2009. ‘Les enjeux de l’affaire “Fortis” Victoire a la Pyrrhus
pour l’Etat belge dans l’affaire Fortis’, Revue pratique des sociétés 7006:
351–67.
Drobrtz, W.,A. Schillhofer and H. Zimmermann 2003. ‘Corporate Governance
and Expected Stock Returns: Evidence from Germany’ ECGI Finance
Working Paper 11.
Fana, J. P. H. and T. J. Wong 2004. ‘Do external auditors perform a corporate
governance role in emerging markets? Evidence from East Asia’, Journal of
Accounting Research 43: 35–72.
Fleischer, H. 2011. ‘Zukunftsfragen der corporate governance in Deutschland und
Europa: Aufsichtsräte, institutionelle investoren, proxy advisors und
whistleblowers’, Zeitschrift für Unternehmens- und Gesellschaftsrecht 40:
155–81.
Gebhardt, D. 2012. ‘Aktuelle Rechtsprechung zur corporate governance’,
Zeitschrift Corporate Governance 1: 179–82
Gompers, P., J. Ishii and A. Metrick 2003. ‘Corporate governance and equity
prices’, The Quarterly Journal of Economics 118: 107–56.
140 eddy wymeersch
Hommelhoff, P. and D. Mattheus 2003. ‘Die Rolle des Abschlusspruefers bei der
Corporate Governance’, in Hommelhoff, Hopt and von Werder (eds.)
Handbuch Corporate Governance, pp. 639–45.
Hopt, K. J. 2009. ‘Obstacles to Corporate Restucturing: Observations from a
European and a German Perspective’, in Tison, De Wulf, Van der Elst
and Steennot (eds.), Perspectives in Company Law and Financial
Regulation, Cambridge University Press, pp. 373–96.
and G. Wohlmannstetter (eds.) 2011. Handbuch Corporate Governance von
Banken, Munich, C.H. Beck.
Kunz, P. V. 2010. ‘Swiss Corporate Governance – An Overview’, Working Paper,
available at www.iwr.unibe.ch/content/e7137/e7138/e8539/e8655/099–
134KUNZ_ger.pdf.
Lorsch, J. W. and K. Simpson 2009. ‘Executive Remuneration at Royal Dutch
Shell’, Harvard Business School Case No. 409–126.
Lutter, M. 2011. ‘Der Kodex und das Recht’, in Grundmann, Merkt and Haar
(eds.), Festschrift für Klaus J. Hopt zum 70. Geburtstag am 24. August 2010,
Unternehmen, Markt und Verantwortung, De Gruyter 2, p. 1025.
MacNeil, I. and L. Xiao 2006. ‘Comply or explain: Market discipline and non-
compliance with the combined code’, Corporate Governance: An
International Review 14: 486–96.
McCahery, J. and E. P. M. Vermeulen 2009. ‘Role of Corporate Governance
Reform and Enforcement in the Netherlands’, in Tison, De Wulf, Van der
Elst and Steennot (eds.), Perspectives in Company Law and Financial
Regulation, Cambridge University Press.
Millstein, I. M., S. G. N. Bajpai, E. Berglöf and S. Claessens 2005. ‘Enforcement and
Corporate Governance: Three Views’, Global Corporate Governance Forum
Focus 3.
Mohd, S. and A. Nariman 2007. ‘Challenges of public and/or private enforcement
of the Corporate Governance Code’, International Corporate and
Comparative Law Journal 6: 15–38.
Nowak, E., R. Rott and T. A. Mahr 2006. ‘The (Ir)relevance of Disclosure of
Compliance with Corporate Governance Codes – Evidence from the
German Stock Market’, Swiss Finance Institute Research Paper No. 11.
Pardo, C. and T. Valli 2012. ‘Exercise of Voting Rights by Asset Management
Companies in 2012’, AFG Research Paper.
Pietrancosta, A. 2011. ‘Enforcement of Corporate Governance Codes: A Legal
Perspective’ in Grundmann, Merkt and Haar (eds.), Festschrift für Klaus
J. Hopt zum 70. Geburtstag am 24. August 2010, Unternehmen, Markt und
Verantwortung, De Gruyter, p.1025.
Poulle, J. B. 2008. ‘L’apparition du principe “se conformer ou expliquer” en droit
français’, Revue trimestrielle de droit financier 4: 41–7.
Prigge, S. 2010. ‘Eine empirische Abschaetzung der Aussagekraft von Studien
ueber den Zusammenhang zwischen Corporate Governance und
Unternehmenserfolg’ in Grundmann Merkt and Haar (eds.), Festschrift
fur Klaus J. Hopt. zum 70. Geburtstag am 24 August 2010, Unternehmen,
Markt und Verantwortung, De Gruyter: S. 1153–66
Shaukat, A. and C. Padgett 2005. ‘The UK Code of Corporate Governance: Link
Between Compliance and Firm Performance’, ICMA Centre Finance
Discussion Paper No. 17.
Thomas, R., A. R. Palmiter and J. F. Cotter 2012. ‘Dodd-Frank’s say on pay: Will it
lead to a greater role for shareholders in corporate governance?’, Cornell
Law Review 97: 1213–66.
Unger, S. 2006. ‘Special Features of Swedish Corporate Governance’, Swedish
Corporate Governance Board, available at www.corporategovernance-
board.se.
Van der Elst, C. 2010. ‘The influence of shareholder rights on shareholder
behaviour’, Corporate Finance and Capital Markets Law Review 1: 50–62.
2011. ‘Revisiting Shareholder Activism at AGMs: Voting Determinants of
Large and Small Shareholders’, ECGI Finance Working Paper No. 311.
2012. ‘Shareholder Rights and Shareholder Activism: The Role of the General
Meeting of Shareholders’, ECGI Law Working Paper No. 188/2012, avail-
able at www.ssrn.com.
and E. P. M. Vermeulen 2011. ‘Europe’s Corporate Governance Green Paper:
Do Institutional Investors Matter?’, Tilburg Law School Research Paper
No. 014.
Von Haartman, A. 2010. ‘The Stockholm Stock Exchange Monitors Application of
the Code’, in Swedish Corporate Governance Board, Annual Report.
Von Werder, A. and T. Talaulicar 2009. ‘Kodex Report 2009, die akkeptanz der
empfehlungen und anregungen des Deutschen Corporate Governance
Kodex’, Der Betrieb 16: 853–1.
Von Werder, A. and J. Bartz 2012. ‘Corporate Governance Report 2012: Kodex-
Regime und Kodexinhalt im Urteil der Praxis’, Der Betrieb 16: 869–78.
Weir, C., D. Laing and P. J. McKnight 2001. ‘An Empirical Analysis of the Impact
of Corporate Governance Mechanisms on the Performance of UK Firms’,
Working Paper, available at www.ssrn.com.
Winter, J. 2012a. ‘The Financial Crisis: Does Good Corporate Governance Matter
and How To Achieve It?’, in Ferrarini, Hopt and Wymeersch (eds.),
Rethinking Financial Regulation and Supervision in Times of Crisis,
Oxford University Press.
2012b. ‘Shareholder engagement and stewardship, the realities and illusions
of institutional share ownership’, Revue trimestrielle de droit financier 1:
4–11.
142 eddy wymeersch
Wong, T. J. 2005, ‘Do external auditors perform a corporate governance role in
emerging markets? Evidence from East Asia’, Journal of Accounting
Research 43: 35–72.
Wymeersch, E. 2005. ‘Enforcement of corporate governance codes’, Common
Market Law Review 26: 113–38.
Wohlmannstetter, G. 2011. ‘Die Rolle des Jahresabschlusspruefers bei der
Corporate Governance von Banken’ in Hopt and Wohlmannstetter (eds.),
Handbuch Corporate Governance von Banken, p. 503.
Zimmermann, J., I. Goncharov and J. R. Werner 2004. ‘Does Compliance with the
German Corporate Governance Code have an Impact on Stock Valuation?
An Empirical Analysis’, University of Bremen Working Paper.
3
Restructuring in family firms: a tale of two crises

christian andres, lorenzo caprio
and ettore croci
1. Introduction
Family firms have attracted a lot of interest from researchers in corporate
finance. Many scholars have examined how family firms perform, both
in terms of stock returns and operating performance, comparing them to
non-family firms. The goal of these studies was, initially, to determine
whether a family firm can be considered as an inferior type of firm that
continues to survive because of the lack of investor protection in some
markets and the existence of private benefits of control that the family
can secure through its voting power. The studies that have examined the
relation between family ownership and firm value have produced mixed
results, usually with more favourable evidence for family control in
Europe (Barontini and Caprio 2006; Maury 2006; Sraer and Thesmar
2007; Andres 2008; Franks et al. 2011) than in the US (Anderson and
Reeb 2003; Villalonga and Amit 2006; Miller et al. 2007) and Asia
(Claessens et al. 2002).1
While the chapter relates to the above literature, it focuses on how family-
controlled firms respond to economic and financial crises. Recent literature
shows that some heterogeneity exists in the financial and investment
corporate policies of different controlling shareholders (Cronqvist and
Fahlenbrach 2009). This behaviour could be pronounced in periods of
crisis, when managers and controlling shareholders are asked to take several
important decisions to keep the company going through hard times. In
particular, families are a category of controlling blockholder whose deci-
sions may differ from those of other blockholders when facing a crisis.
Bertrand and Schoar (2006), among others, argue that family firms have
1
There is also some European evidence that finds a negative effect of family control on firm
performance, for example, Cronqvist and Nilsson (2003), Bennedsen et al. (2007).
143
144 c. andres, l. caprio and e. croci
longer investment horizons than their non-family counterparts, i.e. families

provide family firms, using Bertrand and Schoar’s (2006) words, with
‘patient capital’, allowing them to maximise long-run returns and, therefore,
be less affected by short-term market pressures (Stein 1989). Firms whose
decisions are more sensitive to short-term stock price fluctuations could
select strategies that the market rewards, i.e. rapid growth strategies in
bullish markets and cost-cutting strategies during bearish markets, even if
they are not optimal in the long-run. Family firms, whose controlling
shareholders have longer investment horizons, should be less prone to
react to short-term price movements than non-family firms, avoiding
excessive growth in good times, and downsizing in bad times. A relatively
large literature confirms that controlling families tend to adopt conservative
policies in investment decisions, especially acquisitions (Klasa 2007; Sraer
and Thesmar 2007; Bauguess and Stegemoller 2008; Caprio et al. 2011).
Sraer and Thesmar (2007) show that the long-term commitment of the
controlling family makes it possible to trade off lower employees’ wages for
higher job security.
On the other hand, controlling families tend to maximise their own
utility, which also depends on private benefits of control and not neces-
sarily on the shareholders’ value (Morck and Yeung 2004; Bertrand and
Schoar 2006). The fear of expropriation may penalise companies with
weak corporate governance. An economic crisis can worsen agency
conflicts between controlling and minority shareholders, because con-
trolling shareholders’ incentive to expropriate increases (Baek et al. 2004;
Zhou et al. 2011; Bae et al. 2012). Since family firms are among those
where the threat of expropriation of minority shareholders is highest,
they may experience a sharper decline in their valuation than non-family
firms during crises. To stabilise and maintain private benefits, families
may be averse to excessive growth, especially if this implies a dilution in
their voting power (Caprio et al. 2011). This aversion may lead family
firms to forgo more investment projects and grow less during boom
periods than non-family firms. On the other hand, the existence of
private benefits may lead families to prop up their firms during crises
with their own money to avoid bankruptcy and/or significant
downsizing.
Following the approach of Bae et al. (2012), we use both crisis and
non-crisis periods to test the behaviour of family firms in booms and
busts. We exploit two exogenous shocks that hit European firms to study
the response of family firms and compare it with that of non-family
firms. The two exogenous shocks are the crises that followed the bust of
restructuring in family firms: two crises 145
the dotcom bubble and the 2001 terrorist attacks, i.e. 2001–3, and the
credit and sovereign debt crisis of 2008–10. To observe the different
behaviour in good and bad times, the study covers a relatively long
sample period, from 1997 to 2010, which allows the examination of
two boom periods (1997–2000 and 2004–7), as well as two crises.
Our results show that family firms generally outperform non-family
firms, a finding that is well-documented in the empirical literature. We
find some evidence of differences in performance between the two
groups during economic crises. In fact, by observing the Q ratio, the
effect of the crisis appears stronger for family firms than for non-family
firms. With respect to investment decisions, results show that family
firms invest less and are more likely to downsize in crisis periods. This
finding can be interpreted as evidence for more efficient investment
decisions, as family firms seem to adjust their investment decisions to
changes in the investment opportunity set more quickly. On the other
hand, it could be evidence of the problems that small and closely held
firms face in obtaining outside financing during periods of crises. Our
results carry important policy implications regarding the debate on
corporate governance and the European financial system. They indicate
that access to a broader basis of outside capital is of particular interest to
small and medium-size corporations, a class of firms that is predom-
inantly controlled by families. As these firms tend to base most of their
outside financing on bank debt, the reluctance of banks to grant loans
during a financial crisis could deepen the downturn in an economy that
relies heavily on small and medium-size firms. It thus seems necessary
for attention to be directed from governance issues to better access to
financing and growth. One way could be to further develop a liquid
small-cap corporate bond market for European corporations.
The chapter contributes to the literature providing a direct test of how
family firms in developed countries behave during economic and finan-
cial crises. We are not aware of any study that directly tests whether
family and non-family firms respond to crises differently, with the
exception of Zhou et al. (2011), who examine the response of Thai
firms to the Asian crisis in 1997, finding that family firms and domestic
firms behaved more conservatively than foreign-owned companies. The
chapter also adds to a growing literature that examines the link between
corporate governance and firm value during an economic crisis (Johnson
et al. 2000; Mitton 2002; Baek et al. 2004; Bae et al. 2012). These works
show that firms with weaker corporate governance suffer most during
crises, because crises increase the controlling shareholders’ incentives to
expropriate minorities. Thus, our chapter contributes to the literature

regarding the relationship between corporate governance and firm value.
It is also related to the literature on the allocation of financial resources
and propping in business groups, which are mostly controlled by family
firms (Masulis et al. 2011). When a firm is in financial and/or operating
distress and does not have access to external financing, controlling
shareholders may come to the rescue of the ailing unit (Friedman et al.
2003; Gonenc and Hermes 2008; Bae et al. 2012).
The chapter is organised as follows. Section 2 presents the motivations
and the research question. Section 3 presents our definition of family
firm and of crisis periods. Section 4 describes the sample and the data
used in the chapter. Section 5 is devoted to the empirical analysis.
Section 6 presents the policy implications of the analysis and concludes.
2. Literature, hypotheses development and policy issues

2.1. Hypotheses development
The growing literature on family control highlights several character-
istics related to their behaviour, which allow the hypothesis that family
firms behave differently from firms with different ownership structures
(i.e. non-family firms) when they face a crisis. Since families provide
family firms with ‘patient capital’ (Bertrand and Schoar 2006), family
firms should maximise the long-run value of the firm and be in a position
to ignore, at least partially, short-term price fluctuations.2 Firms whose
decisions are more sensitive to short-term stock price fluctuations could
select the strategies that the market rewards, even if they destroy value in
the long run. Family firms, whose controlling shareholders have longer
horizons, should be less prone to react to short-term price movements
than non-family firms, avoiding growing excessively in good times, and
downsizing in bad times. To express it in a different way, family firms
potentially behave less pro-cyclically than non-family firms to maximise
the long-term value of the firm. We expect this reasoning to be reflected
in measures of market and accounting performance:
H(Performance): The decrease in market valuation and accounting

measures of performance during crises is lower for family firms.
2
Stein (1989) argues that firms with a longer investment horizon will suffer less from
managerial myopia and hence be more profitable in the long run.
The controlling family’s incentives may play an important role in deter-

mining the behaviour of family firms during crises because of private
benefits of control. Morck and Yeung (2004) and Bertrand and Schoar
(2006) observe that families are more concerned with their own utility,
which also depends on private benefits of control, than with maximising
the shareholders’ value. Private benefits of control can take different
forms. While expropriation of the company’s resources to the detriment
of minority shareholders is the most visible and negative form,3 private
benefits of control can also be non-pecuniary, like the pride of the
founder or a family member to run the company she or a relative
founded (Barclay and Holderness 1989) or to build a family legacy
(Bertrand and Schoar 2006). Private benefits of control may also have
a stabilising effect on family firms’ behaviour compared to that of non-
family firms. In fact, during boom periods, family firms may be more
reluctant than non-family firms to pursue aggressive growth strategies
because of the fear that the financing of them may lead to a dilution in
their voting power (Caprio et al. 2011). On the other hand, the presence
of private benefits of control may generate the incentive for the control-
ling family to fight harder for the survival of their companies during a
crisis. Controlling families may be willing to invest additional money in
their firms because they know that, if the company goes bankrupt, not
only will they lose the stream of the company’s cash flow, but also the
private benefits associated with it. Moreover, families may believe that
not only is the survival of their company at stake, but also the family’s
prestige.
However, recent literature shows that the fear of expropriation may
penalise companies with weak corporate governance. In fact, crises may
intensify the agency conflicts between controlling and minority share-
holders: controlling shareholders’ incentive to expropriate increases
when the returns on investment decrease (Baek et al. 2004; Zhou et al.
2011; Bae et al. 2012). To understand this incentive, consider a control-
ling shareholder that has two possible uses for the cash flow generated by
the firm: (1) to invest the cash flow in investment projects; (2) to divert it
for his own benefits. During a crisis, the return on the investment
projects goes down. The lower return on investments projects makes
expropriation, whose return does not usually depend on economic con-
ditions, more appealing. We therefore hypothesise:
3
For example, with intragroup transactions at below/above market transfer prices.
H(Private Benefits): The performance of family firms is negatively affected
by the incentive to maximise private benefits of control. This effect is
amplified during crises.
The potentially more conservative behaviour of family firms is supported

by a relatively large literature, which confirms that controlling families
adopt conservative policies in investment decisions. Families do not
often sell their controlling stakes to outsiders (Klasa 2007; Bauguess
and Stegemoller 2008), and the firms they control also make fewer
acquisitions (Sraer and Thesmar 2007; Bauguess and Stegemoller 2008;
Caprio et al. 2011). Zhou et al. (2011) find that, even during crises,
similar to other domestic institutions, family firms adopt conservative
behaviour in restructuring their business portfolios, avoiding fire sales
and holding on to their assets.
Andres (2011) and Pindado et al. (2011) observe that family-
controlled firms exhibit less investment-cash flow sensitivity than non-
family firms. Family firms are, therefore, less affected by cash flow
availability. While Andres (2011) does not directly analyse the behaviour
of family firms during the business cycle, his findings generate the
expectation that the family firms’ investing behaviour could be more
stable: they increase less in favourable business conditions, when profits
are high and internally generated cash flows plentiful; and they fall less in
recessionary times, when the opposite occurs. However, there is another
side to consider. Andres (ibid.) also shows that investments in family
firms are more responsive to the Q ratio, which proxies for investment
opportunities. This implies that family firms could invest relatively less
when the investment opportunity set deteriorates, which is the case
during recessions.
The effect of an economic crisis on family firms may also be softer
than on other firms because of controlling shareholders’ propping
behaviour. Masulis et al. (2011) observe that business groups throughout
the world, in the vast majority, are controlled by family firms. In several
countries, business groups also exist because they set up internal capital
markets that help the group’s units to overcome inefficiencies in the
external financial markets, especially in times of high uncertainty and
tight capital markets (Khanna and Palepu 2000). Furthermore, when a
firm is in financial and/or operating distress and does not have access to
external financing, other more profitable and financially sounder com-
panies of the business group may help the struggling company, by
propping it through cash injections in various forms (Friedman et al.
2003; Gonenc and Hermes 2008; Bae et al. 2012). This propping behav-
iour is associated with the incentives of the entities that control the
group, i.e. usually a family. These arguments lead us to hypothesise that:
H(Investments): Investments of family firms are less responsive to
economic downturns than those of non-family firms.
Stability during crises in family firms is not limited to investment. Sraer

and Thesmar (2007) find that French family firms, especially those
whose CEO is a member of the controlling family, pay lower wages to
their employees than non-family firms. They argue that this is possible
because family firms can provide their workers with long-term implicit
insurance contracts, under which most workers will keep their jobs even
during crises. Based on economic theory, long-term contracts between
shareholders and employees are needed in order to promote firm-
specific investments by employees (Williamson 1979). Given that formal
contracts are impossible, implicit contracts are a desirable solution.
However, firms may have strong incentives to renegotiate these implicit
contracts (because they cannot be legally enforced) and might not be
credible when making promises to employees. Therefore, shareholders
must gain the trust of employees in order to realise the possible gains of
implicit contracts. Due to their long-term presence in the firm, families
might have an advantage in credibly committing to these agreements.
Consistent with this argument, Astrachan and Allen (2003) find in a US
survey that family firms declare that they are committed to keep employ-
ment levels stable and to avoid downsizing during temporary market
downturns. Their evidence is confirmed by Lee (2006), who finds that in
the 2001–02 downturn family firms played a distinctive role in main-
taining employment stability. This leads to the following hypothesis:
H(employees): Employment in family firms is more stable during crises as
compared to non-family firms.
Summarising our hypotheses, we expect family firms to suffer less during

crises than non-family firms. In particular, family firms’ valuation and
performance is expected to be higher than those of non-family firms
during periods of economic and financial crises. However, potential conflicts
of interest between the dominating (family) shareholder and minority
shareholders – and the adverse consequences for performance – are expected
to be higher in crisis periods. Consistent with the idea that families
pursue long-term value maximisation and are willing to prop up
their firms through cash injections in difficult times, we also expect
the investment outlays of family firms to be less affected by crises. Finally,

family firms are expected to be reluctant to cut their workforce during crises
to avoid breaking up the implicit contract with their employees.
2.2. Policy issues

The topics examined in this chapter offer several policy implications that
are relevant to the current European debate. In the last few years, a
serious concern has been the slow pace of the recovery of many EU
countries after the credit crisis in 2008, which caused a sudden slowdown
in the growth of the world economy that translated into a deep recession
for OECD economies. Our results can shed light on the role of the
ownership structure of European firms and its effect on economic
recovery.
Three of the hypotheses framed in the previous section suggest a
positive role of family control in Europe during crisis periods, i.e.
H(performance), H(investments) and H(employees). According to the
three hypotheses, family control should exert a stabilising role in limiting
the decrease of the firms’ profitability, the curtailing of investments in
real assets, and the downsizing in workforce. Taken together, the three
hypotheses describe the ‘bright side’ of family control. However, the
literature also suggests a ‘dark side’, which is accounted for in H(private
benefits). Although, as noted, private benefits may have a positive role in
rallying the family to prop up the firm during a crisis, they may well
induce the family to extract resources from the firm, especially in times
of crises, to the detriment of minority shareholders and of the viability of
the economy. Therefore, it is interesting to look at the results of the
empirical evidence to determine which side, the bright or the dark,
appears to have prevailed.
The policy implications are the following. If the bright side prevails,
no action to redress the balance between family and non-family control
is necessary.4 If the dark side prevails, a signal that a problem between
controlling shareholders and minority shareholders exists would be
generated. This signal should be taken into account with particular
attention in the debate about corporate governance in the EU, as also
4
Family control is more diffuse in Continental European countries than in other develop-
ment countries like the UK and the US, even though family ownership is also present in
the latter.
mentioned in the European Commission’s 2011 Green Paper (herein-

after 2011 Green Paper).
Our analysis provides an additional policy implication. The recent
credit turmoil includes a banking crisis of unprecedented proportions, at
least in the last few decades. In this respect, the crisis has been different
from many downturns that occurred in the second half of the twentieth
century.5 The majority of middle-sized and – usually – family-controlled
firms in Continental Europe is heavily dependent on banks to provide
external capital, and therefore could have suffered more because of the
limited propensity of banks to lend. This could have affected firm
performance after 2008, possibly more so in the case of family firms,
which tend to avoid equity issues to maintain control. If this proved to be
true, a further policy issue would be raised, i.e. the need to create
financing channels different from bank lending in Continental Europe.
In conclusion, as is clear from the 2011 Green Paper, Continental
Europe’s corporate landscape presents well-known specificities, which
raise the issue of the trade-off between the costs and benefits of a more
stringent corporate governance regime for the medium-size and family-
controlled firms. The empirical evidence presented and discussed in the
following sections has a significant bearing on the cost–benefit analysis
of the evolution of regulation.
3. Family control and crises

3.1. Family control
Several definitions of the term family firm are used in the literature.
Following Caprio et al. (2011), we consider as family controlled any
company in which a family or an individual is the largest ultimate owner
(in terms of voting rights) at the 10 per cent threshold. Ownership
information is from Worldscope and Orbis databases and from stock
market information repositories supplied by private publishers and
regulators. To obtain accurate ownership data for each year in the
sample period, these sources have been integrated with information
disclosed in annual reports, in the investor relations sections of the
companies’ websites and with information in the financial press.
5
Some exceptions exist. For example, the severe banking crisis in Sweden in the early
1990s.
The definition we used is well rooted in the literature, with many

authors using similar definitions, i.e. considering a 10 per cent owner-
ship threshold (La Porta et al. 1999; Faccio and Lang 2002; Maury 2006;
Dahya et al. 2008; Laeven and Levine 2009), a 20 per cent ownership
threshold (La Porta et al. 1999; Faccio and Lang 2002), or even a 25 per
cent stake (Franks et al. 2011). However, using a higher threshold does
not significantly affect the number of family firms. In our European
sample, using 20 per cent as threshold to determine family control would
reduce the observations we classify as family firms (at the 10 per cent
threshold) from 4,614 to 4,230, a reduction of only 384 (8.32 percent).6
In well over half of the observations classified as family firms according
to the 10 per cent definition (53 percent), the controlling family owns
more than 50 per cent of the voting rights.
Other definitions of family firm employ criteria that are not based on
ownership alone. Miller et al. (2007) and Andres (2008; 2011) use
slightly different definitions of family firm, which give more emphasis
to the role of the founding family. In particular, Andres (2008; 2011)
considers as family firms only those where the founder and/or family
members hold more than 25 per cent of the voting shares or, in case the
founding family owns more than 5 per cent but less than 25 per cent of
the voting rights, they are represented on either the management or the
supervisory board. According to Andres (2008; 2011), the founder is the
person who founded the sample company or, when a person acquires a
majority stake in a company and runs the company as CEO, changed the
company’s operational business significantly.
Anderson and Reeb (2003) define family firms as those in which the
founder or a member of the founder’s family by either blood or marriage
is an officer, director, or blockholder, either individually or as a group,
without imposing additional ownership requirements. Villalonga and
Amit (2006) add more restrictive criteria that include stricter require-
ments of ownership and/or the presence of family members in mana-
gerial positions. Following Caprio et al. (2011), this study focuses on the
identity of the largest shareholder holding a large voting power. We
believe this is an effective instrument to identify who has the actual
power to take decisions, especially when ownership is concentrated.
6
In unreported regressions, we test the robustness of our results using a different family
firm definition with a higher ownership threshold of 20 per cent. Our main results are not
sensitive to the change in the definition of a family firm, with coefficients pointing in the
same direction and comparable significance levels.
This study also considers factors beyond mere ownership influence,

like the presence of family members in the top managerial positions.
When this is the case, the family influence in the company is more
pervasive than when the family exerts control only by means of their
ownership stake. To consider such issue, this chapter analyses how
sensitive our results are to the inclusion in the definition of family firm
of requirements that can be added to the ultimate ownership, such as the
presence of family members in management positions. To this end, three
dummy variables are employed: Family CEO; Founder CEO; and Heir
CEO. Family CEO is a dummy variable that takes value 1 if a family
member is also the CEO of the company. Founder (Heir) CEO is a
dummy variable that takes value 1 if the founder (descendant) is the
CEO of the company. Finally, a variable for professional CEOs in family
firms, Professional CEO, is employed if a person unrelated with the
controlling family is the CEO of the family firm.
3.2. Crisis years

Our sample period includes two exogenous shocks that allow us to
examine and compare the decisions of family and non-family firms.
The first shock is the crisis that followed the burst of the dotcom
bubble and the terrorist attacks in the early 2000s (the dotcom crisis).
The second shock is the global financial crisis that started in the US
in the autumn of 2007, peaked after the Lehman Brothers collapse in
September 2008 and worsened in 2010, especially in Europe. We label
this the credit crisis. Different measures that can help identify years
considered as periods of financial crisis around these two exogenous
shocks are investigated. A first obvious measure is the GDP growth
rate.7 From these data, we can easily identify: (1) a marked slowdown
in the economic growth of Continental European countries between
2001 and 2003; (2) a contraction in the period 2008–09, which
extends its effect to 2010, also because of the sovereign debt crisis.
Another measure that can be employed is the annual performance of
the main local stock market index. Using this measure, the dotcom
crisis hit between 2000 and 2002, while the credit crisis had its main
impact in 2008 and in 2010, when the stock market indexes in many
7
GDP data are available from the World Development Indicators & Global Development
Finance database of the World Bank.
large European countries like Italy, France, Spain and Switzerland

posted negative yearly signs.8
While not relying on a single measure, we use the information pro-
vided by these two measures to correctly identify the start and the end of
the crises. In fact, while the GDP growth rate is backward-looking and
enables the capturing of the years in which firms faced economic and
financial difficulties in their product markets, a forward-looking meas-
ure like the market return permits us to time the beginning of the crises
to capture the effect of expectations. Using the two criteria, we consider
the three-year period 2001–03 following the bust of the dotcom bubble
as the dotcom crisis period and the period 2008–10 as the credit crisis
period. Thus, Dotcom crisis is a dummy variable that takes value 1 in the
years 2001–03, and Credit crisis is a dummy variable that takes value 1 in
the years 2008–10. Finally, the variable Crises, which captures all the
crisis years in our sample period, is the sum of Dotcom crisis and Credit
crisis, i.e. a dummy variable that takes value 1 in years 2001–03 and
2008–10.
4. Sample, data and control variables

4.1. The sample
We analyse how family firms react to crises and compare their responses
to non-family firms using a sample of large publicly listed Continental
European firms, in order to examine this issue at the European level.
As argued in the hypothesis section, conflicts of interest between the
controlling family and minority shareholders are expected to influence
firm performance. This type of agency conflict is generally less relevant
in the Anglo-Saxon world, where the classic agency conflict between
managers and shareholders in the sense of Berle and Means (1932) is
more important (Franks and Mayer, 2001). Franks et al. (2011) also
provide evidence that the UK has institutional characteristics very differ-
ent from Continental European countries. We therefore focus on
Continental European firms. Our European sample is an extension of
the sample of 777 large publicly traded companies used in Caprio et al.
(2011), who analyse the 1997–2007 period. We extend the sample period
8
We also rely on an additional criterion to determine the crisis period: the number of
IPOs. IPOs are highly cyclical and tend to be positively correlated with the economic
cycle (Jenkinson and Ljungqvist, 2001). A look at the number of IPOs confirms our
choice of crisis periods, especially the dotcom crisis.
for three more years up to 2010, thus covering the 14-year period from
1997 to 2010. Caprio et al. (2011) create this sample including listed non-
financial firms available from Worldscope whose total assets exceeded
US$250 million at the end of 1997.9 Thus, to be included in this sample,
the firm has to exist at the end of 1997. No later entry is allowed. The
sample includes companies from the following countries: Belgium (24);
Denmark (38); Finland (37); France (161); Germany (144); Italy (72);
Luxembourg (2); Netherlands (77); Norway (40); Spain (46); Sweden
(64); and Switzerland (72).
4.2. Data and variables

Variables unrelated to family control and crises are constructed starting
with data obtained from Worldscope and Datastream. The variables that
will be used as dependent variables in the multivariate analysis are
presented first: Q Ratio, ROA, I/K, Investment, Downsizing (Increase in
size), Wages and Change in employees. Q Ratio and ROA are often used to
measure firm performance (Anderson and Reeb 2003; Barontini and
Caprio 2006; Villalonga and Amit 2006; Andres 2008). Q Ratio is defined
as the sum of total assets and market value of equity minus common
shareholders’ equity scaled by total assets.10 ROA is the return on assets
(ROA), defined as EBITDA over total assets. Firms’ investments are
measured with the variables I/K, Downsizing and Increase in Size. I/K
is a variable traditionally used in the literature that analyses the
investment–cash flow sensitivity (Kaplan and Zingales 1997; Cleary
1999; 2006; Andres 2011). It is measured as the ratio between invest-
ments in fixed assets (capital expenditures) and lagged net fixed assets.
Downsizing and Increase in Size capture the change in the firm’s total
assets, another proxy for the investments and divestitures made by the
firm during a given year. Downsizing (Increase in Size) is based on the
annual growth rate in total assets, and we consider the firm to have
downsized (increased in size) if in year t the growth rate of total assets is
smaller (greater) than −10 per cent (10 per cent). We choose 10 per cent
as a threshold to identify large changes in the firm’s total assets. Finally,
the third set of dependent variables concerns employees and their wages.
9
The cut-off at US$250 million was chosen to limit to manageable terms the data-
collection process.
10
Sometimes the log of the Q ratio is used in lieu of the Q ratio (Barontini and Caprio,
2006). Our results do not change if we use the log transformation of the Q ratio.
Wages is defined as the ratio between salaries and benefits expenses and
the number of employees. Following Sraer and Thesmar (2007), the log
of this ratio is used in the multivariate analysis. Change in employees is
the percentage change in employees between year t-1 and year t.
In addition to family-related variables already discussed in
Section 2.1., we consider explanatory variables that are known to be
associated with firm performance. As argued above, the incentives to
maximise private benefits of control at the expense of minority share-
holder will likely have an effect on firm performance. Therefore another
variable, Wedge, is considered, being defined as the difference between
voting and cash flow rights for the controlling shareholder.11 The sepa-
ration between ownership and control rights, captured by Wedge, affects
the controlling shareholders’ incentives because it signals that the frac-
tion of cash flow rights is lower than the voting power, a situation that
may lead to expropriation. Therefore, Wedge is expected to negatively
affect firm performance. Total Assets is the firm’s total assets, a proxy for
firm size (the log transformation appears in the multivariate analysis).
The variable Leverage is defined as the ratio of the book value of financial
debt to the book value of total assets. Dividend/BV Equity is the ratio
between the cash dividend paid by the firm to their shareholders and the
book value of equity. Both debt and dividends are ways to mitigate the
risk of minority shareholder expropriation by limiting the amount of
cash under the control of the dominant shareholder (Jensen 1986; Faccio
et al. 2001; Villalonga and Amit 2006). Return Volatility is the daily stock
price volatility computed using daily returns over one year. Finally, the
one-year growth rate of sales (Sales Growth) is considered, in order to
control for firm growth.
We also expect the firm’s age to have a significant effect during crises.
In fact, young companies, which in most cases have not accumulated a
large amount of fixed assets and have not yet hoarded huge piles of cash,
are expected to be affected by the crisis more than older companies. It
might be more difficult for young companies to raise external finance,
especially debt, if there are few assets that can be pledged as collateral.
Moreover, they usually lack strong and stable cash flows to borrow
against. Since equity issues during a crisis period are not particularly
11
As proxy for Wedge, we also use a dummy variable that equals 1 if a family firm employs
measures that dilute the one-share–one-vote principle. Results are similar to those
obtained with the difference between voting and cash flow rights and are therefore not
reported.
appealing to existing shareholders because of the depressed stock prices

and because investors are usually reluctant to buy new equity, debt is
often the only outside finance option for firms during crises. Therefore,
young firms can be penalised more than older companies during crises.
The variable Age is computed as the difference between the sample year
and the year the firm was established.
A second set of variables regards investment regressions. Together with
family and crisis variables and the proxy for size, the following are consid-
ered: CF/K is the ratio between the firm’s cash flows and the lagged net fixed
assets; cash flows are defined as the sum of operating income and depreci-
ation; CF/K measures current liquidity (Cleary 1999). Market-to-Book,
which measures growth opportunities (Fazzari et al. 1988; Cleary 1999), is
defined as the ratio of the market value of equity divided by common equity.
CF/K and Market-to-Book are included in the Kaplan and Zingales (1997)
model as well as in Cleary (1999; 2006). Following Almeida et al. (2004) and
Andres (2011), who argue that firms facing financing constraints retain cash
to finance investment projects, we control for the firm’s cash position. The
variable Cash Holdings is the firm’s liquid assets, defined as the ratio of cash
plus tradable securities over total assets. Sales is the firm’s net sales in the
year. Leverage is also included in the investment regressions because firms
with high leverage invest less (Hennessy 2004). Finally, in the wages and
employees regressions, we follow Sraer and Thesmar (2007) and include the
log of age and total assets. Past performance (ROA) and stock market
volatility are also added as control variables.
5. Empirical analysis
5.1. Methodology
The multivariate analysis relies on Ordinary Least Squares (OLS) and
logit regressions with the double-clustered (or Rogers) standard errors
suggested by Petersen (2009) and Thompson (2011) to account for
unobserved time and firm effects. The choice of the type of model
depends on the fact that family ownership is stable over time (see, for
example, Franks et al. 2011). As observed by Sraer and Thesmar (2007),
the stability of family control does not allow identifying firm fixed effects
when the model includes a family status variable. In all regression
models we include industry and country fixed effects to control for
differences at industry and country level. The model employed estimates
the following regression:
Y it ¼ α þ β1 Familyit þ β2 Crisest þ β3 Familyit Crisest þ γXit

þ δIndustryi þ θCountryi þ εit ð1Þ
where Yit is the outcome of interest for firm i by year t; Familyit is the
dummy variable for family control; Crisest is a dummy for whether the
crisis has affected the firm at time t; Familyit x Crisest is the interaction
variable that measures how family control affects the variable of interest
in a crisis year. Industryi and Countryi are fixed effects for industry and
country, respectively, Xit are relevant individual controls and εit is the
error term.
The outcome variable Yit is either Q ratio or ROA in the performance
regressions; I/K, Investment, or Downsizing/Increase in Size in the invest-
ment regressions; and, finally, the log of Wages or the Change in
Employees in the wages/employees regressions. In some regressions,
other family-related variables are employed instead of Familyit, such
as: Family CEO, Founder CEO, Heir CEO, and Professional CEO; while
Dotcom Crisis and Credit Crisis appears in lieu of Crisest.12
The main variable of interest is, of course, Familyit x Crisest, which
captures the different response to the crisis by family firms and non-
family firms, compared to non-crisis years. To put it another way, the
coefficient β3 corresponds to the difference between the change in the
dependent variables for family firms (FF) and the change in the depend-
ent variable for non-family firms (NFF) between crisis periods (CP) and
non-crisis periods (NCP):
b3 ¼ ðFF in CP FF in NCPÞ ðNFF in CP NFF in NCPÞ ð2Þ
We expect the β3 coefficient to be positive and significant in the perform-

ance regressions. A positive coefficient is also expected in the investment
regressions and in the employment/wage regressions, indicating that
family firms do not cut their investments/wages and fire employees as
much as non-family firms.
In addition, to emphasise the effect of the crisis on family firms net of
the effect on non-family firms, the regression model allows the comput-
ing of the effect of family (and non-family) control in both good and bad
times. From the interaction between family control and crisis period,
four types of situations are possible: (1) family firms in a crisis period; (2)
family firms in a non-crisis period; (3) non-family firms in a crisis
12
It goes without saying that the interaction also changes.
period; (4) non-family firms in a non-crisis period. These four cases

correspond to the following combinations of coefficients of the model in
Equation (1):
(1) Family in a crisis period α (Constant)+ β1(Family)+
β2(Crisis)+ β3 (Family*Crisis)
(2) Family in a non-crisis period α (Constant)+ β1(Family)
(3) Non-family in a crisis period α (Constant)+ β2(Crisis)
(4) Non-family in a non-crisis period α (Constant)
Tests between cases (1) and (3) and cases (2) and (4) allow us to verify
whether family perform better than non-family firms during crisis (case
1 vs. case 3) and non-crisis periods (case 2 vs. case 4). It is worth noting
that the coefficient β1 (Family) provides the same information of this
latter test: in fact, the coefficient β1 (Family) estimates the increase/
decrease in the dependent variable due to family control in a non-crisis
period. The test to determine whether family firms outperform non-
family firms during crises, i.e. the test between cases (1) and (3), on the
other hand, corresponds to a test of the linear combination between
coefficient β1 (Family) and β3 (Family*Crisis).
5.2. Univariate evidence

As a preliminary step of our empirical analysis, descriptive statistics for
the sample of 777 European firms from 1997 to 2010 are presented. In
particular, we look for differences between family firms and non-family
firms both in crisis periods and in non-crisis periods.
Table 3.1 presents descriptive statistics of the variables used in the
analysis for the 777 companies in our sample. To remove the effect of
inflation over our sample period, all monetary variables in level (i.e. average
salary, total assets and sales) are expressed in constant 2005 Euros using the
consumer price index from the World Bank’s World Development
Indicators (WDI) & Global Finance Indicators (GDF) database.
Panel A of Table 3.1 shows that the performance of our sample firms is
negatively affected by economic crises, as both the mean and median Q
Ratio (market performance) and ROA (accounting performance) are
significantly lower in crisis periods. As expected, firms seem to invest
significantly less, showing a lower I/K ratio and lower nominal invest-
ments during crises. Accordingly, we observe a significantly higher
(lower) percentage of firms that experience a decrease (increase) in
total assets of more than 10 per cent in the crisis subsample.
Table 3.1 Performance, family ownership and crisis (777 companies)
Panel A Full sample and crisis years
Full Sample No Crisis Crisis

Mean Median No. of Obs. Mean Median No. of Obs. Mean Median No. of Obs.
Family 0.58 1.00 7,929 0.58 1.00 4,876 0.59 1.00 3,053
Family CEO 0.19 0.00 7,929 0.19 0.00 4,876 0.20 0.00 3,053
Professional CEO 0.39 0.00 7,929 0.38 0.00 4,876 0.40 0.00 3,053
Founder CEO 0.05 0.00 7,929 0.05 0.00 4,876 0.04 0.00 3,053
Heir CEO 0.15 0.00 7,929 0.14 0.00 4,876 0.15 0.00 3,053
Wedge 9.23 0.00 7,929 9.68 0.00 4,876 8.51 0.00 3,053
Q 1.48 1.21 7,785 1.58 1.28 4,792 1.31*** 1.11*** 2,993
ROA 0.12 0.12 7,756 0.13 0.12 4,766 0.10*** 0.11*** 2,990
I/K 0.35 0.19 7,747 0.42 0.21 4,761 0.24* 0.18*** 2,986
Downsizing 0.12 0.00 7,929 0.09 0.00 4,876 0.19*** 0.00*** 3,053
Increase in size 0.34 0.00 7,929 0.41 0.00 4,876 0.23*** 0.00*** 3,053
Wages 49.92 44.93 7,554 47.56 44.44 4,623 53.65*** 45.75*** 2931
Employees 27,123.11 7,598.50 7,732 25,033.95 7157.00 4,749 30,449.10*** 8,521.00*** 2,983
Delta Empl. (%) 2.89 0.01 7,691 4.66 0.03 4,717 0.07 0.00*** 2,974
Leverage 0.26 0.25 7,813 0.25 0.24 4,814 0.27*** 0.27*** 2,999
Total Assets 7,273.54 1,430.66 7,814 6,471.84 1294.25 4,814 8,560.01*** 1,676.25*** 3000
Age 91.24 89.00 7,776 89.51 87.00 4,775 94.00*** 91.00*** 3,001
Dividend/BV Eq. 0.05 0.04 7,744 0.05 0.04 4,764 0.06 0.04* 2,980
Return Volatility 0.02 0.02 7,912 0.02 0.02 4,866 0.03*** 0.02*** 3,046
CF/K 0.58 0.34 7,758 0.57 0.34 4,766 0.6 0.34 2,992
Market-to-book 2.46 1.68 7,702 2.75 1.90 4,752 1.99*** 1.35*** 2,950
Cash Holding 0.11 0.08 7,812 0.11 0.08 4,813 0.11 0.08 2,999
Sales 6,221.33 1,487.19 7,804 5,658.26 1,368.45 4,805 7,123.46*** 1,721.18*** 2,999
Sales Growth 0.07 0.06 7,791 0.11 0.08 4,790 0.02*** 0.01*** 3,001
Panel B Time series
Surviving Firms
Inflation Adjusted Values Nominal values
Year Employees Total Assets Sales Wages Total Assets Sales Wages
1997 6.471 1,114.966 1,240.494 43.373 1,004.148 1,111.436 38.075

1998 7.211 1,219.377 1,320.035 43.573 1,096.211 1,167.121 38.154
1999 7.727 1,431.72 1,489.066 44.592 1,313.067 1,344.211 39.913
2000 8.185 1,712.113 1,703.683 45.701 1,578.091 1,564.000 41.787
2001 8.987 1,732.607 1,844.313 46.772 1,608.563 1,753.004 43.674
2002 8.503 1,616.827 1,670.28 46.812 1,538.232 1,621.114 44.399
2003 8.664 1,538.15 1,666.995 46.261 1,491.259 1,619.000 44.878
2004 8.501 1,579.512 1,713.024 45.992 1,560.294 1,686.351 45.302
2005 8.525 1,724.251 1,729.431 46.056 1,724.251 1,729.431 46.040
2006 9.189 1,878.475 1,892.106 45.588 1,906.244 1,921.907 46.211
2007 9.6225 2,083.798 2,087.428 44.928 2,160.029 2,148.896 46.253
2008 10.0815 2,100.62 2,103.634 43.674 2,232.153 2,278.600 46.615
2009 9.690 2,169.465 1,883.655 45.359 2,320.17 2,038.308 48.387
2010 9.944 2,420.863 2,095.229 47.047 2,635.077 2,250.764 51.201
Panel C Family and crisis
No Crisis/No Family Crisis/No Family No Crisis/Family Crisis/Family

Mean Median Mean Median Mean Median Mean Median
Family CEO 0.00 0.00 0.00 0.00 0.34 0.00 0.33 0.00
Professional CEO 0.00 0.00 0.00 0.00 0.66 1.00 0.67a 1.00a
Founder CEO 0.00 0.00 0.00 0.00 0.09 0.00 0.07a 0.00a
Heir CEO 0.00 0.00 0.00 0.00 0.24 0.00 0.25a 0.00a
Wedge 4.28 0.00 3.52 0.00 13.62 9.31 11.95*, a 6.14a
Q 1.56 1.29 1.30*** 1.13*** 1.60 1.28 1.31*** 1.10***,a
ROA 0.13 0.13 0.10*** 0.11*** 0.13 0.12 0.11***, c 0.11***
I/K 0.25 0.20 0.24 0.18*** 0.54 0.21 0.24** 0.18***
Downsizing 0.10 0.00 0.19*** 0.00*** 0.08 0.00 0.19*** 0.00***
Increase in size 0.38 0.00 0.22*** 0.00*** 0.42 0.00 0.23*** 0.00***
Wages 48.44 46.93 56.71*** 48.16*** 46.91 42.23 51.56***, a 43.67***,a
Employees 30,799.13 9,086.00 37,009.47*** 10,599.50** 20,788.59 5,938.00 25,934.44***, a 7,259.00***,a
Delta Empl. (%) 0.08 0.02 0.10 0.00*** 8.05 0.03 0.05 0.00***,b
Leverage 0.24 0.23 0.27*** 0.26*** 0.26 0.25 0.27*** 0.27***, c
Total Assets 9,294.89 1,620.90 12,217.71*** 2,194.39*** 4,418.63 1,122.34 6,053.04***, a 1,490.17***, a
Age 93.38 96.00 95.98 100.00 86.72 83.00 92.66***, c 89.00***,a
Dividend/BV Eq. 0.06 0.04 0.06 0.04 0.05 0.04 0.05 0.04c
Return Volatility 0.02 0.02 0.03*** 0.02*** 0.02 0.02 0.03*** 0.02***
CF/K 0.52 0.34 0.67 0.34 0.60 0.34 0.56 0.34
Market-to-book 2.80 1.93 2.06*** 1.38*** 2.72 1.89 1.95*** 1.32***,c
Cash Holding 0.10 0.07 0.10 0.07 0.12 0.09 0.12a 0.09a
Sales 8,127.55 1,988.00 10,338.55*** 2,412.55*** 3,868.79 1,080.60 4,927.75***, a 1,352.03***,a
Sales Growth 0.10 0.08 0.02*** 0.01*** 0.11 0.08 0.02*** 0.02***
Note: Table 3.1 presents descriptive statistics of the variables used in the analysis for the 777 companies in our sample. All values
are in Euros. Panel A presents statistics for the full sample, and for firm-year observations during crisis (Crisis) and non-crisis
(No Crisis) periods. Family is a dummy variable that takes value 1 if the firm is classified as a family firm. A firm is defined as
a family (non-family) firm if its ultimate owner is (is not) a family member. Family CEO is a dummy variable that takes value 1 if
the firm is classified as a family firm and a family member is CEO. Founder (Heir) CEO is a dummy variable that takes value 1
if the founder (heir) is the family firm’s CEO. Professional CEO is a variable that takes value 1 if the family firm CEO is a
professional manager unrelated to the controlling family. Wedge is a dummy variable that takes value 1 in the case of divergence
between voting rights and cash flows right of the controlling shareholder. Q Ratio is defined as the sum of total assets (Worldscope
Item WC02999) and market value of equity (WC08001) minus common shareholders’ equity (WC03501) scaled by total assets
(WC02999). ROA is the return on assets, defined as EBITDA over total assets (WC18198/WC02999). I/K is the ratio between
investments in fixed assets (capital expenditures, WC04601) and lagged net fixed assets (WC02501); Investments is defined as the
sum of all outlays on capital expenditure (WC04601), acquisitions (WC04355) and R&D (WC01201) less receipts from the sale of
property, plant, and equipment (WC04351), and depreciation and amortization (WC01151). Downsizing (Increase in Size) is a
dummy variable that takes value 1 if the firm’s total assets (WC02999) decreases (increases) by at least 10% with respect to the
previous year. Wages is defined as the ratio between salaries and benefits expenses (WC01084) and the number of employees
(WC07011) in constant 2005 Euros. The change in employees is the percentage change in employees between year t-1 and year t.
We consider control variables that are known to be associated with firm performance; investments; salaries and employment.
Leverage is the ratio of the book value of financial debt as a percentage of the book value of total assets (WC03255/WC02999).
Total Assets is the firm’s total assets, a proxy for firm size (WC02999) in constant 2005 Euros. Age is the firm’s age, computed as
the difference between the sample year and the year the firm was established. Dividend/BV Equity is the ratio between cash
Notes to Table 3.1 (cont.):
common dividends (WC05376) and the book value of common equity (WC03501). Return Volatility is the standard deviation
of daily stock returns over the year. CF/K is the ratio between the firm’s cash flows (operating income (WC01250) plus
depreciation (WC01151)) scaled by lagged net fixed assets. Market-to-book is the ratio of the market value of equity (WC08001)
to common equity in (WC03501). The variable Cash Holding is the ratio of cash plus tradable securities to total assets (WC02001/
WC02999). Sales is the firm’s net sales in the year (WC01001) in constant 2005 Euros. Sales Growth is the one-year growth rate in
sales (WC01001), winsorized at the 1st and 99th percentiles. Panel B reports the median value of the time series of Employees,
Total Assets, Sales, and Salary in constant 2005 Euros and in nominal values for the subsample of firms that did not exit our
sample during the period 1997–2010. Panel C presents the same descriptive statistics of Panel A for the subsamples of family and
non-family firms in crisis/non-crisis periods. We consider crisis years the years 2001–2003 and 2008–2010. In Panel A, ***, **,
and * denote statistical significance of the difference in means (medians) tests at the 1%, 5% and 10% levels, respectively, between
crisis and non-crisis years. In Panel C, ***, **, and * denote statistical significance of the difference in means (medians) tests at
the 1%, 5% and 10% levels, respectively, between crisis and non-crisis years for the two subsamples of family and non-family firms.
The symbols a, b, c, denote statistical significance of the difference in means (medians) tests at the 1%, 5% and 10% levels,
respectively, between family and non-family firms in crisis years.
Surprisingly, the statistics show significantly higher average salaries,

higher levels of employees, as well as higher sales and total assets during
the crisis periods. This is somewhat counterintuitive, but might be
explained by the distribution of crisis and non-crisis periods in our
sample. As each of the two economic boom (or recovery) periods
precedes a crisis period (i.e. the dotcom crisis is preceded by the internet-
driven boom period of the late 1990s and the credit crisis is preceded by
the recovery period of 2004–07), our crisis observations are on average
biased towards the later years of our sample period. Since the sample
statistics that measure nominal figures are expressed in constant 2005
Euros, inflation cannot explain this result, but these values in levels can
still be affected by the compounding of the real growth that took place in
the earlier years. To improve the understanding of these variables in
panel B, we report the median value of time series of Employees, Wages,
Total Assets and Sales for the subsample of firms that survived the entire
sample period, i.e. firms that have observations from 1997 to 2010, both
in constant 2005 Euros and in nominal values.13 The time series clearly
shows a growing trend in these variables, with the sole exception of sales
and the number of employees between 2008 and 2009, which can explain
the results in panel B.
Differences between family and non-family firms during crisis and
non-crisis periods are summarised in Panel C of Table 3.1. In roughly
two-thirds of the family firm-year observations in our sample, family
firms are run by managers outside the family (Professional CEOs).
Accordingly, one-third of the family firm observations are firm-years
in which a family CEO manages the firm. The majority of family CEOs
are descendants (70 per cent of all family CEO observations) who
inherited control from the founder. Regarding the comparison of crisis
to non-crisis periods for the two subsamples, the data reflect the obser-
vations for the pooled sample in Panel A, i.e. lower performance and
lower investments in crises for both groups. In line with previous papers
on family businesses, the average (and median) family firm in our sample
is significantly younger (Age) and smaller (Total Assets and Sales) than
the average non-family firm. As documented in Sraer and Thesmar
(2007), family firms also pay significantly lower wages.
13
Results are similar if we use the full sample.
5.3. Multivariate evidence

Our multivariate analysis begins with an examination of the relationship
between firm performance and the ownership characteristics of our
sample firms. Table 3.2 contains the results of double-clustered OLS
regressions. Models (I) and (II) are based on market performance and
use the Q Ratio as the dependent variable, while models (III) and (IV)
explain accounting performance, measured as ROA. In line with the
existing literature (e.g. Anderson and Reeb 2003; Barontini and Caprio
2006; Andres 2008), we find a significant and positive effect of family
ownership on firm performance, in terms of both market and accounting
performance. The coefficient of Family, which measures the difference in
performance between family and non-family firms in non-crisis years, is
not only statistically significant (at the 0.05 and 0.01 level, respectively),
but also points to an economically significant difference between family
and non-family firms. As expected, crisis periods have a negative and
significant effect on firm performance. This effect seems to be more
pronounced during the second downturn, i.e. the credit crisis.
With regard to our main hypothesis, a significant coefficient is revealed
for the interaction term of Family and the crisis dummy in model (I). The
negative coefficient of Family*Crises in the regression for the Q ratio
indicates that crises have a more severe impact on family firms. In fact,
family firms, in addition to the negative effect measured by the variable
Crises, also bear a higher incremental loss, as measured by the interaction
variable. As Column (I) of panels B and C clearly show, this does not imply
that family firms suffer worse performance than non-family firms in abso-
lute terms during crises. In fact, family firms outperform non-family firms,
in both good and bad times. However, everything else held constant, crises
cause a comparatively larger decrease in the Q Ratio of family firms. In the
ROA regression (Model III), results for the family and crises dummies are
similar to those of the Q Ratio regression, but the interaction variable is not
significant. This indicates that in terms of the impact on ROA, the effect of
the crisis is similar for family and non-family firms. Panels B and C again
show that family firms outperform non-family firms.
While models (I) and (III) contain only one dummy variable that
combines the two crises, models (II) and (IV) aim at analysing the crises
separately. In particular, model (IV) provides further evidence that the
two crises need to be considered separately. The coefficients show a
significantly better accounting performance of family firms during the
dotcom crisis. Combined with the magnitude of the negative and
Table 3.2 Performance, family ownership and crisis (regressions)

Panel A Regressions
Q ROA
I II III IV
Constant (a) 1.3201*** 1.3004*** 0.1097*** 0.1053***
[0.4255] [0.4294] [0.0292] [0.0286]
Family (b) 0.1592** 0.1603** 0.0108*** 0.0110***
[0.0674] [0.0676] [0.0040] [0.0040]
Crises (c) −0.2118*** −0.0158***
[0.0438] [0.0050]
Dotcom Crisis (d) −0.1908*** −0.0159***
[0.0379] [0.0050]
Credit Crisis (e) −0.2426*** −0.0160**
[0.0838] [0.0073]
Family*Crises (f) −0.0154** 0.0043
[0.0071] [0.0036]
Family*Dotcom Crisis (g) −0.0172 0.0097***
[0.0317] [0.0032]
Family*Credit Crisis (h) −0.0112 −0.0028
[0.0324] [0.0046]
Wedge −0.002 −0.002 −0.0001 −0.0001
[0.0014] [0.0014] [0.0001] [0.0001]
Leverage −1.1396*** −1.1435*** −0.1009*** −0.1013***
[0.2490] [0.2488] [0.0127] [0.0128]
Ln(Assets) 0.0351* 0.0360* 0.0035* 0.0037*
[0.0201] [0.0203] [0.0020] [0.0020]
Ln(Age) −0.1684*** −0.1672*** −0.0107*** −0.0105***
[0.0537] [0.0538] [0.0032] [0.0032]
Dividend/BV Equity 1.2739** 1.2746** 0.0863*** 0.0862***
[0.5469] [0.5473] [0.0323] [0.0326]
Return Volatility 3.1235* 3.1856* −1.0708** −1.0598**
[1.7478] [1.7355] [0.4374] [0.4364]
Sales Growth 0.5933*** 0.5926*** 0.0635*** 0.0633***
[0.1215] [0.1227] [0.0083] [0.0079]
Industry/Country FE Yes Yes Yes Yes
Adj. R2 0.1904 0.1904 0.1396 0.1404
Observations 7537 7537 7507 7507
Table 3.2 (cont)

Panel B Tests
Q ROA
I II III IV
Non-Family/Non-Crisis 1.3201*** 1.3004*** 0.1097*** 0.1053***
(a) [0.4255] [0.4294] [0.0292] [0.0286]
Family/Non-Crisis 1.4793*** 1.4607*** 0.1205*** 0.1164***
(a+b) [0.4117] [0.4152] [0.0280] [0.0274]
Non-Family/Crises 1.1083*** 0.0939***
(a+c) [0.4265] [0.0284]
Family/Crises 1.2521*** 0.1090***
(a+b+c+f) [0.4117] [0.0293]
Non-Family/Dotcom Crisis 1.1097** 0.0894***
(a+d) [0.4293] [0.0289]
Non-Family/Credit Crisis 1.0578*** 0.0893***
(a+e) [0.4463] [0.0266]
Family/Dotcom Crisis 1.2527** 0.1101***
(a+b+d+g) [0.4148] [0.0276]
Family/Credit Crisis 1.2069*** 0.0976***
(a+b+e+h) [0.4281] [0.0270]
Panel C Family vs. non-family (F-tests)
Q ROA
I II I II
Non-Crisis 5.58** 5.63** 7.41*** 7.700***
0.0182 0.0177 0.0065 0.0055
Crises 6.42** 8.93***
0.0113 0.0028
Dotcom Crisis 8.75*** 17.16***
0.0031 0.000
Credit Crisis 4.53** 2.19
0.0334 0.1386
Note: In Panel A the table presents the results of OLS regressions with double-
clustered standard errors where the dependent variable is the Q Ratio in Columns
I and II and ROA in Columns III and IV. Q Ratio is defined as the sum of total
assets (Worldscope Item WC02999) and market value of equity (WC08001)
minus common shareholders’ equity (WC03501) scaled by total assets
(WC02999). ROA is the ROA, defined as EBITDA over total assets (WC18198/
WC02999). Family is a dummy variable that takes value 1 if the firm is classified as
family firm. A firm is defined as a family (non-family) firm if its ultimate owner is
(is not) a family member. Crises is a dummy variable that takes value 1 in years
2001–3 and 2008–10. Dotcom Crisis (Credit Crisis) is a dummy variable that takes
value 1 in years 2001–3 (2008–10). Wedge is the divergence between voting rights
and cash flows right of the (ultimate) controlling shareholder. Leverage is the ratio
of the book value of financial debt as a percentage of the book value of total assets
(WC03255/WC02999). Ln(Assets) is the logarithm of the firm’s total assets
(WC02999) in constant 2005 Euros. Ln(Age) is the log of the firm’s age. Dividend/
BV Equity is the ratio between cash common dividends (WC05376) and the book
value of common equity (WC03501). Return Volatility is the standard deviation of
daily stock returns over the year. Sales Growth is the one-year growth rate in sales.
All values are in Euros. All regressions include industry and country fixed effects.
Industry classification is based on the Fama and French 12-industry classification.
Firm and time (double-) clustered standard errors are in brackets and statistical
significance is denoted by ***, **, and * for the 1%, 5%, and 10% levels,
respectively. Panel B reports the coefficients for the combinations of Family and
Crises (Dotcom Crisis/Credit Crisis), obtained as a linear combination of the
coefficients in Panel A. Standard errors are reported below the coefficients. Panel
C reports the results of F-tests, with their p-value, for the tests for differences
between family and non-family in the different periods (non-crisis; crises; dotcom
crisis; credit crisis).
significant coefficient of the Dotcom crisis dummy, the coefficient of

Family*Dotcom crisis implies that the performance of family firms is
still negatively affected by the crisis, but to a significantly lower extent
compared to non-family firms. This result is even clearer looking at
Column IV of Panels B and C: the coefficient of 11.01 per cent for family
firms during the dotcom crisis is reported significantly above the 8.94 per
cent for non-family firms and barely below the 11.64 per cent for family
firms in non-crisis years. It should be noted that the regression analysis
also controls for industry and age, which implies that this effect is not
driven by an industry or life-cycle effect that might lead to a lower
exposure of family firms to internet-related industries.14
14
Also, due to the size threshold of $250 mentioned in Section 4.1., by construction our
sample contains very few internet stocks.
As outlined in Section 2, several empirical papers have documented

that the performance of family firms is particularly strong as long as the
founder is still active as CEO (e.g. Anderson and Reeb, 2003; Villalonga
and Amit 2006), a finding referred to as the ‘founder effect’. The analysis
in Table 3.3 breaks down the family dummy into four sub-categories,
depending on the position of the CEO. In models (I) and (III) we
compare the performance of family firms with a family CEO with the
performance of family firms with a professional (i.e. outside) CEO and
the performance of non-family firms. The coefficient estimates show that
both family firms with a family CEO and family firms with a professional
CEO perform significantly better than non-family firms (see also panels
B and C). The size of the coefficients indicates that family-managed firms
perform better than professionally managed family firms, showing
higher coefficients in terms of Q (model I) and ROA (model III). In
contrast to these findings, the role of the CEO does not seem to affect
family firm performance during the two crises.
However, the differences between family and professional CEO are
not significant. Model specifications (II) and (IV) contain a more fine-
grained definition of the Family CEO dummy, depending on whether the
founder is still active (Founder CEO) or has passed on control to a
descendant (Heir CEO). In contrast to Villalonga and Amit (2006), but
in line with most papers on Continental European samples, the results
show that descendants positively affect the performance of family firms.
Panels B and C show that European family firms outperform non-family
firms in terms of Q ratio, irrespective of the type of CEO. However,
family firms outperform non-family firms only when they are run by a
descendent in the ROA regressions. Again, we find only very limited
evidence of differences for the effect of the two crises between family and
non-family firms. Only one interaction term (Founder CEO*Crises in
model (IV)) shows a positive statistically significant coefficient.
The next step analyses differences in the investment policy of family
and non-family firms. First, investments in net fixed assets (i.e. machin-
ery, real estate, etc.) and capital expenditures as a whole are considered.
The first two columns of Table 3.4 presents the results of OLS regressions
(again, with double-clustered standard errors) of I/K, which stands for
the ratio of investments in fixed assets to lagged net fixed assets. In line
with theoretical considerations, we find that investments are positively
influenced by the availability of investment opportunities (proxied by
the Market-to-book ratio) in all model specifications. Investments in net
fixed assets are, however, also driven by the availability of internally
Table 3.3 Performance, family CEOs and crisis

Panel A OLS regressions
Q ROA
I II III IV
Constant (a) 1.3020*** 1.2851***

0.0998*** 0.1061***
[0.4384] [0.4350]
[0.0290] [0.0290]
Family CEO (b) 0.1850** 0.0207***
[0.0933] [0.0060]
Crises (c) −0.2122*** −0.2131*** −0.0160*** −0.0150***
[0.0439] [0.0432] [0.0050] [0.0051]
Family CEO*Crises (d) −0.0303 0.0079
[0.0266] [0.0053]
Founder CEO (e) 0.3339* 0.0116
[0.1813] [0.0087]
Heir CEO (f) 0.1306* 0.0206***
[0.0745] [0.0067]
Founder CEO*Crises (g) −0.1418 0.0118***
[0.1011] [0.0038]
Heir CEO*Crises (h) 0.0147 0.0032
[0.0163] [0.0052]
Professional CEO (i) 0.1487** 0.1424** 0.0073* 0.0058
[0.0650] [0.0627] [0.0039] [0.0038]
Professional CEO*Crises (j) −0.0079 −0.0079 0.0024 0.0013
[0.0117] [0.0160] [0.0034] [0.0032]
Wedge −0.002 −0.0018 −0.0001 −0.0001
[0.0014] [0.0013] [0.0001] [0.0001]
Leverage −1.1445*** −1.1517*** −0.1035*** −0.1031***
[0.2518] [0.2538] [0.0125] [0.0126]
Ln(Assets) 0.0358* 0.0346* 0.0040** 0.0038*
[0.0205] [0.0201] [0.0020] [0.0020]
Ln(Age) −0.1678*** −0.1582*** −0.0105*** −0.0108***
[0.0535] [0.0517] [0.0033] [0.0033]
Dividend/BV Equity 1.2757** 1.2763** 0.0873*** 0.0876***
[0.5469] [0.5463] [0.0324] [0.0324]
Return Volatility 3.1545* 3.1309* −1.0523** −1.0496**
[1.7647] [1.7719] [0.4350] [0.4366]
Sales Growth 0.5922*** 0.5851*** 0.0629*** 0.0634***
[0.1203] [0.1171] [0.0082] [0.0081]
Adj. R2 0.1903 0.1912 0.1429 0.141
Observations 7537 7537 7507 7507
Table 3.3 (cont.)

Panel B Tests
Q ROA
I II III IV
Non-Family – Non-Crisis 1.3020*** 1.2851*** 0.0998*** 0.1061***

(a) 0.4384 0.4350 0.0290 0.0290
Family CEO Non-Crisis 1.4870*** 0.1204***
(a+b) 0.4176 0.0271
Professional CEO Non-Crisis 1.4507*** 1.4275*** 0.1071*** 0.1120***
(a+i) 0.4255 0.4222 0.0282 0.0283
Founder CEO Non-Crisis 1.6190*** 0.1177***
(a+e) 0.4220 0.0294
Heir CEO Non-Crisis 1.4157*** 0.1267***
(a+f) 0.4177 0.0271
Non-Family – Crisis 1.090** 1.0720** 0.0837*** 0.0912***
(a+c) 0.4401 0.4355 0.0282 0.0283
Family CEO Crisis 1.2446*** 0.1123***
(a+b+c+d) 0.4136 0.0295
Professional CEO Crisis 1.2306*** 1.2065*** 0.0935*** 0.0983***
(a+b+i+j) 0.4282 0.4256 0.0288 0.0289
Founder CEO Crisis 1.2640*** 0.1145***
(a+e+g+h) 0.4060 0.0311
Heir CEO Crisis 1.2172*** 0.1149***
(a+e+i+j) 0.4158 0.0296

Q ROA
I II I II
Non-Crisis Family CEO 3.93** 12.05***

vs. Non-family 0.0474 0.0005
Non-Crisis Professsional 5.23** 3.57*
CEO vs. Non-family 0.0222 0.0587
Non-Crisis Family CEO 0.28 7.07***
vs. Professional CEO 0.5949 0.0079
Non-Crisis Founder 3.39* 1.78
Non-Crisis Heir CEO vs. 3.07* 9.56***
Non-family 0.0799 0.002
Table 3.3 (cont)

Non-Crisis Professional 5.16** 2.4
Non-Crisis Founder 1.64 0.82
CEO vs. Heir CEO 0.2 0.366
Non-Crisis Founder 1.45 0.51
CEO vs. Professional
CEO 0.2284 0.4755
Non-Crisis Heir CEO vs. 0.03 5.86**
Professional CEO 0.852 0.0156
Crisis Family CEO vs. 3.87** 16.85***
Non-Family 0.049 0
Crisis Professional CEO 6.87*** 3.93**
vs. Non-family 0.0088 0.0475
Crisis Family. CEO vs. 0.07 11.74***
Crisis Founder CEO vs. 2.7 4.68**
Non-family 0.1004 0.0306
Crisis Heir CEO vs. 4.13** 11.97***
Non-family 0.0421 0.0005
Crisis Professional CEO 7.16*** 2.32
vs. Non-family 0.0076 0.1274
Crisis Founder CEO vs. 0.19 0
Heir CEO 0.6596 0.9715
Crisis Founder CEO vs. 0.26 2.33
Crisis Heir CEO vs. 0.05 8.78***
Note: The table presents the results of OLS regressions with double-clustered
standard errors where the dependent variable is the Q Ratio in Columns I and II
and ROA in Columns III and IV. Q Ratio is defined as the sum of total assets
(Worldscope Item WC02999) and market value of equity (WC08001) minus
common shareholders’ equity (WC03501) scaled by total assets (WC02999). ROA
is the ROA, defined as EBITDA over total assets (WC18198/WC02999). Family
CEO is a dummy variable that takes value 1 if the firm is classified as a family firm
and a family member is CEO. A firm is defined as a family (non-family) firm if its
ultimate owner is (is not) a family member. Founder (Heir) CEO is a dummy
variable that takes value 1 if the founder (heir) is the family firm’s CEO.
Professional CEO is a variable that takes value 1 if the family firm CEO is a
professional manager unrelated to the controlling family. Crises is a dummy
variable that takes value 1 in years 2001–03 and 2008–10. Dotcom Crisis (Credit
Crisis) is a dummy variable that takes value 1 in years 2001–03 (2008–10). Wedge
is the divergence between voting rights and cash flows right of the (ultimate)
controlling shareholder. Leverage is the ratio of the book value of financial debt as
a percentage of the book value of total assets (WC03255/WC02999). Ln(Assets) is
the logarithm of the firm’s total assets (WC02999) in constant 2005 Euros. Ln
(Age) is the log of the firm’s age. Dividend/BV Equity is the ratio between cash
common dividends (WC05376) and the book value of common equity
(WC03501). Return Volatility is the standard deviation of daily stock returns over
the year. Sales Growth is the one-year growth rate in sales. All values are in Euros.
All regressions include industry and country fixed effects. Industry classification is
based on the Fama and French 12-industry classification. Firm and time-(double-)
clustered standard errors are in brackets and statistical significance is denoted by
***, **, and * for the 1%, 5% and 10% levels, respectively. Panel B reports the
coefficients for the possible combinations of family variables (Family CEO,
Founder CEO, Heir CEO, Professional CEO) and Crises (Dotcom Crisis/Credit
Crisis), obtained as a linear combination of the coefficients in Panel A. Standard
errors are reported below the coefficients. Panel C reports the results of F-tests,
with their p-value, for the tests for differences between family and non-family in
the different periods (non-crisis; crisis; dotcom crisis; credit crisis).
generated funds (measured by CF/K and cash holdings), which should

not happen in frictionless capital markets. Also, older firms seem to
invest significantly less.
Surprisingly, Table 3.4 does not document lower investments during
crisis periods: while the coefficient has the expected negative sign, the
estimates are not statistically significant. Regarding the main variables of
interest, results show that family firms generally invest more. This finding is
statistically significant at the 0.01 level in specifications (I) and (II).
Interacting the family firm dummy with the indicator variables for crisis
periods further reveals that family firms invest significantly less during
crises. The relevant interaction terms are negative and statistically signifi-
cant (at the 0.05 and 0.01 level, respectively) in model specifications (I) and
(II) and seem to be driven by a strong decrease in investments during the
credit crisis. Panels B and C support the results found in Panel A, showing a
significantly higher coefficient during the non-crisis period and more
accentuated decrease because of the crises for family firms. Overall, these
results allow for two possible interpretations: given that family firms gen-
erally invest more, the significant decrease in investments during crises can
be regarded as evidence for efficient investment behaviour as they seem to
Table 3.4 Investments, downsizing and increase in size
Panel A Regressions
I/K Downsizing Increase in Size
I II III IV V VI
Constant (a) 0.3634*** 0.3571*** −0.7533 −0.8072 −0.1308 −0.0301

[0.1190] [0.1188] [0.8322] [0.8760] [0.5400] [0.5632]
Family (b) 0.0361*** 0.0362*** −0.2439** −0.2429** 0.1574*** 0.1548***
[0.0096] [0.0096] [0.1042] [0.1046] [0.0500] [0.0504]
Crises (c) −0.0208 0.7622** −0.7698**
[0.0141] [0.3077] [0.3460]
Dotcom Crisis (d) −0.0215 0.9538*** −0.9545**
[0.0145] [0.3526] [0.4107]
Credit Crisis (e) −0.0202 0.4565 −0.5325
[0.0246] [0.4511] [0.4474]
Family*Crises (f) −0.0226** 0.2763* −0.0891
[0.0090] [0.1598] [0.1153]
Family*Dotcom Crisis (g) −0.0103 0.108 −0.122
[0.0080] [0.1403] [0.1736]
Family*Credit Crisis (h) −0.0384*** 0.5473*** −0.0705
[0.0117] [0.1474] [0.0927]
CF/K 0.0093* 0.0093* −0.0757 −0.0744 0.0379* 0.0356*
[0.0048] [0.0048] [0.0931] [0.0905] [0.0197] [0.0194]
Market-to-book 0.0054** 0.0054* 0.0006 0.0002 0.0434 0.0448
[0.0028] [0.0028] [0.0206] [0.0205] [0.0270] [0.0275]
Cash Holding 0.1157* 0.1162* 0.6972** 0.7494** 0.9733*** 0.9400***
[0.0608] [0.0615] [0.3500] [0.3332] [0.3328] [0.3337]
Leverage −0.03 -0.0301 0.4833 0.4603 0.7220** 0.7303**
[0.0350] [0.0348] [0.4220] [0.4159] [0.3010] [0.2881]
Lagged Sales 0.0086 0.0085 −0.1541 -0.1609 0.1964*** 0.2013***
[0.0100] [0.0100] [0.0989] [0.1001] [0.0549] [0.0530]
Ln(Assets) 0.0003 0.0006
[0.0042] [0.0043]
Ln(Lagged Assets) −0.0644 −0.0627 −0.0267 −0.0319
[0.0503] [0.0529] [0.0423] [0.0425]
Ln(Age) −0.0188** −0.0183** −0.0378 −0.0308 −0.0628 −0.072
[0.0089] [0.0089] [0.0886] [0.0831] [0.0524] [0.0510]
Industry/Country FE Yes Yes Yes Yes Yes Yes
Adj. R2/Pseudo R2 0.0993 0.0998 0.0533 0.0506 0.0595 0.0577
Observations 7,470 7,470 7,500 7,500 7,500 7,500
Panel B Tests
I II III IV V VI
Non-family/Non-Crisis 0.3634*** 0.3571*** −0.7533 −0.8072 −0.1308 −0.0301

(a) 0.1190 0.1188 0.8322 0.8760 0.5400 0.5632
Family/Non-Crisis 0.3995*** 0.3933*** −0.9972 −1.0501 0.0266 0.1247
(a+b) 0.1190 0.1188 0.8084 0.8494 0.5565 0.5784
Non-family/Crises 0.3426*** 0.0088 −0.9007
(a+c) 0.1208 0.8451 0.6887
Family/Crises 0.3561*** 0.0412 −0.8324
(a+b+c+f) 0.1208 0.8725 0.6578
Non-Family/Dotcom Crisis 0.3356*** 0.1465 -0.9846
(a+d) 0.1204 0.8854 0.7282
Non-family/Credit Crisis 0.3369*** −0.3508 −0.5626
(a+e) 0.1234 1.1241 0.8164
Family/Dotcom Crisis 0.3616*** 0.0116 −0.9518
(a+b+d+g) 0.1202 0.8533 0.6443
Family/Credit Crisis 0.3347*** −0.0463 −0.4784
(a+b+e+h) 0.1187 1.0893 0.8479

I II III I II III
Non-Crisis 14.19*** 14.20*** 5.47*** 5.39** 9.91*** 9.44***

0.0002 0.0002 0.0193 0.0203 0.0016 0.0021
Crises 1.66 0.08 0.37
0.198 0.781 0.5449
Dotcom Crisis 5.84** 1.91 0.04
0.0157 0.2517 0.8478
Credit Crisis 0.03 5.29*** 0.91
0.8676 0.0214 0.3411
Note to Table 3.4 (cont.):
The table presents the results of OLS and logit regressions with double-clustered standard errors where the dependent variable is the
firm’s investments. Firms’ investments are measured with the variables I/K (Columns I and II); Downsizing (Columns III and IV)
and Increase in Size (Columns V and VI). I/K is the ratio between investments in fixed assets (capital expenditures, WC04601) and
lagged net fixed assets (WC02501); Downsizing (Increase in Size) is a dummy variable that takes value 1 if the firm’s total assets
decrease (increase) by at least 10% with respect to the previous year. Family is a dummy variable that takes value 1 if the firm is
classified as a family firm. Family is a dummy variable that takes value 1 if the firm is classified as a family firm. A firm is defined as a
family (non-family) firm if its ultimate owner is (is not) a family member. Crises is a dummy variable that takes value 1 in years
2001–03 and 2008–10. Dotcom Crisis (Credit Crisis) is a dummy variable that takes value 1 in years 2001–03 (2008–10). CF/K is the
ratio between the firm’s cash flows (operating income (WC01250) plus depreciation (WC01151)) scaled by lagged net fixed assets.
Market-to-book is the ratio of the market value of equity (WC08001) to common equity in (WC03501). Cash Holding is the ratio of
cash plus tradable securities to total assets (WC02001/WC02999). Leverage is the ratio of the book value of financial debt as a
percentage of the book value of total assets (WC03255/WC02999). Lagged Sales is the firm’s sales (WC01001) in the previous year in
constant 2005 Euros. Ln(Assets/Lagged Assets) is the logarithm of the firm’s total assets (WC02999) in constant 2005 Euros. Ln(Age)
is the log of the firm’s age. All values are in Euros. All regressions include industry and country fixed effects. Industry classification is
based on the Fama and French 12-industry classification. Firm and time-(double-) clustered standard errors are in brackets and
statistical significance is denoted by ***, **, and * for the 1%, 5% and 10% levels, respectively. Panel B reports the coefficients for the
possible combinations of Family and Crises (Dotcom Crisis/Credit Crisis), obtained as a linear combination of the coefficients in
Panel A. Standard errors are reported below the coefficients. Panel C reports the results of F-tests, with their p-value, for the tests for
differences between family and non-family in the different periods (non-crisis; crisis; dotcom crisis; credit crisis).
quickly adjust their investments to changes in the investment opportunity

set. This would imply better investment decisions and could serve as an
explanation for the superior performance of family firms documented in
Tables 3.2 and 3.3 and the empirical corporate finance literature. On the
other hand, it seems plausible that the change in investment policy is
affected by the availability of external funding. As family firms are often
reluctant to issue new equity, the availability of bank loans might affect their
investments. The decrease in investment outlays during crises could thus
also be driven by the reluctance of creditors to finance new investments in
periods of uncertainty. The decrease in investments for family firms, being
primarily driven by the recent credit crisis, supports this interpretation.
The regressions in columns III to VI of Table 3.4 examine investments
from a different angle by looking at changes in the level of total assets.
Specifically, we analyse whether comparatively large changes in total
assets are systematically related to the set of explanatory variables in
several logit regressions. In model specifications (III) and (IV) the
dependent variable captures a downsizing effect and is modelled as an
indicator variable that is set to one if the firm’s total assets decrease by
more than 10 per cent; specifications (V) and (VI) examine the opposite,
i.e. an increase in total assets by more than 10 per cent. The results
corroborate our findings in Columns I and II and show higher invest-
ments in family firms, as evidenced by negative (positive) and significant
coefficients in the downsizing (increase in size) regressions. Again, firms
seem to invest more in periods of economic recovery. Models (III) and
(IV) of Table 3.4 go beyond our previous results and show that firms not
only cut back on their investments during economic crises, but even
decrease firm size in these periods. In line with the investment analysis in
Column I and II, findings reveal that this effect is more pronounced for
family firms. Family businesses are significantly more likely to divest and
reduce their firm size in crisis periods, a finding that again seems to be
driven by the recent credit crisis.
In sum, our analysis of investment policy indicates that the investment
outlays of family firms are more dependent on the overall economic
climate. They seem to invest less in periods of economic crisis, and more
during economic recoveries.
The final part of the analysis deals with the question of how economic
crises affect employment in family and non-family firms. Table 3.5
approaches this question on two dimensions: model specifications (I)
and (II) are OLS regressions (with double-clustered standard errors)
with the log of average per capita salaries (the sum of salaries and benefit
Table 3.5 Crises, wages and employment

Panel A OLS regressions
Ln(Wages) Change in Employees

I II III IV
Constant (a) 4.2068*** 4.2018*** −0.0978 −0.1001

[0.1782] [0.1749] [0.0701] [0.0686]
Family (b) −0.0771*** −0.0770*** 0.0313*** 0.0314***
[0.0270] [0.0270] [0.0075] [0.0075]
Crises (c) 0.0319 −0.0385***
[0.0197] [0.0138]
Dotcom Crisis (d) 0.0179 −0.0395**
[0.0150] [0.0169]
Credit Crisis (e) 0.0509* −0.0373**
[0.0264] [0.0166]
Family*Crises (f) −0.0061 −0.0245**
[0.0113] [0.0117]
Family*Dotcom Crisis (g) 0.0155 −0.0199
. [0.0145]
Family*Credit Crisis (h) −0.0349* −0.0307***
[0.0182] [0.0106]
Wedge 0.0003 0.0003 0.0002 0.0002
[0.0009] [0.0009] [0.0002] [0.0002]
Leverage −0.2082** −0.2072** 0.0713** 0.0711**
[0.0838] [0.0838] [0.0357] [0.0361]
Ln(Assets) −0.0137 −0.0136 0.0109*** 0.0110***
[0.0087] [0.0087] [0.0025] [0.0025]
Ln(Age) 0.0227 0.0228 −0.0228** −0.0227***
[0.0231] [0.0233] [0.0088] [0.0088]
Return Volatility 0.4315 0.4292 −0.1468 −0.1408
[1.0051] [1.0064] [0.3200] [0.3180]
Lagged ROA −0.3269*** −0.3289*** 0.3484*** 0.3475***
[0.1182] [0.1178] [0.0750] [0.0755]
Adj. R2 0.1772 0.1772 0.0471 0.0469
Observations 7335 7335 7476 7476
Table 3.5 (cont)

Panel B Tests

I II III IV
Non-family/Non-Crisis 4.2068*** 4.2018*** −0.0978 −0.1001
(a) 0.1782 0.1749 0.0701 0.0686
Family/Non-Crisis 4.1230*** 4.1248*** −0.0665 −0.0687
(a+b) 0.179176 0.175821 0.0705 0.0691
Non-family/Crises 4.239*** −0.1363**
(a+c) 0.1853 0.0685
Family/Crises 4.1555*** −0.1296*
(a+b+c+f) 0.1809 0.0667
Non-Family/Dotcom Crisis 4.2197*** −0.1396**
(a+d) 0.17901 0.07018
Non-family/Credit Crisis 4.2527*** −0.1374**
(a+e) 0.1818 0.0639
Family/Dotcom Crisis 4.158*** −0.1280*
(a+b+d+g) 0.1780 0.0663
Family/Credit Crisis 4.1408*** −0.1366***
(a+b+e+h) 0.1798 0.0650

I II I II
Non-Crisis 7.35*** 0.65
0.0067 0.4198
Crises 8.16*** 8.11*** 17.37*** 17.52***
0.0043 0.0044 0 0
Dotcom Crisis 4.48** 0.95
0.0344 0.3296
Credit Crisis 11.62*** 0.01
0.0007 0.91034
Note: The table presents the results of OLS regressions with double clustered
standard errors where the dependent variable is the log of average wage in
Columns I and II and the percentage change in employment in Columns III and
IV. Ln(Wages) is defined as the log of the ratio between Salaries and benefits
expenses (Worldscope Item WC01084) and the number of employees
(WC07011). The change in employees is the percentage change in employees
between year t-1 and year t. ROA is the ROA, defined as EBITDA over total assets
(WC18198/WC02999). Family is a dummy variable that takes value 1 if the firm is
classified as a family firm. A firm is defined as a family (non-family) firm if its
ultimate owner is (is not) a family member. Crises is a dummy variable that takes
value 1 in years 2001–3 and 2008–10. Dotcom Crisis (Credit Crisis) is a dummy
variable that takes value 1 in years 2001–3 (2008–10). Wedge is a dummy variable
that takes value 1 in case of divergence between voting rights and cash flows right
of the controlling shareholder. Leverage is the ratio of the book value of financial
debt as a percentage of the book value of total assets (WC03255/WC02999). Ln
(Assets) is the logarithm of the firm’s total assets (WC02999). Ln(Age) is the log of
the firm’s age. Return Volatility is the standard deviation of daily stock returns
over the year. All values are in Euros. All regressions include industry and country
fixed effects. Industry classification is based on the Fama and French 12-industry
classification. Firm and time-(double-) clustered standard errors are in brackets
and statistical significance is denoted by ***, **, and * for the 1%, 5% and 10%
levels, respectively. Panel B reports the coefficients for the possible combinations
of Family and Crises (Dotcom Crisis/Credit Crisis), obtained as a linear
combination of the coefficients in Panel A. Standard errors are reported below the
coefficients. Panel C reports the results of F-tests, with their p-value, for the tests
for differences between family and non-family in the different periods (non-crisis;
crisis; dotcom crisis; credit crisis).
expenses over the number of employees) as the dependent variable, while

specifications (III) and (IV) use the percentage change in the number of
employees as the dependent variable. In line with Sraer and Thesmar
(2007), results show significantly lower pay levels in family firms.
Surprisingly, average salaries do not decrease during crises, probably
because of labour market rigidities that characterise Continental Europe.
However, this finding needs to be considered in conjunction with the
results in specifications (III) and (IV), which analyse the determinants of
the change in the number of employees. These regressions indicate that
firms reduce their workforce in times of crises.15 In contrast to our
hypothesis, results show that family firms are more aggressive in reduc-
ing their workforce during crises and also pay significantly lower wages
15
If pay levels remain constant and staff is reduced, average per capita salaries do not
necessarily increase, as these numbers also incorporate severance pay and other one-off
payments that are a side-product of the reduction in staff.
as a result of the credit crisis. A possible explanation might be our earlier

finding that family firms are also more likely to downsize in crisis
periods. Panel B provides additional support to the evidence in Panel
A, showing lower coefficients for family firms in both crisis and non-
crisis periods. The difference between these coefficients is significant, as
reported in Panel C.
6. Policy implications and conclusions

This chapter analyses how family-controlled firms respond to economic
and financial crises, focusing on two major crises: the dotcom bubble
crisis and the credit crisis. Our results confirm that family firms generally
outperform non-family firms in Europe. These findings contrast with the
private benefits hypothesis, i.e. the dark side of family control. More
importantly, this evidence suggests that family firms invest less and are
more likely to downsize in crisis periods than non-family firms. We
interpret this result as evidence that family firms are more efficient than
non-family firms, as they adjust their investment decisions to changes in
the investment opportunity set more quickly than their non-family
counterparts. However, this sudden change also highlights the fact that
family firms and, more generally, small listed companies, may face major
obstacles when they need fresh capital to finance investments or even
daily operations during a crisis. Our evidence suggests that this effect is
not generalised to all crises, but tends to be more relevant in severe
recessions like the credit crisis.
In many Continental European countries, small listed companies,
whose majority is represented by family firms, primarily rely on bank
debt. This source of financing is particularly sensitive to credit market
conditions and it can dry up during severe crises such as happened in the
recent global financial crisis. Difficulties in obtaining and renewing bank
loans may lead family firms to cut investments and lay off employees, as
shown by our evidence. Since families are known to be reluctant issuers
of equity because of the fear of diluting their ownership stakes in the
firm, the excessive reliance on bank debt may be detrimental to them,
especially in crisis periods. Crisis periods are also the periods when the
reluctance to issue equity is generally stronger. In fact, because of low
stock prices, the dilution for the controlling shareholder would be even
greater. The development of a public bond market for small firms may
provide an answer to this problem. The European Commission’s 2011
Green Paper stresses the importance of establishing a differentiated and
proportionate corporate governance regime for small and medium-size

listed companies. Our evidence suggests that this type of consideration
should not be limited to corporate governance codes; rather, it should be
applied also to provide a better (and more differentiated) access to
financing. The corporate bond markets is not commonly used by small
firms in continental Europe (Allen et al. 2004), and the creation of a bond
market dedicated to the issues of small and medium-size listed firms may
help small and relatively unknown companies sell their bonds to the
public.
This analysis sheds light on important implications of the ownership
structure of Continental European companies, which exhibit a large
percentage of family firms (Faccio and Lang 2002). The 2011 Green
Paper advocates a larger shareholder engagement. The engagement on
the part of long-term investors is particularly encouraged, because they
have incentives to avoid excessive risk-taking behaviours that can put the
company in danger. Controlling families can be considered part of this
category of long-term investors. Our analysis shows that this recipe,
while theoretically sound, may not produce the desired outcomes in
terms of more stable performance and investments. In fact, the evidence
suggests that family firms behave pro-cyclically during crises. Therefore,
countries where family firms are more common are not expected to
experience milder crises because of the counter-cyclical behaviour of
family firms. To put it another way, stability cannot be achieved with
fine-tuning of firms’ ownership structures. Policies aimed at favouring
the presence of large shareholders to increase engagement with this goal
in mind may backfire.
Although family firms are less affected by short-termism and market
pressures than non-family firms with diffuse ownership, evidence also
suggest that it could be dangerous to encourage family control only to
improve economic stability, especially during downturns. While family
control is, indeed, stable as many academic articles show (for example, see
Franks et al. 2011), this does not imply that their investment policies are less
sensitive to the business cycle than those of non-family firms. Conversely,
family firms may exacerbate the crisis in the short term because they tend
less gradually to adjust their investment policies. While this can have a
negative effect in the short term, it could produce benefits in the long term
because the crisis allows family firms to dismiss unproductive assets as the
overall better performance of family firms seems to indicate. The stability of
control in family firms may well be the reason behind their investment
policies. To preserve control and fearing dilution, family firms could avoid
issuing equity in time of crisis to finance new investments, which may lead
them to rely more on bank debt. Loans can be more difficult to obtain
during economic crises, in particular during the recent crisis, which may
lead family firms to allocate their capital to the best possible uses, i.e.
investment with higher returns.
We also find evidence that family firms used the recent credit crisis to
reduce their workforce and wages. Since family firms already pay lower
wages than non-family firms, and they further reduce wages during
credit crises, the reduction in the number of employees could signal
the break-up of long-term implicit contracts with their employees. These
contracts are based on the trade-off between low wages and higher job
security, which could be detrimental to firms’ long-term growth and
profits. These findings may point to a wealth transfer from labour to
shareholders during crises. With fewer outside options available, work-
ers may be willing to accept a downward revision in their wages to
preserve their jobs. However, our evidence also highlights the fact that
this redistribution does not take place in all crises: family firms did not
reduce wages and workforce during the dotcom crisis, i.e. the first crisis
we examine, but only during the most severe recession that followed the
credit crisis. These quick adjustments would be more difficult to carry
out if employees owned a significant fraction of the equity capital.
Employee share ownership, which, as also mentioned in the 2011
Green Paper, has a long tradition in some European countries, may be
a double-edged sword: on the one side, it would probably lead to a
smoother transition during crises; on the other side, it would prevent
efficient restructuring of the firms.16
Finally, our results suggest that family firms do not use periods of
crisis to expropriate minority investors. In fact, the decrease in the
valuation of family firms during crisis is similar to that of non-family
firms, which is a sign that markets do not believe that the fewer invest-
ment opportunities will lead the controlling family to increase the
expropriation of minority investors. On the other hand, the analysis
based on investment suggests that family controlling shareholders do
not prop up their companies with cash injections either. In fact, family
firms tend to shed assets more than non-family firms during crises. Thus,
more than propping up during crises, the overall better performance of
16
Another related issue with employee share ownership is that it does not allow employees
to diversify: both their savings and their jobs depend on one firm.
family firms in the long run is the result of the more efficient investment
policy adopted by family firms.
Results showing that family firms outperform non-family firms in
good and bad times have another important consequence at policy
level. The policy debate has focused primarily on the improvement of
investor protection, in particular where a conflict of interest exists
between controlling shareholders and minority investors. The 2011
Green Paper of the European Commission clearly states that ‘Minority
shareholder engagement is difficult in companies with controlling
shareholders’, and it worries about rights to represent their interests
effectively in companies with a controlling shareholder. Overall, the
results show that families protect their fellow shareholders quite well,
delivering superior stock and accounting performance. While this does
not imply that minority shareholder protection is a second-order prob-
lem as financial scandals like Parmalat SpA in 2003 show,17 it signals
that, on average, this problem is to some extent overstated. Proposed
remedies like reserving the appointment of some board seats to minority
shareholders, as currently done in Italy, could increase the information
that flows to minority shareholders, but there is no guarantee that they
will generate better performance. Moreover, it is highly questionable that
one (or few) director(s) appointed by minority shareholders will really
affect the firm’s decision when the controlling shareholder has a majority
in the board. It would be advisable to shift attention from pure govern-
ance problems to better access to financing and growth.
The evidence presented in this chapter has important implications for
the evolution of the regulation of corporate governance in EU countries.
Overall, our work provides evidence about the desirability of family
control in listed companies in Europe that should increase our under-
standing of the costs and benefits of corporate governance proposals.
References
Allen, F., M.K.F. Chui and A. Maddaloni 2004. ‘Financial systems in Europe, the
USA, and Asia’, Oxford Review of Economic Policy 20: 490–508.
Almeida, H., M. Campello and M.S. Weisbach 2004. ‘The cash flow sensitivity of
cash’, Journal of Finance 59: 1777–804.
17
At the time Parmalat was a family firm controlled by the Tanzi family with more than 50
per cent of the voting capital.
Anderson, R.C. and D.M. Reeb 2003. ‘Founding-family ownership and firm
performance: Evidence from the S&P 500’, Journal of Finance 58: 1301–28.
Andres, C. 2008. ‘Large shareholders and firm performance: An empirical exami-
nation of founding-family ownership’, Journal of Corporate Finance 14:
431–45.
2011. ‘Family ownership, financing constraints and investment decisions’,
Applied Financial Economics 21: 1641–59.
Astrachan, J.H. and I.E. Allen 2003. MassMutual/Raymond Institute American
Family Business Survey, Family Business Review.
Bae, K.H., J.S. Baek, J.K. Kang and W.L. Liu 2012. ‘Do controlling shareholders’
expropriation incentives imply a link between corporate governance and
firm value? Theory and evidence’, Journal of Financial Economics, forth-
coming, doi:10.1016/j.jfineco.2012.02.007.
Baek, J.S., J.K. Kang and K.S. Park 2004. ‘Corporate governance and firm value:
evidence from the Korean financial crisis’, Journal of Financial Economics
71: 265–313.
Barclay, M.J. and C.G. Holderness 1989. ‘Private benefits from control of public
Barontini, R. and L. Caprio 2006. ‘The effect of family control on firm value and
performance: Evidence from Continental Europe’, European Financial
Bauguess, S. and T. Stegemoller 2008. ‘Protective governance choices and the
value of acquisition activity’. Journal of Corporate Finance 14: 550–66.
Bennedsen, M., K.M. Meiner Nielsen, F. Perez-Gonzalez and D. Wolfenzon 2007.
‘Inside the family firm: The role of families in succession decisions and
performance’, Quarterly Journal of Economics 122: 647–91.
Berle, A.A. and G.C. Means 1932. The Modern Corporation and Private Property.
Harcourt, Brace & World, New York.
Bertrand, M. and A. Schoar 2006. ‘The role of family in family firms’. Journal of
Economic Perspective 20: 73–96.
Caprio, L., E. Croci and A. Del Giudice 2011. ‘Ownership structure, family
control, and acquisition decisions’, Journal of Corporate Finance 17:
1636–57.
Claessens, S., S. Djankov, J.P.H. Fan and L.H.P. Lang 2002. ‘Disentangling the
incentive and entrenchment effects of large shareholdings’, Journal of
Finance 57: 2741–71.
Cleary, S. 1999. ‘The relationship between firm investment and financial status’,
Journal of Finance 54: 673–91.
2006. ‘International corporate investment and the relationships between finan-
cial constraint measures’, Journal of Banking and Finance 30: 1559–80.
Cronqvist, H. and R. Fahlenbrach 2009. ‘Large shareholders and corporate
policies’, Review of Financial Studies 22: 3941–76.
Cronqvist, H. and M. Nilsson 2003. ‘Agency costs of controlling minority
shareholders’, Journal of Financial and Quantitative Analysis 38:
695–719.
Dahya, J., O. Dimitrov and J. McConnell 2008. ‘Dominant shareholders, corporate
boards, and corporate value: A cross-country analysis’, Journal of Financial
European Commission 2011. ‘The EU corporate governance framework’, Green
Paper COM(2011) 164.
Faccio, M. and L.H.P. Lang 2002. ‘The ultimate ownership of Western European
and L. Young 2001. ‘Dividends and expropriation’, American Economic Review
91: 54–78.
Fazzari, S.M., G. Hubbart and B.C. Petersen 1988. ‘Financing constraints and
corporate investment’, Brookings Papers on Economic Activity 1: 141–206.
Franks, J. and C. Mayer 2001. ‘Ownership and control of German corporations’,
P. Volpin and H.F. Wagner 2011. ‘The life cycle of family ownership: A
comparative study of France, Germany, Italy and the U.K.’, Review of
Financial Studies, forthcoming.
Friedman, E., S. Johnson and T. Mitton 2003. ‘Propping and tunneling’, Journal of
Comparative Economics 31: 732–50.
Gonenc, H. and N. Hermes 2008. ‘Propping: Evidence from new share issues of
Turkish business group firms’, Journal of Multinational Financial
Hennessy, C. A. 2004. ‘Tobin’s q, debt overhang, and investment’, Journal of
Finance 59: 1717–42.
Jensen, M. C. 1986. ‘The agency costs of free cash flow: Corporate finance and
takeovers,’ American Economic Review 76: 323–9.
Jenkinson, T. and A. Ljungqvist 2001. Going Public: The Theory and Evidence on
how Companies raise Equity Finance, 2nd edn, Oxford University Press.
Johnson, S., P. Boone, A. Breach and E. Friedman 2000. ‘Corporate governance in
the Asian financial crisis 1997–98’, Journal of Financial Economics 58:
141–86.
Kaplan, S.N. and L. Zingales 1997. ‘Do investment–cash flow sensitivities provide
useful measures of financing constraints?’, The Quarterly Journal of
Economics 112: 169–215.
Khanna, T. and K. Palepu 2000. ‘Is group affiliation profitable in emerging
markets? An analysis of diversified Indian business groups’, Journal of
Finance 55: 867–91.
Klasa, S. 2007. ‘Why do controlling families of public firms sell their remaining
ownership stake?’, Journal of Financial and Quantitative Analysis 42:
339–68.
Laeven, L. and R. Levine 2009. ‘Bank governance, regulation and risk taking’,
Lee, J. 2006. ‘Family firm performance: Further evidence’, Family Business Review
19: 103–14.
Masulis, R., P. Pham and J. Zein 2011. ‘Family business groups around the world:
Costs and benefits of pyramids’, Review of Financial Studies, forthcoming,
doi:10.1093/rfs/hhr052.
Maury, B. 2006. ‘Family ownership and firm performance: Empirical evidence
from Western European corporations’, Journal of Corporate Finance 12:
321–41.
Miller, D., I. Le Breton-Miller, R.H. Lester and A.A. Cannella, Jr. 2007. ‘Are family
firms really superior performers?’, Journal of Corporate Finance 13: 829–58.
Mitton, T. 2002. ‘Across-firm analysis of the impact of corporate governance on
the East Asian financial crisis’, Journal of Financial Economics 64: 215–41.
Morck, R. and B. Yeung 2004. ‘Family control and the rent-seeking society’,
Entrepreneurship: Theory and Practice 28: 391–409.
Petersen, M. 2009. ‘Estimating standard errors in finance panel data sets: compar-
ing approaches’, Review of Financial Studies 22: 435–80.
Pindado, J., I. Requejo and C. de la Torre 2011. ‘Family control and investment–
cash flow sensitivity: Empirical evidence from the Euro zone’, Journal of
Corporate Finance 17: 1389–409.
Sraer, D. and D. Thesmar 2007. ‘Performance and behavior of family firms:
Evidence from the French stock market’, Journal of the European
Economic Association 5: 709–51.
Stein, J. 1989. ‘Efficient capital markets, inefficient firms: A model of myopic
corporate behavior’, Quarterly Journal of Economics 103: 655–69.
Thompson, S. 2011. ‘Simple formulas for standard errors that cluster by both firm
and time’, Journal of Financial Economics 99: 1–10.
Villalonga, B. and R. Amit 2006. ‘How do family ownership, control and manage-
ment affect firm value?’, Journal of Financial Economics 80: 385–417.
Williamson, O. E. 1979. ‘Transaction-cost economics: The governance of con-
tractual relations’, Journal of Law and Economics 22: 233–61.
Zhou, Y.M., X. Li and J. Svejnar 2011. ‘Subsidiary divestiture and acquisition in a
financial crisis: Operational focus, financial constraints, and ownership’,
Journal of Corporate Finance 17: 272–87.
4
Corporate boards in Europe: size, independence

and gender diversity
daniel ferreira and tom kirchmaier
1. Introduction
The board of directors is one of the most important governance mech-
anisms in modern corporations. In principle, the board is responsible for
approving major strategic and financial decisions. It has access to priv-
ileged and timely information about the firm, meets regularly to discuss
this information and has a fiduciary duty towards the shareholders it
represents. The role of the board is to advise and monitor management,
and for that purpose the board is typically staffed with distinguished
individuals who have the required skills to fulfil this role. As Adams and
Ferreira (2007) point out, the degree to which a board can fulfil its
function also depends on the quality of information provided by
management.
Some observers believe that the board is the first line of defence in
corporate governance. The importance of corporate boards is reflected in
an ample academic literature on this topic and in the regulatory focus
that they attract. For reasons of data availability and comparability,
much of the academic literature on boards concentrates on US firms.
This chapter provides comparable board data for many European
countries.
It is the first study to provide a comprehensive analysis of the deter-
minants of board structure variables in European countries. The chapter
focuses on the determinants of board characteristics, rather than on the
consequences of these characteristics for firm policies and performance.
Thus, this study fits into the literature that shows that the composition
of boards is related to a number of firm characteristics such as size,
growth opportunities, leverage and proxies for information asymmetry,
amongst others (Boone et al. 2007; Coles et al. 2008; Linck, Netter and
191
192 d aniel ferreira and tom kirchmaier
Yang 2008; Lehn et al. 2009; Ferreira et al. 2011). However, this literature
focuses almost exclusively on the boards of US firms.
Understanding the variation in board structure across European
countries is important, because many of the regulatory proposals that
have emerged since the crisis aim at reforming European boards. The
most ambitious proposals have been directed at reforming the boards of
financial firms (Kirkpatrick 2009; Walker 2009; European Commission
2010). But there have also been more general trends toward the regu-
lation of board composition in non-financial firms. In particular, pro-
posals that aim to give women better representation on corporate boards
have recently gained momentum. For example, the Davies Report rec-
ommends that all FTSE 100 boards should aim for a minimum of 25 per
cent female representation by 2015, and also that the UK Corporate
Governance Code should be amended to introduce an explicit policy
concerning board diversity (Davies 2011). Recently, countries such as
Spain and France have introduced legislation with explicit quotas for
female directors on corporate boards.
This chapter examines some of the determinants of board size, direc-
tor independence and board gender diversity in 28 European countries
(22 of which are from the EU). Sample data include roughly 2,600
European firms from many different sectors in 2010. Part of the collected
data date back to 2000, which allows us to provide a thorough descrip-
tion of the evolution of these variables in the pre-crisis and post-crisis
periods. Similar data are available on 4,014 US firms, which allow us to
compare the evolution of these variables in Europe with that of US firms.
In sum, this study provides the most comprehensive analysis to date of
these three board characteristics in European countries.
The chapter first describes the aggregate time trends in EU countries and
then compares them to the trends in the US. The findings show that board
size, on average, has been declining in both EU and US firms. The average
board size in both the EU and the US was about 8.5 directors in 2010. Most
of the reduction in the average board size is explained by composition
effects: young firms enter the dataset more often than old firms, and the
former tend to have smaller boards. Once this is taken into account, the
change in board size from 2000 to 2010 is not very remarkable.
Quite a different situation is found when looking at the board inde-
pendence data. Board independence (the proportion of directors who are
classified as independent non-executive directors) has been increasing
both in the EU and in the US, but the levels of independence are much
higher in the US (74 per cent) than in the EU (34 per cent). Results also
corporate boards in europe 193
show that although board gender diversity (the proportion of female

directors on the board) has been on the rise everywhere, the average
female representation in EU boards is still quite low (8 per cent) and not
much different from that of US companies.
The chapter next analyses the cross-section of board characteristics in
European firms after the crisis (our benchmark year is 2010). The
analysis shows that variation in board size is reasonably well explained
by basic firm characteristics, such as firm size and industry classification.
In contrast, both board independence and board gender diversity are
better explained by country characteristics, a finding that suggests an
important role for regulation in shaping those variables. Governance
regulation varies across Europe at the country level, whilst the various
European countries vary considerably in terms of size, income and
political and economic history. The economic differences are most
prominent in the former communist countries towards the east, many
of which joined the EU as recently as 2007.
The chapter further investigates the determinants of the changes in
these board characteristics in the post-crisis period (2007 to 2010). The
most interesting results are those that indicate that small and poorly-
performing firms in 2007 tended to decrease both board size and board
independence. Such changes are consistent with poorly performing firms
changing their boards to increase focus and improve their expertise.
Our findings are consistent with some of the existing evidence in the
international corporate governance literature, such as the finding that
most of the cross-sectional variation in governance variables is explained
by country characteristics (Doidge et al. 2007; Aggarwal et al. 2009).
Some recent papers have focused on the relation between board charac-
teristics in financial firms and their performance during the crisis (exam-
ples include Minton et al. 2010; Beltratti and Stulz 2009; Erkens et al.
2010; Chesney et al. 2010; Adams 2012). One common theme across
these papers is the finding that financial firms with pro-shareholder
boards seemed to have performed worse during the crisis than firms
with pro-management boards.
The structure of the chapter is as follows. Section 2 briefly discusses
the sample and the data. Section 3 presents the main facts and trends
related to EU corporate boards. Section 4 presents the main results
concerning the cross-section distribution of board characteristics in
Europe. Section 5 presents a series of robustness tests. Section 6 discusses
the potential policy implications of this study and provides some brief
concluding remarks.
2. Data
The sample consists of an unbalanced panel of 2,812 listed firms in 28
European countries, of which 2,661 are from the 22 EU countries
covered. The panel data stretch over an eleven-year time period, from
2000 to 2010, and are complemented by a panel of 4,014 US-based firms.
The panel has a strong bias towards UK and US data in its early years.
Overall, the data contain 60,060 firm-year observations spanning the
whole period. Only a subset of these data is used in this analysis, which
depends on the availability of data.
The analysis is based on a copy of the entire BoardEx database drawn
in September 2011. We exclude all non-European and non-US firms, as
well as all non-listed firms, firms that were only traded in over-the-
counter (OTC) markets and the so-called ‘shell companies’, which are
not economically active. As the European coverage of BoardEx improved
substantially over the course of the 2000s, the analysis is anchored on the
years 2007 and 2010, covering 2,375 and 2,600 firms, respectively. The
findings from Europe are compared with data from 4,014 US-based
firms. The analysis does not exclude any industry or sector group, but
includes controls for 44 sectors. The sector data also come from
BoardEx.
The dataset is complemented with financial data, sourced from
CapitalIQ. We were able to match 2,553 European firms to the
BoardEx dataset and 3,939 firms that have their legal seat in the US.
To make the data comparable, for those countries that are not part of the
Eurozone all financial data are converted into EUR at market prices.
Country level economic indicators, such as the size of the economically
active population and the gross national income per capita, are sourced
from Euromonitor.
As we obtain director level data from BoardEx, we aggregate the key
variables (board size, board independence and board gender diversity)
from individual director data. Board size is defined as the number of
board members in a given year. In the case of two-tier boards, we
combine both the management and the supervisory board. We calculate
board independence as the proportion of self-declared independent
directors over board size. Although the definition of director independ-
ence varies slightly across countries, a typical definition considers a
director independent if he or she is not an employee, a former executive,
a relative of a current corporate executive, or someone who has business
relations with the company. Board gender diversity is measured as the
proportion of women on a board. We have a complete set of independ-

ence and gender data available.
3. European corporate boards: facts

This section presents an overview of the corporate board data for
selected European countries.
Figure 4.1 plots the average board size, board independence and
gender diversity (the proportion of women on boards) for twenty-two
EU countries from 2000 to 2010. The equivalent figure including all of
the European countries in the sample looks very similar, and is omitted
for the sake of brevity.
The three time series in Figure 4.1 display clear, if not remarkable,
patterns. Board size has declined significantly, from an average of 11.4 in
2000 to an average of 8.6 in 2010. The decline in the size of the board is
monotonic, but it is most dramatic from 2000 to 2006, when the average
board size fell by almost half a director per year. As can be observed from
0,40 12
0,35
10
0,30
8
0,25
0,20 6
0,15
4
0,10
2
0,05
0,00 –
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Independence Gender Diversity Boardsize
Figure 4.1: Time trends in board characteristics: European Union, 2000–10
Note: This figure shows the averages of board size (the number of directors), board
independence (the proportion of independent directors on the board) and board
gender diversity (the proportion of female directors on the board) for all EU firms in
the sample.
12,00
11,00
10,00
9,00
8,00
7,00
6,00
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
All firms Existing firms in 2000
Existing firms in 2003 Existing firms in 2006
Figure 4.2: Time trends in board size: European Union, 2000–10, stable samples
Note: This figure shows the averages of board size (the number of directors) for all EU
firms in the sample and for stable samples of EU firms from 2000, 2003 and 2006.
Figure 4.2, part of this trend is explained by composition effects, as the

sample size in 2010 is three times larger than the sample size in 2000.
Most of the newly added firms in the sample are small, and small firms
have small boards. However, this trend looks similar if the sample is kept
stable, although the fall in board size is less dramatic. In a stable sample
of existing firms in 2000 (the top line in Figure 4.2), board size falls from
an average of 11.5 directors to 10.8 directors in 2010. In a stable sample
of existing firms in 2003, board size falls from an average of 10.1
directors to 9.7 directors in 2010.
Board independence has been on the rise since 2000, but the changes
are not very impressive. The fraction of independent directors on the
board has increased from 29 per cent in 2000 to 34 per cent in 2010.
However, here the composition effect works in a different direction (see
Figure 4.3). In a stable sample of existing firms in 2000, board independ-
ence increases from an average of 29 per cent to 42 per cent in 2010. In a
stable sample of existing firms in 2003, board independence increases
from an average of 33 per cent to 40 per cent in 2010. The new firms that
enter the sample in each year have on average lower levels of board
independence, which tends to attenuate the observed trend towards
increasing board independence.
0,45
0,40
0,35
0,30
0,25
0,20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Figure 4.3: Time trends in board independence: European Union, 2000–10, stable
samples
Note: This figure shows the averages of board independence (the proportion of
independent directors on the board) for all EU firms in the sample and for stable
samples of EU firms from 2000, 2003 and 2006.
The improvement in gender diversity has been relatively more pro-

nounced, but the average levels of board gender diversity are still quite
low. The fraction of female directors on the board has doubled from 2000
to 2010, but at an average of 8 per cent in 2010, such an increase is
also not very impressive. Importantly, composition effects here appear
to have little effect on this trend. Figure 4.4 shows that the newly
added firms tended to have fewer women on their boards. However,
the difference is not economically meaningful. The breakdown in
Figure 4.4, however, has the advantage of making it clear that, on
average, all firms have clearly been increasing female participation on
their boards. In a stable sample of existing firms in 2000, female repre-
sentation has increased from 4.5 per cent in 2000 to a little more than 10
per cent in 2010.
To provide a basis for comparison, Figure 4.5 plots the same averages
for US firms. In the US, as in Europe, board size has been monotonically
declining, from 9.8 in 2000 to 8.4 in 2010. In the US sample, however,
this effect is almost fully explained by the addition of new firms to the
sample (the corresponding figures are omitted here for brevity). Board
0,110
0,100
0,090
0,080
0,070
0,060
0,050
0,040
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Figure 4.4: Time trends in board gender diversity: European Union, 2000–10, stable
samples
Note: This figure shows the averages of board gender diversity (the proportion of
female directors on the board) for all EU firms in the sample and for stable samples of
EU firms from 2000, 2003 and 2006.
independence has increased more dramatically, from 53 per cent in 2000

to 74 per cent in 2010. Most of this increase occurs until 2003, probably
reflecting the influence of the US governance reforms in 2001–03 (see
also Ferreira et al. 2010). Composition effects are of little importance to
explain the US trend towards more independent boards. Gender diver-
sity has remained basically flat throughout this period at roughly 8 per
cent–9 per cent, but it has increased for those firms that were in the
sample since 2010, from 8 per cent to 13 per cent in 2010.
The evidence reveals that the average EU corporate board is currently
very similar to the average US board on two dimensions: size and gender
diversity. In both of these cases it is observed that EU firms have been
catching up with US firms. However, US firms and EU firms remain very
different in one dimension: board independence. The EU ‘independence
gap’ with respect to the US was 19 percentage points in 2000. It then
increased dramatically to 40 percentage points in 2010. This evidence
makes it clear that, at least on average, the EU and the US appear
strikingly different in terms of corporate board independence.
0,80 10,0
0,70
9,5
0,60
0,50
9,0
0,40
8,5
0,30
0,20
8,0
0,10
– 7,5
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Independence Gender Diversity Boardsize
Figure 4.5: Time trends in board characteristics: United States, 2000–10
Note: This figure shows the averages of board size (the number of directors), board
independence (the proportion of independent directors on the board) and board
gender diversity (the proportion of female directors on the board) for all US firms in
the sample.
The average picture for EU firms is somewhat misleading. There is

much more heterogeneity across EU firms than across US firms.
Table 4.1 reports the summary statistics of board size in 2010 for each
country in the sample (which includes EU countries, some non-EU
European countries and the US). The fact that UK firms have a very
small average board size (6.5 directors) is the main reason that the
average EU board size looks similar to the average US board size. Most
of the other EU countries have significantly larger boards. The difference
in board size is not simply due to countries with dual board structures:
while Germany has an average board of 14 directors, the Netherlands has
an average of 8.9 directors, which is not much higher than the US
average. The next section discusses the determinants of the variation in
board sizes amongst European countries.
Table 4.2 reports the summary statistics for board independence in
2010. Here the picture is different. Although there is much variation
across countries, board independence levels are quite low everywhere,
except in the US and Finland.
Table 4.1 Board size across countries (2010)

Country Mean SD Min. Max. N
Austria 14.34 4.99 6 24 41
Belgium 9.46 3.57 5 23 65
Bulgaria 5.00 5 5 1
Croatia 14.50 0.71 14 15 2
Cyprus 8.33 4.44 4 18 9
Czech Republic 17.67 3.79 15 22 3
Denmark 11.41 3.43 6 20 34
Finland 7.87 1.26 6 12 31
France 10.76 4.25 3 27 247
Germany 13.99 6.83 3 30 183
Greece 9.63 3.52 5 19 52
Hungary 15.50 9.19 9 22 2
Iceland 6.67 1.53 5 8 3
Ireland 8.51 3.37 3 21 73
Italy 12.72 5.39 4 34 97
Liechtenstein 9.00 2.83 7 11 2
Luxembourg 8.67 2.78 5 17 21
Netherlands 8.88 2.96 3 16 90
Norway 7.43 2.33 3 12 63
Poland 13.43 3.41 7 20 21
Portugal 13.00 6.38 6 26 28
Romania 11.00 11 11 1
Russia 13.04 5.51 7 27 23
Slovenia 14.00 14 14 1
Spain 12.49 3.64 5 24 71
Switzerland 9.12 4.10 4 28 101
Turkey 9.20 1.99 7 14 10
United Kingdom 6.50 2.56 2 25 1,326
United States 8.42 2.51 2 33 3,799
Note: This table shows summary statistics of board size (the number of directors
on the board) across countries in 2010.
Table 4.3 reports the facts on corporate board gender diversity. There
are a few outliers, such as Norway (38 per cent), Finland (26 per cent),
Iceland (29 per cent) and Slovenia (43 per cent). Without exception,
explicit regulation can explain the higher proportions of female directors
on corporate boards in these countries. Norway has had a binding female
Table 4.2 Board independence across countries (2010)

Country Mean SD Min Max N
Austria 0.39 0.26 0.00 0.83 41
Belgium 0.43 0.16 0.00 0.83 65
Bulgaria 0.00 0.00 0.00 1
Croatia 0.03 0.05 0.00 0.07 2
Cyprus 0.33 0.25 0.00 0.67 9
Czech Republic 0.11 0.09 0.00 0.18 3
Denmark 0.29 0.23 0.00 0.89 34
Finland 0.78 0.19 0.38 1.00 31
France 0.35 0.22 0.00 1.00 247
Germany 0.05 0.16 0.00 0.86 183
Greece 0.38 0.19 0.00 0.80 52
Hungary 0.30 0.04 0.27 0.33 2
Iceland 0.33 0.58 0.00 1.00 3
Ireland 0.46 0.28 0.00 0.91 73
Italy 0.43 0.17 0.00 0.84 97
Liechtenstein 0.49 0.08 0.43 0.55 2
Luxembourg 0.51 0.29 0.00 1.00 21
Netherlands 0.52 0.22 0.00 0.88 90
Norway 0.36 0.34 0.00 1.00 63
Poland 0.17 0.18 0.00 0.56 21
Portugal 0.25 0.17 0.00 0.57 28
Romania 0.00 0.00 0.00 1
Russia 0.31 0.25 0.00 1.00 23
Slovenia 0.00 0.00 0.00 1
Spain 0.38 0.18 0.00 0.88 71
Switzerland 0.42 0.38 0.00 1.00 101
Turkey 0.18 0.23 0.00 0.63 10
United Kingdom 0.34 0.25 0.00 1.00 1,326
United States 0.74 0.17 0.00 1.00 3,799
Note: This table shows summary statistics of board independence (the proportion
of independent directors on the board) across countries in 2010.
director quota of 40 per cent since 2008. Iceland has passed a similar law,
which will become binding in 2013. Finland requires boards to have at
least one man and one woman. Slovenia, from which we have only one
observation, has rules governing the gender balance of state-owned
companies. Interestingly, Spain has passed a quota of 40 per cent that
Table 4.3 Board gender diversity across countries (2010)

Country Mean SD Min Max N
Austria 0.05 0.06 0.00 0.22 41
Belgium 0.10 0.12 0.00 0.50 65
Bulgaria 0.00 0.00 0.00 1
Croatia 0.14 0.09 0.07 0.20 2
Cyprus 0.12 0.11 0.00 0.29 9
Czech Republic 0.05 0.05 0.00 0.09 3
Denmark 0.11 0.08 0.00 0.30 34
Finland 0.26 0.12 0.00 0.43 31
France 0.12 0.11 0.00 0.75 247
Germany 0.06 0.08 0.00 0.38 183
Greece 0.08 0.12 0.00 0.43 52
Hungary 0.02 0.03 0.00 0.05 2
Iceland 0.29 0.12 0.20 0.43 3
Ireland 0.07 0.09 0.00 0.36 73
Italy 0.06 0.08 0.00 0.40 97
Liechtenstein 0.05 0.06 0.00 0.09 2
Luxembourg 0.05 0.11 0.00 0.38 21
Netherlands 0.08 0.10 0.00 0.38 90
Norway 0.38 0.10 0.00 0.63 63
Poland 0.08 0.06 0.00 0.17 21
Portugal 0.05 0.08 0.00 0.27 28
Romania 0.09 0.09 0.09 1
Russia 0.07 0.09 0.00 0.33 23
Slovenia 0.43 0.43 0.43 1
Spain 0.10 0.08 0.00 0.27 71
Switzerland 0.07 0.08 0.00 0.29 101
Turkey 0.11 0.14 0.00 0.38 10
United Kingdom 0.06 0.10 0.00 0.60 1,326
United States 0.09 0.10 0.00 0.67 3,799
Note: This table shows summary statistics of board gender diversity (the
proportion of female directors on the board) across countries in 2010.
will become binding in 2015, but the fraction of female directors is still
only 10 per cent.
The facts presented in this section raise the question of what explains
the variation in corporate board characteristics amongst the countries in
our sample. This is the question to which we now turn.
4. The cross-section of corporate board structure in Europe

This section focuses on the cross-sectional variation in board structure.
As data span over eleven years, we focus initially on a representative year
and choose the most recent year for which we have data – 2010 – as the
benchmark.
4.1. Methodology
How much of the cross-sectional variation in board structure is explained
by country effects, industry effects and firm characteristics?
Methodologically, we follow the approach of Ferreira et al. (2011) and run
linear regressions of board structure variables (size, independence and
gender diversity) on firm characteristics, industry dummies and country
dummies. We then compare the incremental (adjusted) R2 of each set of
explanatory variables (this is also the approach adopted in Doidge et al.
2007). The goal of this analysis is not to make inferences about the estimated
parameters, but to compare the explanatory power, or goodness of fit, of
these different specifications.
The main variables of interest are the size of the board, the proportion
of independent directors and the proportion of female directors on the
board. The set of firm characteristics includes two measures of firm size
(the book value of assets and sales), the market to book ratio, return on
assets (which is a measure of accounting profitability) and 44 industry
dummies. In the robustness section, we also run regressions using sales
growth instead of sales and regressions that include controls for leverage.
The list of firm characteristics is kept short, for the sake of simplicity.
To address the question of which country characteristics affect board
structure, we also run regressions with country characteristics on the
left-hand side, such as (the log of) the gross national income per capita,
(the log of) the size of the economically active population, a dummy
variable indicating a mandatory one-tier board structure and a dummy
indicator for former communist countries. Parsimonious model specifi-
cations are chosen in order not to lose too many observations due to
missing data.
The per capita gross national income serves as a proxy for economic
development. The economically active population measures both the
size of the country and the depth of labour markets. We expect both
variables to have some influence on board characteristics. The former
communist country dummy may capture the unique characteristics of
countries that underwent recent waves of privatisation.
204 daniel ferreira and tom kirchmaier
One of the few board regulations that can be compared across coun-
tries is the requirement that firms must be run by a single board, as in the
US, or by two different boards, as in Germany. In the two-tier structure,
the advisory and monitoring functions of boards are formally separated
into a management and a supervisory board (see Adams and Ferreira
2007). The rules on one-tier and two-tier board structures may also serve
as a proxy for the overall governance system of a country. Table 4.4
provides a classification of countries into one-tier or two-tier board
structures. As can be seen, some European countries adopt a single-tier
board structure (e.g. the UK), some adopt a dual board structure (e.g.
Germany) and some allow for a choice between the two (e.g. France).
A two-tier board structure is often coupled with stronger labour
representation on boards. European boards are fundamentally different
from US boards in respect of labour participation rights. For example,
whilst these rights are unknown in the US and UK, German laws allocate
substantial participation, or co-determination rights, to labour repre-
sentatives. There is substantial variation in participation rights across the
other European countries. The particular German situation is a direct
result of the German post-war consensus, in which the labour unions
agreed to a growth rate in wages that was below the growth rate in
productivity for many years after World War II. This compromise
helped the industry to re-capitalise itself more quickly and to rebuild
the country after the war. Whilst this structure gives labour a de facto
ownership over some part of the capital stock and, with it, representation
rights on boards in Germany, other European countries adopted this
model voluntarily and without the historical context. After the end of the
communist area, co-determination principles were adopted widely in
Eastern Europe as many countries modelled their own corporate law
along the lines of the German code. Co-determination rights and a two-
tier board structure often go hand-in-hand. Roe (2003) argues that the
politically awarded power of co-determination on board level brings
about concentrated ownership as a counter-balance to employee/stake-
holder strength.
Finally, to study the determinants of recent changes in board structure,
we also run regressions of the change in the relevant board characteristic
of each firm from 2007 to 2010. All other variables are defined as before,
but are now measured as of 2007. This last specification is used to study
the determinants of the most recent changes, which may have occurred
in the post-crisis period.
Table 4.4 One-tier versus two-tier board structures

One-Tier Board Two-Tier Board Choice over Board
Country Structure Structure Structure
Austria 0 1 0
Belgium 1 0 0
Bulgaria 0 0 1
Croatia 0 1 0
Cyprus 1 0 0
Czech Republic 0 1 0
Denmark 0 1 0
Finland 1 0 0
France 0 0 1
Germany 0 1 0
Greece 0 0 1
Hungary 1 0 0
Iceland 1 0 0
Italy 0 0 1
Liechtenstein 0 0 1
Luxembourg 0 0 1
Netherlands 0 1 0
Norway 1 0 0
Poland 0 1 0
Portugal 0 0 1
Republic of Ireland 1 0 0
Romania 1 0 0
Russian Federation 0 0 1
Slovenia 0 1 0
Spain 1 0 0
Switzerland 0 0 1
Turkey 1 0 0
United Kingdom 1 0 0
United States 1 0 0
Note: The table is based on Adams and Ferreira (2007) and Ferreira et al. (2010),
and was extended by consulting the individual governance codes and, where
necessary, the corporate law provisions, in March 2012. For the Nordic countries
(except Denmark) we follow note 148 by Hansen, in Geens and Hopt (2010).
Denmark was categorised as a two-tier board after observing that firms calculate
the size of the board by combining both tiers.
4.2. Board size

Table 4.5 reports the estimates of the parameters of the models (a)–(e)
(as reported in the Table), in which the dependent variable is the log of
board size. All regressions are cross-sectional and estimated by ordinary
least squares with robust standard errors.
Column (a) in Table 4.5 shows results for a regression of the log of
board size on a set of firm characteristics: (log) assets; (log) sales; market-
to-book; return on assets; and 44 industry dummies. We find that firm
size, as measured either by total assets or by revenue, is positively related
to board size. We also find that book profitability (measured by return on
assets) has a statistically weak positive association with board size. The
model in column (a) explains a sizeable part of the cross-sectional
variation in board size, with an adjusted R2 of 51 per cent.
Column (b) shows results for a regression of board size on a set of
country dummies (all dummy coefficients are omitted from the table).
This exercise reveals that country dummies alone can explain 33 per cent
of the observed variation in board size.
Finally, column (c) includes firm characteristics, industry dummies
and country dummies. The incremental explanatory power of country
dummies is not very large; the adjusted R2 increases by less than 10
percentage points when moving from (a) to (c). The incremental R2 for
the firm characteristics (plus industry dummies) is larger: moving from
(b) to (c), the R2 increases by roughly 27 percentage points. We conclude
that few firm characteristics, such as firm size and profitability, suffice to
explain much of the observed variation in board size in Europe.
Column (d) confirms the relative unimportance of country effects for
board size. The only country characteristic that shows a statistically
strong association with board size is the one-tier board structure
dummy. This association is expected; boards are likely to be larger if
they can be organised as two separate boards. Because the R2 only
falls from about 60 per cent to 55 per cent as one moves from (c) to
(d), we also conclude that the parsimonious model in (d) performs
relatively well.
We conclude that the cross-sectional distribution of board size in
Europe is mostly determined by the cross-section of firm sizes, industry
classifications and other firm characteristics. Except for the influence of
the regulations affecting the choice between single versus dual board
structures, countries matter relatively little for board size. For example,
although UK firms appear to have quite small boards, this effect is mostly
Table 4.5 Corporate board size in Europe: the impact of firm characteristics, industries and countries (2010)
Dependent Variable: Board size
Independent Variables (a) (b) (c) (d) (e)
Assets (log) 0.089*** 0.079*** 0.082*** 0.001
[7.389] [5.893] [5.704] [0.319]
Revenue (log) 0.029*** 0.023*** 0.023*** −0.003
[3.969] [3.171] [2.853] [−0.697]
Return on Assets −0.051* −0.052* −0.048* 0.093***
[−1.723] [-2.028] [−1.992] [2.912]
Market to Book −0.001 −0.000 −0.000 0.001**
[−1.458] [−0.566] [−0.832] [2.111]
One tier board structure −0.201*** −0.054***
[−3.572] [−6.441]
Former communist country 0.085 −0.081**
[0.640] [−2.415]
GNI per capita (log) −0.125 −0.057**
[−1.365] [−2.412]
Economically active pop. (log) 0.016 −0.002
[0.539] [−0.489]
Country Dummy No Yes Yes No No
Industry Dummy Yes No Yes Yes Yes
Table 4.5 (cont.)

Observations 2,288 2,288 2,288 2,285 1,951

Adj. R-square 0.506 0.329 0.599 0.545 0.005
Note: This table shows OLS regressions of corporate board size on firm characteristics, industry dummies, country dummies and
country characteristics in 2010. The sample consists of 2,288 firms from 23 European countries. The number of observations varies
because of missing data. The dependent variable in columns (a)–(d) is the natural logarithm of board size. The dependent variable in
column (e) is the change in board size from 2007 to 2010. All independent variables are as of 2010 in columns (a)–(d) and as of 2007
in column (e). Assets is the book value of total assets (in millions of EUR). Revenue is measured by sales (in millions of EUR). Market-
to-book is the market value of equity over the book common equity. Return on assets is net income over assets. GNI per capita (in
EUR, at constant 2011 prices and fixed 2011 exchange rates) is sourced from the World Bank’s World Development Indicators. The
economically active pop. is the economically active population (in thousands) sourced from the International Labour Organisation.
One-tier board is a dummy variable that equals 1 if boards are required to have a unitary board structure; this variable was hand-
collected from various sources. Former communist country is a dummy variable. Robust t-statistics are in brackets. Asterisks indicate
significance at 0.01 (***), 0.05 (**) and 0.10 (*) levels.
explained by the presence of many small UK firms in our sample. This

effect is magnified by the fact that UK firms have one-tier boards.
A different picture emerges when considering the changes in board
size since the 2007 crisis. Column (e) shows that both market-to-book
ratios and return on assets are positively related to changes in board size.
One possible explanation for this finding is that firms that performed
poorly during the crisis decided to reduce their boards. We also find that
firms in countries with one-tier board structures reduced the sizes of
their boards by more than those firms in countries with two-tier boards.
Former communist countries have also been reducing the sizes of their
boards. Finally, the decrease in board size is more pronounced in richer
countries.
Our conclusions are as follows. Most of the cross-sectional variation
in board size in Europe is explained by differences in firm characteristics,
in particular by differences in firm size and industry characteristics.
Country effects do not seem to matter much, except for the effect of
the rules governing the choice between one-tier and two-tier board
structures. Changes in board size since 2007 are explained by different
factors. Interestingly, poorly performing firms have chosen to reduce the
sizes of their boards. Country effects seem to matter more in this case:
former communist countries, countries with one-tier boards and richer
countries were all more likely to reduce the sizes of their boards.
4.3. Board independence

Table 4.6 reports the estimates for regressions in which the dependent
variable is the fraction of independent directors on the board. All right-
hand-side variables are as before.
Column (a) in Table 4.6 shows results for a regression of the log of
board independence on a set of firm characteristics (which includes 44
industry dummies). We find that firm size, profitability and market-to-
book are positively related to board independence. The most striking
result, however, is the finding that the model in column (a) explains very
little of the cross-sectional variation in board independence, with an
adjusted R2 of just 8 per cent.
Column (b) shows results for a regression of board independence on
22 country dummies. The explanatory power of country effects is far
superior to that of firm and industry characteristics: country effects alone
can explain up to 18 per cent of the cross-sectional variation in board
independence. In column (c), the complete specification with firm
Table 4.6 Corporate board independence in Europe: the impact of firm characteristics, industries and countries (2010)
Dependent Variable: Board Independence
Assets (log) 0.027*** 0.035*** 0.032*** 0.006**
[3.622] [3.747] [3.504] [2.764]
Revenue (log) 0.001 0.007* 0.007* −0.004**
[0.235] [1.981] [1.774] [−2.320]
Return on Assets 0.030** 0.028** 0.019* 0.023*
[2.075] [2.337] [1.973] [1.987]
Market-to-Book 0.002*** 0.002*** 0.002*** 0.001*
[6.985] [6.210] [6.632] [1.810]
One-tier board structure 0.168** 0.003
[2.483] [0.496]
Former communist country 0.089 0.042**
[0.806] [2.564]
GNI per capita (log) 0.087 0.012
[1.629] [0.928]
Economically active pop. (log) 0.059* 0.003
[−1.978] [0.943]
Observations 2,288 2,288 2,288 2,285 1,951
Adj. R-square 0.081 0.179 0.296 0.172 0.005
Note: This table shows OLS regressions of independence on firm characteristics, industry dummies, country dummies and country
characteristics in 2010. The sample consists of 2,288 firms from 23 European countries. The number of observations varies because
of missing data. The dependent variable in columns (a)–(d) is the proportion of independent non-executive directors on the board.
The dependent variable in column (e) is the change in independence from 2007 to 2010. All independent variables are as of 2010 in
columns (a)–(d) and as of 2007 in column (e). Assets is the book value of total assets (in millions of EUR). Revenue is measured by
sales (in millions of EUR). Market-to-book is the market value of equity over the book common equity. Return on assets is net income
over assets. GNI per capita (in EUR, at constant 2011 prices and fixed 2011 exchange rates) is sourced from the World Bank’s World
Development Indicators. The economically active pop. is the economically active population (in thousands) sourced from the
International Labour Organisation. One-tier board is a dummy variable that equals 1 if boards are required to have a unitary board
structure; this variable was hand-collected from various sources. Former communist country is a dummy variable. Robust t-statistics
are in brackets. Asterisks indicate significance at 0.01 (***), 0.05 (**) and 0.10 (*) levels.
characteristics, industry dummies and country dummies is used. The

model can explain roughly 30 per cent of the cross-sectional variation in
board independence.
The evidence here is quite different from that in section 4.2. Country
effects are more important than firm characteristics for explaining board
independence. This result is similar to the evidence reported by Ferreira
et al. (2010) in a sample of commercial banks. They argue that differ-
ences in regulations and governance practices can explain the import-
ance of countries for bank board independence. However, unlike
Ferreira et al., here much of the cross-sectional variation in board
independence remains unexplained.
Moving to column (d), we find that the adjusted R2 drops significantly
from 30 per cent to 17 per cent. This drop suggests that the country
characteristics included in model (d) capture only part of the importance
of country effects. As expected, boards are more independent in coun-
tries with one-tier boards. Larger countries appear to have less inde-
pendent boards, but this effect is statistically weak.
Column (e) considers the determinants of changes in board independ-
ence. Good performing firms (as measured by ROA) appear to have
increased board independence, but these effects are statistically weak.
The effect of firm size on independence is ambiguous: the two different
proxies for size have effects of different signs on independence. Board
independence has been increasing in former communist countries,
which is to be expected as these countries catch up with others.
Our conclusions are as follows. Countries appear to matter for
board independence. Board regulations and business practices vary
substantially across European countries, which could explain the
importance of country effects for board independence. Both firm size
and firm performance are positively related to board independence in
European countries. But much of the variation in board independence
is not explained by firm size, performance, industry effects or country
effects.
Poor performance in the crisis has led, if anything, to reductions in
board independence levels. Thus, there is no evidence that European
firms reacted to the crisis by increasing the independence of their
boards. In fact, the average board independence in European firms
has remained stable at 34 per cent from 2004 to 2010. Despite the
existence of considerable variation across European countries, there
seems to have been no structural change in board independence levels
in recent years.
4.3. Board gender diversity

As discussed in Section 3, gender diversity in the boards of European firms
has been monotonically improving since 2000, but the average proportion
of women on boards is still quite low, at levels below 10 per cent.
Table 4.7 considers the cross-sectional determinants of board gender
diversity, as measured by the proportion of female directors on the board.
Column (a) shows that firm and industry characteristics can explain only a
trivial fraction of the cross-sectional variation in board gender diversity. The
adjusted R2 is quite low: 4 per cent. Not surprisingly, country effects do a
much better job of explaining the cross-section of board gender diversity. In
column (b), the adjusted R2 is just under 25 per cent. This is strong evidence
of the power of gender balance rules and quotas which have been adopted
by some countries in Europe. The combination of country effects with firm
characteristics can explain 30 per cent of the variation (column (d)).
There are two important results here. First, firm profitability, as
measured by return on assets, is positively related to the proportion of
women on the boards of European firms. This evidence is quite robust in
the data. A positive correlation between gender diversity and profit-
ability is reported in many different studies (see Ferreira 2010 for a
review of the literature). It is tempting to conclude that board gender
diversity improves firm performance. However, equally plausible is the
hypothesis that more profitable firms select more women to their boards.
Adams and Ferreira (2009) provide some evidence in support of this
hypothesis. For a recent analysis of the performance effects of the
introduction of gender quotas in Norway, see Ahern and Dittmar (2012).
Second, the importance of gender-specific rules for explaining the evi-
dence is underscored by the results in column (d). Country characteristics
such as economic development and country size have little effect on board
gender diversity. The country effects that explain board gender diversity are
very idiosyncratic. The countries with explicit gender balance policies, such
as Norway, Iceland and Finland, are the ones that explain most of the
variation. It is expected that Spain and France will also show significant
effects in the near future, once their gender balance policies become binding.
Column (e) shows that larger firms have been increasing the propor-
tion of female directors on their boards since 2007. As board gender
diversity has become a more important policy issue, it is natural to expect
that more visible (larger) firms would be the first to employ more
women. Another possibility is that larger firms may find it easier to
recruit top female directors.
Table 4.7 Board gender diversity in Europe: the impact of firm

characteristics, industries and countries (2010)
Dependent Variable: Board Gender Diversity
Assets (log) 0.002 0.003** 0.002 0.000
[0.705] [2.333] [0.884] [0.149]
Revenue (log) 0.003 0.003** 0.002 0.003**
[1.601] [2.544] [0.974] [2.181]
Return on Assets 0.017** 0.020** 0.018*** −0.001
[2.520] [2.402] [2.858] [−0.217]
Market-to-Book 0.000 0.000* 0.000 −0.000
[1.002] [1.847] [1.377] [−0.890]
One-tier board structure −0.004 −0.000
[−0.302] [−0.080]
Former communist country 0.112 −0.029
[1.254] [−1.702]
GNI per capita (log) 0.102 −0.019**
[1.374] [−2.420]
Economically active pop. (log) −0.010 −0.006*
[−1.052] [−2.058]
Observations 2,288 2,288 2,288 2,285 1,951
Adj. R-square 0.044 0.247 0.303 0.118 0.005
Note: This table shows OLS regressions of the gender ratio on firm characteristics,
industry dummies, country dummies and country characteristics in 2010. The
sample consists of 2,288 firms from 23 European countries. The number of
observations varies because of missing data. The dependent variable in columns
(a)–(d) is the proportion of female directors on the board. The dependent variable
in column (e) is the change in board gender diversity from 2007 to 2010. All
independent variables are as of 2010 in columns (a)–(d) and as of 2007 in column
(e). Assets is the book value of total assets (in millions of EUR). Revenue is
measured by sales (in millions of EUR). Market-to-book is the market value of
equity over the book common equity. Return on assets is net income over assets.
GNI per capita (in EUR, at constant 2011 prices and fixed 2011 exchange rates) is
sourced from the World Bank’s World Development Indicators. The economically
active pop. is the economically active population (in thousands) sourced from the
International Labour Organisation. One-tier board is a dummy variable that
equals 1 if boards are required to have a unitary board structure; this variable was
hand-collected from various sources. Former communist country is a dummy
variable. Robust t-statistics are in brackets. Asterisks indicate significance at 0.01
(***), 0.05 (**) and 0.10 (*) levels.
The main conclusions are as follows. The proportion of female direc-

tors on European corporate boards has been on the rise since 2000.
However, this fraction is still quite small, at about 8 per cent. Much of
the progress was only made possible by active intervention; only those
countries with explicit gender balance policies have averages above 20
per cent. Left to their own devices, European firms have done little to
increase female representation on their boards. We also find that profit-
ability and board gender diversity are strongly associated, but we would
caution against making causal statements on the basis of this correlation.
5. Additional evidence and robustness

This section examines the robustness of the conclusions and provides
some additional evidence.
5.1. Alternative specifications

Specifications (a) to (d) in Section 4 use contemporaneous firm and
country characteristics. Our goal is not to infer causal relationships,
but simply to investigate how much of the variation in board character-
istics is ‘explained’ by cross-sectional variation in firm characteristics.
The use of contemporaneous independent variables mitigates concerns
about firm characteristics being measured with error, as they better
reflect the current state of the firm. One drawback, however, is that
contemporaneous firm variables may be temporarily affected by board
characteristics, and as such should not be considered as ‘long run’
determinants of board characteristics. This section replicates the pre-
vious tables now using all independent variables as of 2007. It is import-
ant to note that the use of lagged variables is not aimed at establishing
causality. Rather, the goal here is simply to ascertain which correlations
seem to be persistent and which are short-lived.
To address possible concerns with multicollinearity of measures of
firm size, this section also reruns all regressions after replacing sales with
sales growth. For brevity of exposition, only results in which sales growth
is used are reported. There is no important difference between the results
reported in this section and the results from unreported regressions that
use lagged sales as a control variable.
Table 4.8 reports the results for board size. Overall, the results are very
similar to those in Table 4.5. There are two important differences,
though. First, past performance (as measured by ROA in 2007) is
Table 4.8 Board size in 2010 and firm characteristics in 2007

Assets (log) 0.122*** 0.102*** 0.105*** −0.003
[9.191] [9.766] [9.754] [−1.681]
Sales growth 0.000 0.000 0.000 0.001***
[0.877] [0.985] [0.748] [6.383]
Return on Assets 0.081* 0.047 0.078** 0.096**
[2.024] [1.542] [2.261] [2.765]
Market-to-Book 0.002*** 0.002*** 0.002*** 0.001**
[2.970] [5.447] [4.197] [2.182]
One-tier board structure −0.218*** −0.057***
[−3.178] [−6.596]
Former communist country −0.235 −0.128***
[−1.057] [−3.277]
GNI per capita (log) −0.167 −0.057**
[−1.193] [−2.188]
Economically active pop. (log) −0.009
[−0.351] [−1.355]
Observations 1,655 1,655 1,655 1,655 1,665
Adj. R-square 0.503 0.326 0.600 0.543 0.023
Note: This table shows OLS regressions of corporate board size in 2010 on firm
characteristics, industry dummies, country dummies and country characteristics
as of 2007. The sample consists of 2,288 firms from 23 European countries. The
number of observations varies because of missing data. The dependent variable in
columns (a)–(d) is the natural logarithm of board size. The dependent variable in
column (e) is the change in board size from 2007 to 2010. All independent
variables are as of 2007. Assets is the book value of total assets (in millions of EUR).
Sales growth is the one-year change in sales divided by sales in the previous year.
Market-to-book is the market value of equity over the book common equity.
Return on assets is net income over assets. GNI per capita (in EUR, at constant
2011 prices and fixed 2011 exchange rates) is sourced from the World Bank’s
World Development Indicators. The economically active pop. is the economically
active population (in thousands) sourced from the International Labour
Organisation. One-tier board is a dummy variable that equals 1 if boards are
required to have a unitary board structure; this variable was hand-collected from
various sources. Former communist country is a dummy variable. Robust t-
statistics are in brackets. Asterisks indicate significance at 0.01 (***), 0.05 (**) and
0.10 (*) levels.
positively related to board size. Second, market-to-book now appears to

be positively related to board size.
Table 4.9 reports the results for board independence. Again, the main
conclusion remains the same: country effects are much more important
than firm characteristics for explaining board independence. The posi-
tive correlation between ROA during the crisis and changes in board
independence appears slightly stronger in this specification.
Table 4.10 reports the results for board gender diversity. Again, the
results are similar. One key difference is that the positive correlation
between ROA and board diversity is now statistically weak. This is
further discussed in Section 5.3.
5.2. Leverage
We also re-run all regressions after including leverage (assets over
equity) as a control. Leverage is robustly negatively related to board
size and independence. It does not have a statistically reliable relation
with board diversity. All previous results remain unchanged after the
inclusion of leverage in the regressions.
5.3. UK versus non-UK firms

About half of the sample firms come from the UK. It is thus natural to
ask which results are influenced by the disproportionate importance of
UK firms in the sample. To address this question, all previous regressions
are rerun using only non-UK firms. Tables are omitted for the sake of
brevity.
There is virtually no difference in the board size regressions. There
seems to be nothing special to UK firms when it comes to the relation
between firm characteristics and board size. The same is true in the
regressions with board independence. Obviously, because of the reduced
sample size (about 800 firms), most estimates appear statistically weaker.
Nevertheless, all point estimates are similar to those estimated in the full
sample.
There is one remarkable difference between the full sample and the
non-UK sample regressions with board diversity. In the non-UK sample,
there is a positive and statistically strong relation between past perform-
ance (ROA in 2007) and current levels of board gender diversity (in
2010). In the full sample, this relationship is weaker and not statistically
significant (see Table 4.10). We conclude that ROA is not a good
Table 4.9 Board independence in 2010 and firm characteristics in 2007

Dependent Variable: Board Independence
Assets (log) 0.028* 0.042*** 0.041*** 0.003**
[1.968] [3.942] [3.838] [2.344]
Sales growth −0.000 −0.000 −0.000 −0.001***
[−1.264] [−1.233] [−0.542] [−6.318]
Return on Assets 0.014 0.014 −0.006 0.029**
[0.608] [0.645] [−0.227] [2.405]
Market-to-Book 0.001** 0.001* 0.001* 0.000
[2.840] [2.051] [2.095] [1.084]
One-tier board structure 0.168** 0.006
[2.512] [1.071]
Former communist country 0.118 0.018
[0.959] [1.202]
GNI per capita (log) 0.021 0.006
[0.302] [0.339]
Economically active pop. (log) −0.049*
[−1.951] [0.831]
Observations 1,655 1,655 1,655 1,655 1,665
Adj. R-square 0.060 0.213 0.313 0.171 0.007
Note: This table shows OLS regressions of corporate board independence in 2010
on firm characteristics, industry dummies, country dummies and country
characteristics as of 2007. The sample consists of 2,288 firms from 23 European
countries. The number of observations varies because of missing data. The
dependent variable in columns (a)–(d) is the proportion of independent non-
executive directors on the board. The dependent variable in column (e) is the
change in independence from 2007 to 2010. All independent variables are as of
2007. Assets is the book value of total assets (in millions of EUR). Sales growth is
the one-year change in sales divided by sales in the previous year. Market-to-book
is the market value of equity over the book common equity. Return on assets is net
income over assets. GNI per capita (in EUR, at constant 2011 prices and fixed 2011
exchange rates) is sourced from the World Bank’s World Development Indicators.
The economically active pop. is the economically active population (in thousands)
sourced from the International Labour Organisation. One-tier board is a dummy
variable that equals 1 if boards are required to have a unitary board structure; this
variable was hand-collected from various sources. Former communist country is a
dummy variable. Robust t-statistics are in brackets. Asterisks indicate significance
at 0.01 (***), 0.05 (**) and 0.10 (*) levels.
Table 4.10 Board gender diversity in 2010 and firm characteristics in

2007
Dependent Variable: Board Gender Diversity
Assets (log) 0.005 0.009*** 0.009*** 0.004**
[0.863] [13.875] [5.071] [2.698]
Sales growth −0.000** −0.000*** −0.000*** −0.000
[−2.406] [−8.956] [−3.793] [−1.294]
Return on Assets 0.037 0.010 0.020 0.002
[0.986] [0.840] [0.881] [0.341]
Market-to-Book 0.000 0.000 0.000 −0.000
[0.191] [1.349] [0.083] [−0.469]
One-tier board structure 0.039 0.000
[1.489] [0.038]
Former communist country 0.218 −0.009
[1.669] [−0.680]
GNI per capita (log) 0.153 −0.021**
[1.640] [−2.651]
Economically active pop. (log) −0.005
[−0.680] [−2.811]
Observations 1,655 1,655 1,655 1,655 1,665
Adj. R-square 0.036 0.291 0.338 0.153 0.005
Note: This table shows OLS regressions of the gender ratio on firm characteristics,
industry dummies, country dummies and country characteristics in 2010. The
sample consists of 2,288 firms from 23 European countries. The number of
observations varies because of missing data. The dependent variable in columns
(a)–(d) is the proportion of female directors on the board. The dependent variable
in column (e) is the change in board gender diversity from 2007 to 2010. All
independent variables are as of 2010 in columns (a)–(d) and as of 2007 in column
(e). Assets is the book value of total assets (in millions of EUR). Revenue is
measured by sales (in millions of EUR). Market-to-book is the market value of
equity over the book common equity. Return on assets is net income over assets.
GNI per capita (in EUR, at constant 2011 prices and fixed 2011 exchange rates) is
sourced from the World Bank’s World Development Indicators. The economically
active pop. is the economically active population (in thousands) sourced from the
International Labour Organisation. One-tier board is a dummy variable that
equals 1 if boards are required to have a unitary board structure; this variable was
hand-collected from various sources. Former communist country is a dummy
variable. Robust t-statistics are in brackets. Asterisks indicate significance at 0.01
(***), 0.05 (**) and 0.10 (*) levels.
predictor of board gender diversity in the UK, while it appears to be so in

other European countries.
6. Concluding remarks and policy notes

This chapter provides the first comprehensive account of the distribu-
tion of board size, independence and gender diversity in Europe in the
post-crisis period. Understanding the determinants of board character-
istics is a necessary step to assess the likely success or failure of many of
the governance reforms that have recently been proposed.
We find that European firms have been reducing the sizes of their
boards. This pattern is more pronounced for those firms that performed
poorly during the crisis period. We also find that firm size and industry
affiliation are the most robust determinants of board size. Despite the
differences in regulations mandating either a one- or a two-tier board
structure, the differences in average board sizes in Europe and in the US
are not significant.
European firms have also been increasing the independence of their
boards, but from a distinctly lower level than their US counterparts.
Whilst in the US almost three out of four directors are independent, in
Europe independent directors are still in the minority. Interestingly,
European firms that performed poorly during the crisis have chosen to
reduce board independence. This finding and the existing evidence
relating pro-shareholder boards in financial firms to poor performance
in the crisis (e.g. Beltratti and Stulz 2009) question the received wisdom
that more board independence is beneficial in crisis periods.
Board gender diversity is a fairly recent – but passionately debated –
topic in both public and academic debates. There have been many policy
initiatives, at both country and European level, aimed at increasing
female participation on corporate boards. Some countries, such as
Norway or Finland, have chosen to increase the female participation
rate through quotas or explicit rules, whilst others, such as the UK, are
attempting to achieve gender balance through voluntary targets. The
evidence in this chapter is relevant for this debate as well: differences in
regulations across countries are the single most important factor that
explains the differences in board gender diversity across European
countries. Whilst female participation on European boards has been
increasing continuously since 2000, significant changes should be attrib-
uted mainly to policy initiatives.
This chapter does not show direct evidence of specific country-level

characteristics affecting board gender diversity, but such evidence is
available from contemporaneous studies. For example, Adams (2009)
examines the extent to which female labour force participation and
institutional and country-level factors are related to board diversity.
They find that female labour force participation is not related to the
representation of women in executive ranks. In fact, female executive
participation cannot be explained by any country-specific variables. In
contrast, female labour force participation is positively correlated with
the representation of women in non-executive positions. This suggests
that policies that facilitate female labour force participation may also
eventually have an impact on corporate boards. However, cultural norms
are also correlated with female participation in non-executive positions
even after controlling for labour force participation. This raises the
question of whether policies directly targeting boardroom diversity will
be sustainable in the long run.
Given the importance of the corporate board in the governance frame-
work of firms, questions of board structure, composition and conduct
feature highly in the various governance codes and policy initiatives on
both country and European levels. In fact, such an initiative was first
introduced by the UK’s Treasury in 1982. At that time it was hoped that
more effective work by non-executive directors could help improve the
performance of the UK’s commercial sector, which was perceived to be
lagging behind other countries in terms of competitiveness. While it will
never be known whether this initiative had the desired effect, it did,
however, form the foundations of the UK’s governance code. Thirty
years of governance debate produced a code that is mature and widely
accepted within the UK’s business community and allows for flexibility
through its ‘comply or explain’ rule (even if the actual use of the explain
rule is rare – see Arcot et al. 2010). The maturity and flexibility of the
code gives it credibility within the boardroom. Interviews with company
chairmen revealed just how important this is: boards that see the use-
fulness of certain code provisions are happy to apply them, while less-
accepted provisions are seen as ‘box-ticking exercises’ with little impact
on boardroom behaviour (Owen and Kirchmaier 2008).
Herein we see the dangers of regulation that is not sufficiently tied to
the individual needs of a company, the institutional structure of a
country, or both. Disconnected regulation can easily be seen in the
boardroom as an unnecessary burden, and therefore dismissed. Given
that boards are typically staffed with people of extraordinary talent, what
might be needed is a vivid public debate about the benefits of certain best
practices that convinces boards of their merits, rather than new regu-
lation. Given that board structures in Europe are so diverse, far-reaching
regulation with little connection to the individual needs of firms runs the
risk of being dismissed as another ‘box-ticking exercise’.
References
Adams, R. 2012. ‘Governance and the financial crisis’, International Review of
Finance 12: 7–38.
and D. Ferreira 2007. ‘A theory of friendly boards’, Journal of Finance 62:
217–50.
2009. ‘Women in the boardroom and their impact on governance and
performance’, Journal of Financial Economics 94: 291–309.
Aggarwal, R., I. Erel, R. Stulz and R. Williamson 2009. ‘Differences in governance
practices between U.S. and foreign firms: measurement, causes, and
consequences’, Review of Financial Studies 22: 3132–69.
Ahern, K. R., and A. K. Dittmar 2012. ‘The changing of the boards: the impact on
firm valuation of mandated female board representation’, Quarterly Journal
of Economics 127: 137–97.
Arcot, S., V. Bruno and A. Faure-Grimaud 2010. ‘Corporate governance in the
UK: Is the comply or explain approach working?’ International Review of
Law and Economics 30: 193–201.
Beltratti, A. and R. Stulz 2009. ‘Why Did Some Banks Perform Better During the
Credit Crisis? A Cross-Country Study of the Impact of Governance and
Regulation,’ Working Paper No. 254/2009, Bocconi University and The
Ohio State University, forthcoming Journal of Financial Economics.
Boone, A., L. Field, J. Karpoff and C. Raheja 2007. ‘The determinants of corporate
board size and composition: an empirical analysis’, Journal of Financial
Chesney, M., J. Stromberg and A. F. Wagner 2010. ‘Risk-taking Incentives,
Governance, and Losses in the Financial Crisis’, Swiss Finance Institute
Research Paper Series No. 10–18.
Coles, J., N. Daniel and L. Naveen 2008. ‘Boards: does one size fit all?’ Journal of
Financial Economics 87: 329–56.
Davies, M. 2011. ‘Women on Boards’, The Davies Report.
Doidge, C., A. Karolyi and R. Stulz 2007. ‘Why do countries matter so much for
corporate governance?’ Journal of Financial Economics 86: 1–39.
Erkens, D., M. Hung and P. Matos 2010. ‘Corporate Governance in the 2007–2008
Financial Crisis: Evidence from Financial Institutions Worldwide’,
Working Paper, University of Southern California, forthcoming Journal of
Corporate Finance.
European Commission 2010, Corporate Governance in Financial Institutions and
Remuneration Policies. Green Paper available at http://ec.europa.eu/inter-
nal_market/company/modern/corporate_governance_in_financial_insti-
tutions_en.htm.
Ferreira, D. 2010. ‘Board Diversity’, Chapter 12 in H. K. Baker and R. Anderson
(eds.), Corporate Governance: A Synthesis of Theory, Research, and Practice,
pp. 225–42, Hoboken, NJ: John Wiley & Sons.
Ferreira, D., M. Ferreira and C. Raposo 2011. ‘Board structure and price
informativeness’, Journal of Financial Economics 99: 523–45.
Ferreira, D., T. Kirchmaier and D. Metzger 2010. ‘Boards of Banks’, EGGI Finance
Working Paper No. 289/2010.
Geens, K. and K. Hopt 2010. The European Company Law Action Plan Revisited:
Reassessment of the 2003 Priorities of the European Commission, Leuven
University Press.
Kirkpatrick, G. 2009. ‘The corporate governance lessons from the financial crisis’,
Financial Markets Trends 96: 2009/1.
Lehn, K. M., S. Patro and M. X. Zhao 2009. ‘Determinants of the size and compo-
sition of US corporate boards: 1935–2000’, Financial Management 38:
747–80.
Linck, J., J. Netter and T. Yang 2008. ‘The determinants of board structure’,
Minton, B. A., J. A. Taillard and R. Williamson 2010. ‘Do Independence and
Financial Expertise of the Board Matter for Risk Taking and
Performance?’ Working Paper, The Ohio State University, Boston College
and Georgetown University.
Owen, G. and T. Kirchmaier 2008. ‘The changing role of the chairman: Impact of
corporate governance reform in the United Kingdom 1995–2005’,
European Business Organization Law Review 9: 187–213.
Roe, M. 2003. Political Determinants of Corporate Governance: Political Context,
Corporate Impact, Oxford University Press.
Walker, D. 2009. A Review of Corporate Governance in UK banks and other
Financial Industry Entities: Final Recommendations. Online document
available from www.hm-treasury.gov.uk/walker_review_information.htm.
5
Board on Task: developing a comprehensive

understanding of the performance of boards
jaap winter and erik van de loo
1. Introduction
The governance crisis of 2001–03 and the financial crisis of 2007–08
have sparked and continue to spark extensive debate on and regulation
of boards of directors of companies.1 The term ‘board’ in this chapter is
used to describe the interaction between non-executives and executives,
regardless of whether the organisational structure is a one-tier board
which comprises executive directors and non-executive directors, or a
two-tier model in which executive directors and non-executive directors
take seats in two separate boards.
In short, the governance crisis revealed that boards of listed compan-
ies apparently were unable to stop executives from manipulating finan-
cial statements in order to boost their bonuses, stock options and
performance shares. In the financial crisis, boards of financial institu-
tions were unable to ensure proper risk management (Winter 2012a).
The two crises have left us wondering what boards and, within boards,
what non-executives are actually doing or should be doing. The regu-
latory response in both cases has been to enforce the monitoring role of
non-executives towards executives, fuelled by the dominant economic
corporate governance model of the agency theory. Non-executive duties
in fields like audit, internal control and risk management, which we refer
to as ‘corporate hygiene’, have been described and detailed explicitly in
mandatory regulation and corporate governance codes. At the same
time, the personal responsibility of non-executives for a professional
execution of this monitoring role has been emphasised.
1
The governance crisis refers to the corporate scandals that occurred in that period,
including, in particular, Enron and Worldcom.
225
226 jaap winter and erik van de loo
All of this creates a new reality for boards and new dynamics between
executives and non-executives. It also creates new expectations of what
good board performance is supposed to deliver. But the regulatory
approach and the economic approach to boards fail to provide a com-
prehensive view of sound board performance. They fail to integrate the
roles of executives and non-executives and tend to focus solely, or at least
overly, on the latter and only on one particular aspect of their role. This
single-minded focus on the monitoring role of non-executives in the
board distorts what board performance is truly about. The legal and
economic approaches also ignore that the board is a social phenomenon,
an organisational institution through which people cooperate, debate
and take decisions in order to achieve certain objectives. Personal behav-
iour of individual board members and behaviour of the group that
constitutes the board unavoidably affect the performance of the board
(Bainbridge 2002). In order to understand board performance and what
drives it we need to move beyond the narrow economic and legal
perceptions of the role of boards and develop an integrated, compre-
hensive understanding of the functioning of boards.
Based on an Organisational Role Analysis, we first develop the concept of
the Board on Task. This concept clarifies roles of executives and non-
executives in their interaction. These roles and the way they are perceived
are a key driver of board behaviour of executives and non-executives. To
provide a comprehensive view of the performance of the board, we comple-
ment Board on Task with Board GPS, a set of lenses that allows us to
distinguish three key areas of factors that determine board performance:
Group, Person and System.
This chapter first sets out how the economic approach and the legal
approach to boards provide an incomplete view of the role of boards and
the factors that determine board performance (Section 2). It then pro-
ceeds to describe our model for understanding boards and board per-
formance, built on two concepts: Board on Task and Board GPS
(Section 3). The chapter focuses on the first concept: Board on Task
(Section 4) as a critical building block for understanding board perform-
ance. A lack of clarity of the role of the board and of the non-executive
and executive players is a recipe for board failure. In our experience,
perceptions of executives and non-executives and, among non-
executives, of a particular board, often differ as to what the role of the
board encompasses. The classic legal and economic approaches to
boards provide only an incomplete perspective on the role of the board
and what it means to be ‘on task’. We show that, even in defining the
developing an understanding of performance 227
role of the board and what it means to be ‘on task’, a broader perspective
is needed, encompassing management and behavioural theories, in
order to start to make sense of the performance of boards. We then
describe how this model can facilitate reviewing and developing board
performance in practice and that it can be a source of inspiration for
novel academic research into board performance (Section 5) and finally,
draw some policy conclusions (Section 6).
2. The unhappy (or at least, incomplete) marriage of law and

economics to understanding boards
Law, at least law on the law books, typically has a minimalistic approach
to what the role of the board is and what it requires of executives and
non-executives. Dutch law, for example, provides that the duty of the
management board (het bestuur) is to manage the company, full stop.2
No further explanation is provided of what managing the company
actually entails. Specific duties are scattered around the Companies
Act, such as the duty to draw up and publish annual accounts, organise
general meetings, etc. The duties described in the law are almost entirely
of a formalistic nature, ensuring the legal operation of the company as a
construction of the law. The duty of the supervisory board (raad van
commissarissen) is described as supervising and advising the manage-
ment board, again full stop and with some scattered specific, formal
duties.3 In addition, the law provides some procedural prescriptions on
decision-taking by corporate bodies and representation of the company,
again ensuring its functioning as a legal entity.
The other involvement of the law with the operation of boards is in the
case of derailment when it all goes wrong. The law imposes personal
liability on directors when they have breached their duties. Generally
such breaches occur when either shareholders or creditors of the com-
pany are prejudiced by the acts of directors. Law typically allows direc-
tors a large discretion on how to conduct their roles and intervenes only
at the margins. Legal concepts like the business judgement rule or the
marginal review of conduct by courts are expressions of this notion
2
Article 2:129 Dutch Civil Code.
3
Interestingly, management literature on boards of directors describes the key role of the
board in the one-tier system in remarkably similar terms: control and service, suggesting
that the board of directors in a one-tier board has the same function as the supervisory
board in the two-tier board. See, for example, Forbes and Milliken (1999).
(Assink 2007). Law stays at the margins of what boards are doing by
focusing on formalities, process and liability. Within these margins the
law is agnostic about what boards are or should be doing (Lorsch 2012).
Not so economics and economists. Economists do take a view on what
boards should be doing, and also on how to regulate boards to ensure
that they do it. The key objective from an economic perspective is maxi-
mising efficiency, understood as maximising social wealth. The dominant
economic approach in relation to corporate governance and boards is the
agency theory developed by Jensen and Meckling (1976). In this theory,
managers of a company are the agents for the shareholders as principals.
Shareholders run the ultimate economic risk of the company but (at least in
a dispersed ownership context where ownership and control are separated,
which typically is the starting point of this analysis) do not take the crucial
decisions determining the direction of the company. This is what they hire
management for.
Managers as rational utility maximising agents, however, do not have the
same interests as shareholders as principals. Their constant involvement in
the affairs of the company provides them with opportunities to extract
benefits from the company to the detriment of the principals either by
slacking: persistent underperformance without being corrected, by empire
building and being rewarded for running a bigger empire (Tosi and Gomez-
Mejia 1994), or by cheating: creating an illusion of good performance by
misleading financial statements or outright stealing.
Corporate governance is all, or at least primarily, about providing
mechanisms to discipline the managers as agents for the benefit of
shareholders as principals. Shareholder rights, such as appointment
and dismissal rights and voting rights on specific transactions, takeover
bids, performance-based executive pay and oversight of executives by
non-executive directors are all governance mechanisms in this sense:
focused on disciplining management (Kraakman et al. 2009).
Fundamental criticism has been raised against the agency theory. For
one thing, it ignores other stakeholders, such as creditors and employees
as principals. The financial crisis has, at the least, cast serious doubts on
the wisdom of the singular focus of financial institutions on serving
shareholder interests, to the detriment of deposit holders and other
creditors.4 The theory has also been challenged in its core notion of the
4
See more broadly, Masouros (2012) on the contribution of the agency theory and its legal
application in corporate governance to create short-term focused financial capitalism and
the value destruction this has caused.
rational economic behaviour of managers as agents, causing them to

serve their own interests as much as they can (Donaldson and Davis
1990; Blair and Stout 1999). Lawyers have also criticised the theory
because it focuses only on economic efficiency objectives and appears
to disregard notions intrinsically relevant to law, such as fairness
(Kronman 1980; Tamanah 2006).
Regardless of this criticism, the agency theory has rapidly become the
dominant economic theory in the field of corporate governance. It has
allowed economic and financial academics to conduct research into the
efficiency of various disciplining mechanisms, applying ever more
advanced mathematical and statistical tools, telling us what is efficient
and what is not. This rising understanding of and focus on disciplining
mechanisms helped institutional investors in the 1990s and 2000s to
demand more rights, spurring the modern corporate governance debate
and practice. And then the corporate governance crisis happened, trig-
gering a strong regulatory response. In the absence of any meaningful
view of the law and lawyers on what boards should be doing, the
dominant economic theory also became the key driver for regulatory
change relating to boards. The failing of non-executive directors as
gatekeepers was revealed and the regulatory response was to strengthen
their monitoring role. Surprisingly, the effectiveness and efficiency of
performance-based executive pay as a mechanism to align the interests
of managers with the interests of shareholders was not challenged. In the
US, performance-based executive pay was completely ignored as a
source of the problems with misleading financial statements, while in
the EU the focus was again more on disclosure and the monitoring role
of non-executives in the field of executive pay than on the problems of
variable executive pay itself.5
The financial crisis and subsequent research indicate that the prob-
lems are deeper than mere design and governance failures, as identified
by Bebchuk and Fried (2004) and Jensen and Murphy (2004) and that
5
The first regulatory reflection in the EU on remuneration as a governance problem was
from the High Level Group of Company Law Experts in 2002, recommending disclosure
of remuneration policy and individual director pay and shareholder say on pay, the
setting up of remuneration committees on the board and proper accounting for the costs
of stock option and share grant schemes, and finally, for the Commission to issue a
Recommendation on a proper regulatory regime for director pay, see http://ec.europa.eu/
internal_market/company/docs/modern/report_en.pdf. This led to the Commission
Recommendation of 14 December 2012, see http://eur-lex.europa.eu/LexUriServ/
LexUriServ.do?uri=OJ:L:2004:385:0055:0059:EN:PDF.
the risks and costs of substantial incentives for executives may exceed the
benefits of any alignment they seek to bring (Winter 2012b). And so
agency theory infused regulation primarily sought to strengthen the
monitoring role of non-executive directors.
As the governance crisis was triggered by massive frauds through
misleading financial statements, the focus of the monitoring role of
non-executives became the audit of financial statements and the internal
controls on which they are based. Audit committees of non-executive
directors were required to pay closer scrutiny in these areas. Non-
executive directors had to improve their understanding of audit and
control matters, and at least some of them should be financial experts.
A strong emphasis was put on the independence of non-executive
directors, as it was found that strong ties with executives could seriously
impair their monitoring role as non-executives. The financial crisis
added a new non-executive focus on risk and risk management to the
spectrum of their monitoring role.
In between the governance and financial crisis, a new notion was
developed from a different angle: diversity. It was mainly a societal
debate on the lagging representation of women in leading positions
that triggered the diversity debate, later followed by behavioural insights
that diversity can also contribute to the quality of decision-taking.
Diversity, typically reduced to gender diversity, is now also finding its
way into regulation.6
Economic and finance research in the meantime has focused on trying
to establish a link between the performance of the firm and aspects of the
regulation of boards, in particular requirements of independence, exper-
tise and diversity. A wealth of studies has been published investigating
whether such a link exists. They typically take as a starting point input
variables on the composition of boards that are publicly available based
on annual reports of companies, reporting on the independence, exper-
tise and diversity aspects of their non-executive board members. These
data are then combined with data on the performance of the companies
in the data set. Statistical methods are applied to find whether there is a
6
See art. 2:166 Dutch Civil Code, requiring a representation of at least 30% men and women on
both management boards and supervisory boards in the Netherlands. If the 30% is not
reached, the company must explain what efforts it has made and intends to make to reach the
target. Other EU countries have developed similar diversity requirements, either mandatory
or on the basis of comply or explain. EU Commissioner Reding intends to impose a quota for
women on boards across the EU, see www.theparliament.com/latest-news/article/newsar-
ticle/reding-urges-meps-to-back-women-on-board-quotas/.
meaningful correlation between the input variables at board level and

firm performance. The research is often inconclusive and, even when
clear conclusions are drawn, the meaningfulness of the conclusions can
often be doubted.
Take the example of the independence of non-executives. This notion
has been developed to deal first of all with conflicts of interests. The idea
is that if a person has or has had a substantial relationship with the
company or its management, for example as a previous executive direc-
tor, as a director or employee of a bank providing financial services to the
company on a more than incidental basis, as a lawyer providing legal
services to the company etc, such a relationship may impair the ability of
the person to make objective, independent judgements as a non-
executive. As it is difficult substantively to regulate independence of
mind, regulation has chosen to designate formal current or previous
relationships that are deemed to make the non-executive not independ-
ent, and to require that a minimum number of non-executives can be
deemed to be independent by these standards.7
This approach is intuitively sound and is also supported by behav-
ioural research indicating that people consistently overestimate their
own objectivity. People generally tend to find themselves less susceptible
to biases than others (Pronin and Kugler 2007). This points at a self-
serving over-confidence bias that makes us believe that we can separate
the various interests, including our personal interests, when taking
decisions, while in reality we cannot (Bazerman and Moore 2009;
Kahneman 2011). Protecting boards from such over-confidence seems
to be sensible. Nonetheless, the formalistic approach to independence is
highly problematic. The list of independence-disqualifying relationships
is never complete. Personal friendships are not included in the lists –
understandingly so, for how to describe friendship in precise regulatory
terms? – but they have a fundamental impact on the objectivity of non-
executive directors. And what to say of a banker whose bank does not yet
have a substantial financial relationship with the company, but would
like to have such relationship: would he truly be more independent and
objective in dealing with management proposals than the banker whose
bank already has such a relationship with the company? Furthermore,
7
See the independence requirements in, for example, the UK and Dutch Corporate Governance
Codes, the US Sarbanes-Oxley Act and the EU Commission’s Recommendation on the role
of non-executive or supervisory directors of listed companies of 15 February 2005
(2005/162/EC).
the more we insist on non-executive independence and exclude persons

with close relations to the company, the more non-executive members
become dependent on executive directors as the single source of infor-
mation about the business and the assessments of risks and opportuni-
ties. By insisting on non-executive independence, governance regulation
may actually have contributed to poorer monitoring by boards of finan-
cial institutions (Lorsch 2012; Winter 2012a). The Dutch corporate
governance code, insisting on all non-executives being formally inde-
pendent with the exception of a maximum one member,8 is perhaps the
pinnacle of this paradox of independence. Van Zijl (2012) has recently
reviewed 48 studies on the relation between independence of non-
executive directors and firm performance and concludes that on the
aggregate the findings are inconclusive. In light of the problematic
nature of independence requirements, this is not surprising.
Similar problems exist in trying to establish a link between non-
executive expertise and firm performance. Hau and Thum (2009) have
researched the relation between the financial expertise of non-executive
directors of German public and German private banks and the perform-
ance of these banks in the financial crisis. They show that state-owned
banks in Germany performed significantly worse in the financial crisis
than private banks. They also find a large competence gap between non-
executives of private banks and of state-owned banks with respect to
management experience and financial market competence. Both corre-
late significantly statistically. This appears to be a comfortable conclu-
sion: expertise of non-executives matters. The opposite conclusion,
expertise does not matter, certainly would have been more uncomfort-
able. But how, and how much precisely, does expertise of non-executives
matter? It seems likely that not only the expertise of non-executives of
German private banks exceeds that of their colleagues at state-owned
banks, but also the expertise of the executives and staff of private banks
exceeds those of public banks. Hau and Thum suggest that better non-
executives appoint better executives, but it seems likely that the eco-
nomic potential of private banks, including their ability to pay their staff
and executives more than public banks, has a much bigger effect on the
quality of executives and staff and their performance. Also, how likely is
it that non-executives can consistently and competently correct the
judgments of their executives? The fact that private banks also sustained
substantial losses in the crisis indicates that the higher level of expertise
8
Best Practice Provision III.2.1 Dutch Corporate Governance Code (2009).
of their non-executives was not enough to correct executive failures.

Finally, the policy conclusion of the research is to ensure board com-
petence, by permanent education9 and by extending fit and proper
requirements for board appointment to requirements of specific finan-
cial and risk expertise and knowledge.10
But what would we have taught non-executive directors in, say, 2005?
That a substantial crisis was in the making, revealing the fallacy of the
subprime business and unprecedented systemic risks? That an infectious
greed was taking over the industry and would soon show its destructive
force?11 If the herds of executives and even banking regulators are
insufficiently aware of such risks, how can we expect non-executive
directors to lead the right way? New regulatory admission tests for the
expertise of non-executive directors of financial institutions restrict the
pool of possible candidates for such functions to those who have explicit
financial institution experience. A potentially highly valuable addition of
outsiders to the board, who, for example, can bring in much wanted
expertise on how to build long-term trust-based client relationships, may
be practically impossible in light of expertise requirements.
Finally, research has focused on the effects of gender diversity on firm
performance. The outcomes are mixed. McKinsey (2012) reports that for
companies in the top quartile of executive board diversity, return on
equity (ROE) and earnings before interest and taxes (EBIT) were sig-
nificantly higher than for those in the bottom quartile return on equity
They acknowledge that their findings are not proof of a direct relation-
ship between diversity and financial success. At high-performing com-
panies, the board or the CEO may simply have greater latitude to pursue
diversity initiatives. Adams and Ferreira report that attendance rates go
up (female directors have better attendance rates than male directors and
male directors attend more when the board is more gender diverse) and
9
See, for example, the Dutch Banking Code (2009) s. 2.1.8, providing for permanent
education of non-executive directors in specific financial fields such as risk, financial
reporting and audit. Interestingly, a similar responsibility is imposed on executive
directors: s. 3.1.3. The CRD IV Directive will introduce further regulation in the field
of permanent education for non-executive directors of financial institutions, see Winter
(2012a).
10
As of 2012, the Dutch Central Bank, DNB, tests the expertise of executive and non-
executive directors prior to their appointment, art. 3:8 Wft.
11
In 2003, Frank Partnoy published his book, Infectious Greed; How Deceit and Risk
Corrupted the Financial Markets, not triggering much of a debate, let alone any regu-
latory response. It was a clear and chilling analysis of the state of the financial industry
five years before the financial crisis finally hit.
suggest that the monitoring effort is stronger in more diverse boards. But
overall, the gender effect on firm performance is negative (Adams and
Ferreira 2008). For gender diversity the same is true as for independence,
even if these factors in principle have a beneficial effect on the quality of the
debate and decision-taking, they are not absolute indicators of good per-
formance. They come at a cost, in the case of diversity (not only gender, but
also geographic, age, fields of expertise etc.) at the cost of cohesion. A board
composed of widely diverse members may find it more difficult to come to a
joint understanding of the company’s situation, and even if they think they
have such joint understanding, in fact they may not, due to different
interpretations of the same facts, figures and words.
The problem with this research is that it only looks at input variables
of board demographics and tries to establish a link between those
variables and the outcome variable of firm performance. The research
does not attempt to explain or reveal any relation between the input
variables and board performance as such, let alone between board
performance and firm performance.
The extent to which the performance of the firm can be linked to board
performance is hard to establish. We would assume and would certainly
hope that better board performance in the aggregate leads to better firm
performance. But we should also acknowledge that many other company
specific and external factors determine firm performance, some of which
are within the board’s control and many others are not.
Whatever the impact of board performance on firm performance, the
performance of the firm can certainly not solely, or even primarily, be
based on the potential quality of the monitoring role performed by the
non-executives, judged by board demographic input variables. Forbes
and Milliken (1999) refer to a wealth of research indicating that board
demographics–firm performance research must remain inconclusive if
the intermediating process of the operation of the board itself is not
researched. Agency theory-based understanding of boards, focusing only
on the monitoring role of non-executives and researching only input
variables of board demographics, provides much too narrow an
approach to understanding board performance. It takes the exclusive
perspective of the non-executive gatekeeper as an indicator of perform-
ance of the board and leaves out the primary contribution of the execu-
tives who are being monitored. Furthermore, it only takes the
monitoring perspective in a formal way, addressing the demographic
prerequisites for monitoring and not the substance of the matters to be
monitored. In short, the monitoring perspective only provides an
incomplete picture of board performance when the executive and non-

executive roles are not integrated. Only by developing an integrated
perspective exploring the interactive nature of these roles, can we begin
to see what it means for a board to perform well.
A further flaw in the exclusive agency theory-based approach to board
performance is that it leaves the human interaction in the board com-
pletely out of the equation. The board as a combination of executive and
non-executive directors is a social organisational structure allowing
people in different roles to co-operate, debate and take decisions in
order to achieve specific corporate objectives. The potential self-serving
behaviour of executives and the monitoring role of non-executives to
prevent this are part of this interaction, but they certainly do not provide
a complete representation of the human interaction within the board
(Levrau and Van den Berghe 2007). Both at the personal level and at the
level of the dynamics of the group as a whole, behaviour is driven by
factors that cannot be explained and cannot even be brought to the
attention by the agency theory. It is also essential to see that the mere
fact that regulation and economic theory (in short, the system) demand
and impose that certain roles are to be fulfilled by executives and non-
executives do not have the immediate and direct consequence that the
people who perform these roles actually fulfil them as expected. There is
an intermediate step between the demands and expectations of the
system and the actual behaviour of the board and of board members.
This step is the perception of the role, or the role idea that is generated
both at the level of individual board members and at that of the
group. These perceptions are not based solely on the role definitions as
laid down in regulation and elaborated upon in legal, economic and
governance debates, but are interpreted on the basis of personal experi-
ences, examples from others, world views and frames of reference that
each person and also groups develop over time (Erhard et al. 2011).
Together with other factors that determine behaviour at individual and
group level, these perceptions determine actual board performance (see
further Section 4).
3. An integrated approach to board performance: Board on

Task and Board GPS
To build a better understanding of board performance, we suggest
the concept of Board on Task. When we talk about performance, we
inherently assume a certain task that is to be performed. Lawyers and
economists have a limited view of this task of the board, and our model
intends to broaden the scope of the task of a board. Being on task
basically means a board doing the right things and doing them right.
Doing the right things requires clarity of the roles to be performed. In the
absence of clear guidance from the law, boards need to develop and
define their own role with precision (Lorsch 2012). As the non-executive
role by definition is derived from the executive role and only exists in
relation to the executive role, any comprehensive perspective on the role
of the board necessarily includes a view on both the executive role and
the non-executive role, as well as the connection between them. This is
what the board is about. As we will see in the next paragraph, clarifying
the various roles is not just a descriptive exercise. Organisational Role
Analysis (ORA) provides a helpful framework in order to identify how
the individuals on the board are connected to the organisation and each
other through their roles (Borwick 2006). Both executives and non-
executives take up and have interacting roles in the board where they
come together. Only by taking the perspectives of both can we develop an
integrated understanding of the task of boards and what this requires of
executives and non-executives.
Doing the right things right puts the focus on the actual behaviour of
board members in their roles. To a large extent, such behaviour is
determined by the role board members have or are perceived to have.
Additional perspectives are also generated by looking at board interac-
tion through three different lenses: the lens of the Group, the lens of the
Person performing the role and the lens of the System. We suggest using
the concept of Board GPS to look at the functioning of a board through
these three lenses. (1) The Group lens focuses on a series of group
dynamics, group processes and group phenomena that occur as a result
of the fact that the board is a group of individuals. The Group lens helps
to look at factors such as formal and informal leadership, cohesion,
information sharing, conflict resolution, reflection, biases and group-
think. (2) The Person lens focuses on individual characteristics and
person-related aspects, with an actual or potential significant impact
on their functioning as a board member. One may think of factors
such as personal styles and natural roles, skills, empathy, biases and
the need to look good. (3) The System lens focuses on the formal roles
and processes designed by general law and the rules and regulations of
the company itself. Examples of relevant factors at System level are:
board structure, board size, board composition, committees, decision
rules, conflicts of interests and liability.
Group
On
Task
Person System
Figure 5.1: The Group lens
For all three lenses we have defined Essences, Abilities and Traps that
further refine the views provided by each of the lenses. Essences are the
core features of a board; they define the make-up of a particular board.
Formal and informal leadership and cohesion, for example, are Essences
of a board seen through the Group lens. Essences as seen through the
Person lens include personal style and natural roles, and Essences
through the System lens include board structure and committees.
Abilities are the functional abilities as seen through the different lenses,
for example: information sharing, conflict resolution and reflection for
the Group lens, expertise, skills and empathy for the Person lens and
decision-making and managing conflicts of interests for the System lens.
If Essences and Abilities are not balanced, boards can fall into Traps such
as group biases and groupthink as seen through the Group lens, indi-
vidual biases and dominance by the need to look good as seen through
the Personal lens, and compliance attitude and liability as seen through
the System lens. Combining Board GPS with the ORA of the Board on
Task provides a richer, more comprehensive view of the complex inter-
actions that make up the board. They enable us to make well-founded
assessments of the performance of a board. See further, Figure 5.1.
4. Board on Task
In Section 2, we described how both the legal and the economic per-
spective of boards provide a limited view of the role of the board,
dominated by the perspective of the non-executives and emphasising
their monitoring role to a large extent focused on Hygiene factors such as
audit, control, risk management and compliance. Many organisational

theories distinguish two or three basic roles for the board: control,
service and sometimes also, strategy (Forbes and Milliken 1999; Levrau
and Van den Berghe 2007). Strategy is sometimes seen as part of the
control role, as the board is expected to oversee the strategy set by the
managers.12 Others see the role of the board in the field of strategy as part
of its service role (Langevoort 2000; Levrau and Van den Berghe 2007).
These organisational theories also primarily take the perspective of non-
executives to describe the roles of the board.
An organisational theory taking a different approach is the steward-
ship theory. This theory starts with the presumption that managers are
driven primarily by a need for achievement and recognition, intrinsic
satisfaction, respect for authority and ethics. In this theory the board
should facilitate and empower management and the CEO lead the
company, rather than monitor and control them (Donaldson and
Davis 1990). Boards should be composed primarily of executive direc-
tors (Donaldson 1990) and the focus of the board role in this theory is
therefore primarily on the executive function.
By taking either a predominantly non-executive or executive perspec-
tive, none of these theories provide a comprehensive view of the inter-
action between executives and non-executives. We believe that this
interactive nature of boards constitutes the essence and core of its
functioning. The basic distinction between control and service is insuf-
ficiently refined to describe this essence. The two notions indicate some-
thing about the nature and intensity of the interaction of executives and
non-executives, but not about the fields in which such interaction takes
place or is expected. This probably causes the confusion of the role of the
board in the field of strategy. Is that role part of the control function or of
the service function, or is it a role in itself? Strategy is a Field of
Interaction, but says little about whether the interaction is more moni-
toring/controlling or more advising/servicing. We develop our model of
Board on Task by first distinguishing and then combining Fields of
Interaction from and with Types of Involvement. Based on the current
12
US non-executive directors interviewed by Jay Lorsch believe their role in strategy is to
oversee management as they set strategy, i.e. strategy is part of the control role of the
board (Lorsch 2012). This is very similar to the approach taken by the Dutch Supreme
Court for supervisory directors in the Dutch two-tier model in recent cases: the manage-
ment board determines the strategy under the supervision of the supervisory board: HR
13 July 2007, JOR 2007, 178 (ABN AMRO) and HR 9 July 2010, JOR 2010, 228 (ASMI).
Again we see that one-tier and two-tier boards may not be far apart.
regulatory framework and board practices, there are five fields where
interaction between executives and non-executives is expected to take
place: Hygiene, Strategy, Performance, People and Stakeholders.
Hygiene is the field that is dominant in the regulation of boards. After
the governance crisis, regulation imposed stronger monitoring demands
on non-executives in the fields of audit, internal control, financial
reporting and compliance, a reflection of the perceived problems of the
scandals triggering the regulation. The financial crisis added risk and risk
management as an explicit further area of monitoring by non-executives.
To strengthen the involvement of non-executives in the hygiene field, it
is typically required or recommended that the board set up committees
with exclusively or predominantly non-executive members. The com-
mittee interacts extensively with executives on both a more detailed as
well as a much more principled level, requiring non-executives to gain a
better understanding of both the details of the issues as well as the way
these affect the running and performance of the business.
Strategy is the field where executives and non-executives need to deter-
mine the direction of the company, acknowledging and understanding the
uncertainty and complexity of the external environment of markets, com-
petitors, products, innovation, etc. in which the company operates. The
strong focus on hygiene in the past decade of governance reform and the
time that non-executives need to spend in this field has put pressure on and
constrained their ability to play a meaningful role in the field of strategy
(Lorsch 2012). Within strategy, incidental key strategic decisions may come
up, such as major acquisitions, divestments or a bid for the company. Those
decisions typically require an even more intensive interaction between
executives and non-executives.
Performance is the field where traditionally, and still today, the inter-
action between non-executives and executives is most prominent. The
natural execution of the monitoring or control role of non-executives is
to review the performance of the company on a regular basis. For many
boards this is the dominant field of interaction. Substantial parts of
board meetings are typically dedicated to this role. Often the interaction
is restricted to understanding and discussing performance as measured
by financial indicators.
People is the field where the non-executives take on the role of
employer. In this field, the remuneration of executives has received
most of the attention, both in regulation as well as in practice. Other
key elements, such as performance assessment and succession of execu-
tives, have received significantly less attention. Finally, this field includes
the responsibility to evaluate the interaction between executives and

non-executives, to review the performance of the board.
Stakeholders is the field dealing with the external parties who are key
stakeholders in the company. It typically includes shareholders. In some
companies, certain groups of other stakeholders are critical for the core
of the business (e.g. deposit holders of banks, insurance policy holders of
insurance companies). Stakeholders other than shareholders may also
become more important at different times, such as creditors when near-
ing insolvency, or under different corporate law and political models,
such as employees in co-determination models.
In these various Fields of Interaction between executives and non-
executives, the Type of Involvement of non-executives may be different.
Control and service divide the interaction into two broad categories.
These categories are no longer sufficient to describe the variety in the
nature and intensity of the interaction that results from modern regu-
lation and the development of the board practice. We provide a more
refined distinction by introducing the notions of Ratifying, Probing,
Engaging and Directing as distinct Types of Involvement of non-
executive directors.
Ratifying is the interaction where non-executives are asked to approve
the decisions prepared by executives. What is discussed between execu-
tives and non-executives is the outcome of the executive process of
analysis, deliberation and decision-making. Non-executives have not
been involved in any of the analysis, deliberation and decision-making
and are asked to bless it. The interaction is therefore limited, typically
scratches the surface and is mostly focused on the process followed to
reach the decision. Although a refusal of approval by non-executives is
theoretically possible, in practice this is exceptional and not expected.
Probing is the interaction which, in general, is now required by regula-
tion, whether hard or soft law, following from governance crisis. It is not
good enough for non-executives to have discussion with executives about
their plans and intended decisions and then to give their blessing. Non-
executives need to satisfy themselves that they have received all relevant
information and that the analysis made is proper. If not immediately
satisfied, they should test and challenge what is presented to them, seeking
additional information where needed in order to satisfy themselves of the
soundness of the decisions before agreeing to them.
Engaging is the interaction where the non-executives are seeking to
contribute beyond the probing and challenging of the analysis and
actions of executives. It requires even further in-depth knowledge and
Table 5.1. Matrix of board interaction

Type of Involvement
Fields of Interaction Ratifying Probing Engaging Directing
Hygiene
Strategy
Performance
People
Stakeholders
Focus Process Process Content Process and Content
understanding of (aspects of) the business of the company. It typically

assumes an involvement in a much earlier phase of the process that leads
to decision-taking. Servicing does not fully encapsulate this notion, as it
suggests a somewhat non-committal involvement of non-executives.
Engaging interaction cannot be seen as separate from the Probing or
sometimes Directing interaction.
Directing is the interaction where the non-executives are the owners of
both the process and of the decision to be made. Their role is no longer
derived from the role of non-executives, but becomes an independent
role. In the Field of People, the Directing role of non-executives directly
relates to executives. Non-executives are required to take this Directing
employer role, for example, when selecting and appointing a new CEO
and when setting remuneration or assessing the performance of execu-
tives. In other Fields such as Hygiene and Strategy, the Directing involve-
ment is likely to be more incidental. The notions of monitoring or
control are insufficient to cover such a Directing role of non-executives.
We can plot the five Fields of Interaction we have identified and the
four different Types of Involvement in a Matrix of Board Interaction in
Table 5.1.
Cross-hatching indicates the nature and intensity of involvement that
is minimally required following the governance and financial crisis. It is
the expectation embedded in regulation, without seeking to be specific
on the details of what is required and on whether it is legally enforceable
hard law or only best practices based on soft law. In four of the five Fields
of Interaction, Probing is generally the minimum Type of Involvement.
The governance and financial crisis and the regulation following from
them gave a strong signal that Ratifying is never good enough, in any of
the Fields of Interaction and under any circumstances. This represents a

major shift in expectation of the role of non-executive directors and
affects both the position and role execution of non-executives and of
executives. In one field (People), the minimum requirement is Directing,
non-executives owning the process and the decisions to be made. Non-
executives are not just monitoring remuneration decisions relating to
executives, but are expected take these remuneration decisions and to
develop the process leading up to them. The same applies to assessment
of executives and succession and to assessment of the performance of the
board as a whole: non-executives are expected to take full responsibility
for the process and content of the decisions.
Vertical lines indicate the fields where the interaction is occasionally
Directing. An example of this would be when a potential fraud is
reported by internal audit that potentially involves executive directors.
In such a case, non-executives need to take control of the decision to start
an investigation, secure the company’s operation in the relevant area
during the investigation and to draw conclusions from the investigation.
After decisions have been taken and implemented, in principle the
involvement should return from Directing to Probing, plus, potentially,
Engaging. Other examples of an occasional Directing involvement could
include a key strategic decision like a bid for the company where
executives may have a conflict of interest, or when a conflict has arisen
with shareholders where executives can no longer effectively engage in
dialogue. These situations may momentarily require the non-executives
to step in, control the process and the decision.
Shading indicates that the board has discretion. This relates particu-
larly to the Engaging interaction. Engaging is not a consistent or occa-
sional ‘must’ in any of the fields. Executives and non-executives need to
agree on how to ensure that an Engaging interaction can be effective and
truly adds value. This may not always involve all non-executive directors,
sometimes one or two non-executives in particular may be able to add
value in a specific field.
The Matrix does not provide a static description of the Fields of
Interaction and the Types of Involvement of non-executives. On the
contrary, the interaction is dynamic by nature, both from field to field,
from time to time and from company to company.
Starting with the Matrix of Board Interaction, a particularly valuable
framework in order to deepen our understanding of these board realities
is Organisational Role Analysis (ORA). ORA is a relatively new model,
conceiving the organisation as a system of interrelated tasks, roles and
role-holders (Newton et al. 2006). The concept of ‘role’ comprises the

‘place’ or ‘area’ that is the interface between a person and an organisa-
tion, or between personal and social systems’ (Sievers and Beumer 2006).
It represents a space impacted on the one hand by the organisation
and its definitions (tasks, other roles, system boundaries, resources,
etc.) and on the other hand by the way this specific person/role holder
fills and shapes this space, fuelled by the specific needs, aspirations,
values, attitudes and perceptions of that person. From this perspective
the ‘role’ is the place where the formal role (as defined by the organisa-
tion/system) blends with the informal role (the specific way a specific
person takes up his or her role). Crucial in all of this is that the role-
holder assumes what the role requirements are: what and how is the
primary task to be fulfilled? Individual role-holders construct implicit
and explicit task ideas. These task ideas to a large extent will determine
how an individual will take up his role. Taking up one’s role represents a
complex and interrelated configuration of interpretations: of the system,
of the tasks to be fulfilled, of one’s role, of oneself in one’s role and of
others in their roles. For a meaningful explanation of the functioning of a
system, it is important to map out how the various role-holders interpret
their own as well as one another’s roles. We may refer to role ideas. As a
consequence, for any group with a joint task or with interrelated tasks, it
is essential to become aware of both one’s own and of others’ role and
task ideas. This requires role awareness and role dialogues (Long et al.
2006). It is an alarming finding that while the complexities and inter-
dependencies in the financial-economic system have increased rapidly
over the last twenty years, this has not yet led to adequate and effective
dialogues between key actors and stakeholders about their respective
roles in that system (Van de Loo and Kemna 2012).
From the Matrix of Board Interaction we can develop role definitions
and distinguish role ideas from both the perspective of non-executives
and executives. For example, let us take the Probing Involvement.
Regardless of the specific Field of Interaction, the Probing Involvement
requires non-executives to have relevant information so that they can
really verify, challenge and probe. In order to be Probing, non-executives
need to review the information critically and to make efforts to under-
stand and prepare for a discussion on the basis of the information
received. The classic image of a non-executive director in the board
meeting opening the envelope with meeting documents does not reflect
the Probing interaction. If non-executive directors do not commit suffi-
cient time and care, what they are actually doing does not exceed
Ratifying. The role idea of non-executives does not, then, fit with the
role definition. Probing also involves a role definition for executives. A
Probing involvement of non-executives requires executives to facilitate
this, by providing timely and relevant information, and by providing
access to staff or external advisers, etc. In practice, executives may
reduce the real involvement of non-executives to no more than
Ratifying by practices such as: overloading non-executives with infor-
mation without clarifying what is crucial and what the key questions
or dilemmas are; providing information only very shortly prior to the
meeting, or even during the meeting; and providing new information
in the meeting and discussing this information rather than the infor-
mation provided before the meeting. These practices do not allow any
involvement of non-executives beyond Ratifying. The role idea and
role execution of executives frustrates the role idea and role execution
of non-executives.
Not only the Types of Involvement create specific role definitions and
role ideas that generate mutually interdependent behavioural patterns,
but the nature of the Fields of Interaction also has a distinct impact on
the interaction. For example, Hygiene, to a certain extent, requires a
critical, strict attitude of non-executives and a healthy suspicion. It also
requires a willingness at least sometimes to make the life of executives
difficult and not to feel inhibited to intervene when not satisfied with the
information or the performance. If non-executives are not up to being
tough when needed, they will not be able to be Probing in this field.
Similarly, executives will need to be able to deal with a critical and robust
Probing attitude of non-executives. If they perceive criticism as a failure
or a sign of distrust and develop a practice of delaying and hiding crucial
information from non-executives, they frustrate the role of non-
executives. In the field of Strategy on the other hand, the nature of the
interaction is more of a partnering kind, ensuring best choices are made.
Non-executives can improve the thinking process of the executives by
providing external perspectives and specific expertise. Lack of specific
company knowledge of non-executives may lead both executives and
non-executives to assume that non-executives cannot contribute to the
development of strategy in depth. Furthermore, non-executives contrib-
ute a different risk-reward–cost-opportunity sensitivity to the analysis.
Non-executives have little explicit incentives to take risk: success is often
regarded as the succes of executives, who are rewarded financially and
reputationally. However, failure quickly also reflects negatively on non-
executives, certainly reputationally and possibly with liability risks.
These factors have an impact on the different role ideas that non-
executives and executives may develop in the field of strategy.
These examples show how the concept of Board on Task, the distinc-
tion between Fields of Interactions and Types of Involvement, the Matrix
of Board Interaction and ORA applied to these concepts, can shed light
on board performance in practice.
5. Board reviews and board research

Corporate governance codes now typically require that the board reviews
its own performance regularly.13 Our concepts of Board on Task and the
Matrix of Board Interaction can be powerful tools to facilitate such
reviews. In our own experience of conducting board reviews, boards do
not regularly engage in a discussion on their task and the role definitions
that follow from it for executives and non-executives. It is even more
exceptional for a board to discuss the different role ideas of individual
board members. At the same time, we see that perceptions of task
and roles regularly differ widely or subtly, leading to suboptimal and,
sometimes, dysfunctional board behaviour. When tasks and roles as
such are not (and cannot be) discussed, discrepancies typically come to
the surface in discussions on the substance in any of the Fields of
Interaction. ORA suggests that a board, in order to be effective, should
regularly reflect on the board task, role definitions and individual role
ideas. Different perceptions of tasks and roles that are not revealed lead
to disconnected behaviours, which potentially undermine board
performance.
Executives and non-executives are encouraged to become aware of the
way they understand their own role, as well as the roles of others
(executives and non-executives) and joint interaction. What do they
see as their primary tasks, what do they see as their most important
contribution to make, and what actual behaviours are assumed to be
conditional in order to be effective? Our Matrix of Board Interaction
helps to facilitate and structure meaningful dialogues about the interplay
of task and role definitions and ideas, both within and between the
groups of executive and non-executive board members. In pursuing
this, it is important to beat in mind that the person-role-organisation
model asks for regular dialogues and exchanges, respecting the dynamic
13
Best Practice Provision III.1.7 Dutch Corporate Governance Code (2009) and S. B 6 of
the UK Corporate Governance Code (2012).
nature of board reality. Board reality encompasses elements of change

and stability, both at the level of persons (stable individual character-
istics and styles versus shifting feelings, opinions and behaviours) and at
the organisational or system level (for example, stable formal responsi-
bilities versus shifting lines of power, interest and authority).
As a consequence, roles and role-relationships need to be calibrated
regularly. ORA advocates that it is helpful also to differentiate between
the formal definitions and requirements of roles and tasks and the
experiences linked to these tasks and roles. In this perspective, boards
should see it as part of their ongoing work to develop the capacity to self-
reflect and engage in meaningful and effective dialogues about all of
these aspects. We would advise that this self-reflection is applied not
only at annual formal evaluation sessions. Ideally a board develops the
capacity to self-reflect on the way it is performing its task and executives
and non-executives perform their roles while it engages in its substantive
discussions. In a metaphor suggested by Heifetz and Linsky (2002), a
board should regularly be able to get on to the balcony in order to see
better what is happening on the dance floor. What dance is being danced,
who is dancing with whom, in what patterns, who is sitting out what kind
of dance. The ability to switch from active to reflective mode and back is
a core quality of leadership and it would serve boards well if they would
develop this ability.
Turning back to the topic of academic research, it seems that the
current economic and financial research on board performance suffers
from the observational bias called the ‘streetlight effect’. People tend to
search for lost keys under the lamp post, because there observations can
be more easily than in the dark. For economic and financial academics,
the lamp post is called ‘data’. In order to be able to apply the research
toolkit based on statistical analysis, data need to be processed and
correlated to other data. With respect to boards, typically the data
available to researchers are demographic input variables derived from
publicly disclosed information on the composition of the board and on
individual board members. As we have commented, this type of research
ignores the actual performance of boards and has little, if any, explan-
atory value when it comes to understanding board performance.
This is not to say that board performance cannot be researched; we
simply need to add different research methods. If we really want to
understand what makes boards perform and how, regardless of demo-
graphics, board performance can be frustrated, we need to observe and
study actual board behaviour. This is a challenge, as researchers mostly
do not witness board interactions first-hand. Actual board performance

occurs in a black box that cannot be observed by outsiders. Nonetheless,
there are various research methods that can provide us with valuable
insights into this black box of actual board behaviour. Executives and
non-executives can be interviewed and asked about their experience,
their perceptions of roles and the interaction between them. Such inter-
views can be qualitative, asking for board members’ views, perceptions
and impressions: they can also be much more factual and empirical,
asking for information about aspects of the board process and interac-
tion as they have been observed by board members. Board simulations
can be developed in which executives and non-executives (real, and
others such as students as control groups) play scripted roles and are
observed while doing so.
These research methods come with their own limitations, such as the
risk of self-serving responses in interviews and the subjective nature of
observations of behaviour, to name just a few. Nonetheless, such
research methods can provide relevant insights into what drives board
performance in practice. These types of research have in common that
they focus on the actual board process and interaction. For each type of
research, our concept of Board On Task, the Matrix of Board Interaction
and ORA may offer valuable notions and tools that contribute to better
and deeper understanding of the reality of board performance. By
approaching board performance through studying board process and
interaction with qualitative, empirical and circumstantial research tech-
niques, we can acquire a better understanding of board performance
than through precise and objective, but only partially relevant quantita-
tive data-driven research. Also applicable here, as Keynes said, ‘it is
better to be approximately right than to be precisely wrong’.
6. Policy implications
Our approach to board performance has several policy implications.
(1.) Regulation should avoid imposing strict and mandatory rules for
the composition of the board of directors. Factors such as inde-
pendence, expertise and diversity may be relevant for the function-
ing of the board, but there is no direct and clear relation between the
level of independence, expertise and diversity and actual board
performance. Research seeking to establish a relation between
board demographic input variables and output of firm performance
is mostly and necessarily inconclusive and cannot explain actual

board performance.
(2.) Imposing various mutually conflicting demographical requirements
on boards at the same time is even less convincing. Independence
comes at the cost of expertise. Insisting on specific expertise limits
the benefits of diversity in the board’s composition. Insisting on
diversity reduces expertise. Regulation cannot, for all companies
and under all circumstances, decide on the appropriate mix of
independence, expertise and diversity.
(3.) Regulation of board performance by demographic requirements is
not only impossible, but also reduces the board’s own responsibility
to ensure an appropriate composition in light of the size, complexity
and nature of the company’s business and to make it work in
practice.
(4.) Regulation can and should require boards to account for the way in
which they have exercised their responsibility for proper composi-
tion and proper performance, for example, through corporate gov-
ernance codes based on ‘comply or explain’.
(5.) In corporate governance codes, as in mandatory regulation, so far
the emphasis has been to strengthen the monitoring role of non-
executive directors. A more balanced approach in corporate govern-
ance codes to the task of boards would contribute to better board
performance.
(6.) Board evaluations are potentially powerful tools to improve board
performance. If one provision was to remain or become a manda-
tory requirement, it should be that boards should regularly evaluate
their own performance.
References
Adams, R. and D. Ferreira 2008. ‘Women in the boardroom and their impact on
governance and performance’, available at www.ssrn.com.
Assink, B.F. 2007. ‘Rechterlijke toetsing van bestuurlijk gedrag’, Series Instituut
voor Ondernemingsrecht, 59, Kluwer.
Bainbridge, S. 2002. ‘Why a board? Group decision making in corporate
governance’, Vanderbilt Law Review 55: 1–55.
Bazerman, M.H. and D. Moore 2009. Judgment in Managerial Decision Making,
7th edn., John Wiley & Sons.
Bebchuk, L. and J. Fried 2004. Pay Without Performance: The Unfulfilled Promise
of Executive Compensation. Cambridge, MA, Harvard University Press.
Blair, M. and L. Stout 1999. ‘A team production theory of corporate law’, Virginia
Law Review 85: 247–328.
Borwick, I. 2006. ‘Organizational Role Analysis: Managing Strategic Change in
Business Settings’, in John Newton, Susan Long and Burkard Sievers (eds.),
Coaching in Depth; The Organizational Role Analysis Approach. London,
Karnac, pp. 3–28.
Donaldson, L. 1990. ‘The ethereal hand: Organizational economics and manage-
ment theory’, Academy of Management Review 15: 377.
and J.H. Davis 1990. ‘Stewardship theory or agency theory: CEO governance
and shareholder returns’, Australian Journal of Management 16: 51.
Erhard, W., M. Jensen, S. Zaffron and K. Granger 2011. ‘Introductory Reading For
Being a Leader and The Effective Exercise of Leadership’: An Ontological/
Phenomenological Model, Harvard Business School NOM Research Paper
No. 10–091, available at www.ssrn.com.
Forbes, D.P. and F.J. Milliken 1999. ‘Cognition and corporate governance:
Understanding boards of directors as strategic decision-making groups’,
Academy of Management Review 24: 489–505.
Hau, H. and M. Thum 2009. ‘Subprime Crisis and Board (In)Competence: Private
vs. Public Banks in Germany’, ECGI Working Paper Series in Finance No.
247/2009, available at www.ssrn.com.
Heifetz, R.A. and M. Linsky 2002. Leadership on the Line: Staying Alive through
the Dangers of Leading. Harvard Business School Press.
Jensen, M. and W. Meckling 1976. ‘Theory of the firm: Managerial behavior,
agency costs and ownership structure’, Journal of Financial Economics 3:
305–60.
Jensen, M. and K. Murphy 2004. ‘Remuneration, Where We’ve Been, How We
Got There, What Are the Problems and How We Fix Them’, ECGI Finance
Working Paper No. 44/2004, available at www.ssrn.com.
Kahneman, D. 2011. Thinking, Fast and Slow, London, Allen Lane: pp. 261–4.
J.L. Knetsch and R.H. Thaler 1986. ‘Fairness and the assumptions of
economics’, The Journal of Business 59: 285–300.
Kraakman, R., D. Davies, H. Hansmann, G. Hertig, K. Hopt, H. Kanda and E. Bock
2009. The Anatomy of Corporate Law. Oxford University Press, 2nd edn.
Kronman, A.T. 1980. ‘Wealth maximization as a normative principle’, 9 Journal of
Legal Studies 227.
Langevoort, D. 2000. ‘The Human Nature of Corporate Boards: Law, Norms and
Unintended Consequences of Independence and Accountability’,
Georgetown University Law Center Business, Economics and Regulatory
Law Working Paper No. 241402, available at www.ssrn.com.
Levrau, A. and L. Van den Berghe 2007. ‘Corporate Governance and Board
Effectiveness: Beyond Formalism’, Vlerick Leuven Gent Working Paper
Series 2007/03, available at http://public.vlerick.com.
Long, S., J. Newton and J. Chapman 2006. ‘Role dialogue: Organizational Role
Analysis with pairs from the same organization’, in Newton, Longs and
Sievers (eds.), Coaching in Depth, pp. 95–112.
Lorcsh, J. 2012. ‘Boardroom challenges, lessons from the financial crisis and
beyond’, in Jay Lorsch (ed.), The Future of Boards: Meeting the
Governance Challenges of the Twenty-First Century, Harvard Business
Review Press, p. 13.
Masouros, P. 2012. ‘Corporate law and economic stagnation: How shareholder
value and short-termism contribute to the decline of the western
economies’, Ph.D. Leiden University, Eleven International Publishers,
pp. 112–14.
McKinsey Quarterly (April 2012), available at www.mckinseyquarterly.com/
Is_there_a_payoff_from_top_team_diversity_2954.
Newton, J., S. Long and B. Sievers (eds.) 2006. Coaching in Depth: The
Organizational Role Analysis Approach. London, Karnac.
Partnoy, F. 2003. Infectious Greed: How Deceit and Risk Corrupted the Financial
Markets. Times Books.
Posner, R.A. 2003. Economic Analysis of Law. New York, Aspen Publishers.
Pronin, E. and M. Kugler 2007. ‘Valuing thoughts, ignoring behavior: the intro-
spection illusion as a source of the bias blind spot’, Journal of Experimental
Social Psychology 43: 565–78.
Sievers, B. and U. Beumer 2006. ‘Organizational Role Analysis and consultation:
the organization as inner object’, In Newton, Long and Sievers (eds.),
Coaching in Depth, pp. 65–81.
Tamanah, B. 2006. Law as a Means to an End: Threat to the Rule of Law.
Cambridge University Press.
Tosi, H.L. and L.R. Gomez-Mejia 1994. ‘CEO remuneration monitoring and firm
performance’, Academy of Management Journal 37: 1002–123.
Van de Loo, E.L.H.M. and A. Kemna 2012. ‘Roles, risks and complexity: An
exploration of the triangle institutional investors, executive boards and
supervisory boards in the Netherlands’ in S. Long and B. Sievers (eds.),
Towards a Socioanalysis of Money, Finance and Capital(ism). London and
New York, Routledge, pp. 178–91.
Van Zijl, N.J.M. 2012. ‘The importance of board independence: A multidiscipli-
nary approach’, Series Instituut voor Ondernemingsrecht 90, Kluwer.
Winter, J. 2012a. ‘The Financial Crisis: Does Good Corporate Governance Matter
and How to Achieve It?’, in Eddy Wymeersch, Klaus J. Hopt and
Guido Ferrarini (eds.), Financial Regulation and Supervision: A Post-
Crisis Analysis. Oxford University Press, pp. 368–88.
2012b. ‘Corporate Governance Going Astray: Executive Remuneration Built to
Fail’, in Randall Thomas and Jennifer Hill (eds.), Research Handbook on
Executive Pay. Cheltenham, Edward Elgar, pp. 199–216.
6
Directors’ remuneration before and after

the crisis: measuring the impact of reforms
in Europe
roberto barontini, stefano bozzi,
guido ferrarini and maria-cristina
ungureanu
1. Introduction
In this chapter we measure the impact of recent reforms on directors’
remuneration by comparing the compensation practices at large
European listed companies before and after the recent financial crisis.
Our dataset is composed of the FTSE Eurofirst 300 Index constituent
companies, save for the adjustments indicated below. The firms included
in our sample are distributed across 16 European countries, of which 14
are in the EU. We analyse the data concerning directors’ remuneration at
these firms for the years 2007 and 2010, assuming that the changes
occurring between these two years reflect both the economic crisis
determined by the 2008 financial turmoil, and the remuneration and
corporate governance consequent reviews. Our analysis reveals system-
atic differences across countries. In fact, country-specific characteristics
such as corporate governance and the nature and quality of the legal
system have an impact on the agency problems between managers and
shareholders, and affect the level and structure of management pay
(Jensen and Meckling 1976; Fama 1980; Fama and Jensen 1983;
Ferrarini et al. 2009). Particular attention is dedicated to the financial
sector, due to the long-standing view that executive compensation in
financial institutions is, on average, higher than in other sectors, and to
the current pressure for reforming the compensation structure at these
institutions.
The present section briefly connects our work with previous studies in
this area. The following section introduces some core aspects of recent
251
252 r. barontini, s. bozzi, g. ferrarini, m.-c. ungureanu
EU and national reforms, the impact of which we intend to measure by

examining remuneration practices. In Section 3, we analyse the data
concerning remuneration governance and disclosure. In Section 4, we
analyse the data concerning pay structure and levels. Section 5 concludes
by advancing some policy suggestions.
Several papers show that the governance and disclosure of executive
compensation are impacted by factors related both to the institutional
setting of a given country and to firm-specific characteristics. Lang and
Lundholm (1993) provide empirical evidence that financial disclosure is
positively related to firm size and performance, as measured by earnings
and return variables. Chizema (2008) explores the determinants that led
German firms to behave differently as to the disclosure of individual pay
recommended by the German Code of Corporate Governance. His
results indicate that firm size and ownership types (institutional, dis-
persed, state) are correlated with disclosure. Muslu (2010) explores
whether the influence of the executives and insider directors on a
board affects remuneration disclosure, finding a positive relationship
with higher levels of disclosure, and with more performance-based pay
contracts. A similar effect is also related with investor protection, sug-
gesting that institutional settings have an influence on executive pay
disclosure.
In Section 3, we highlight a substantial improvement in remuneration
disclosure across the EU. However, the effects of better remuneration
disclosure on shareholder wealth are still intensely debated. Recent
reforms show that politicians and regulators share the view that an
increase in the level and quality of disclosure enhances shareholders’
scrutiny on compensation practices, reducing the power of executives
vis-à-vis the board, thus producing a more efficient pay contract.
However, the empirical evidence is mixed. Grinstein et al. (2011)
point out that, following the adoption by the Securities and Exchange
Commission (SEC) of new disclosure requirements for the non-wage
perquisites (perks) paid to executive officers of public corporations,
the firms that disclosed large perks for the first time subsequently
experienced a larger decrease in their level. This evidence is
consistent with the disciplinary role of the market in curbing large
perks and excessive compensation. Other studies support the idea
that disclosure of executive pay may be associated with the rise of the
level of executive compensation. Hayes and Schaefer (2009) theoretically
predict that CEO wages will be distorted upward, as a CEO’s wage
may serve as a signal of a firm’s performance and therefore affect
directors’ remuneration: impact of reforms 253
the value of the firm. Bizjak et al. (2008) find that the use of benchmark-
ing is widespread and has a significant impact on CEO compensation, a
practice that, in the presence of better disclosure, may lead to an esca-
lation in CEO pay. Similar results suggest that the impact of increased
disclosure on the overall amount of management pay is an empirical
issue, depending on several factors that may make the final result
difficult to predict.
In Section 3 we also focus on the existence and independence of
remuneration committees and remuneration consultants. Similar mech-
anisms increase the quality of corporate governance by allowing a more
intense scrutiny of management actions and limiting managers’ power to
shape their own pay (Bebchuk et al. 2002; Bebchuk and Fried 2003;
2004). Several studies find that better governance quality reduces man-
agerial opportunism (e.g. Andres and Vallelado 2008; Ahn and Choi
2009; Morey et al. 2009). Core et al. (1999) find that weak corporate
governance structures lead to excess compensation paid to CEOs which,
in turn, may negatively affect future firm performance, while Sun et al.
(2009) find that the relationship between future firm performance and
CEO stock option grants is positively affected by the quality of the
compensation committee.
Ownership structure is also likely to affect the quality of corporate
governance. Since Berle and Means (1932) and Jensen and Meckling
(1976), dispersed ownership is associated with agency costs affecting the
relationship between shareholders and management, while incentives
can be used to alleviate such costs. However, even management com-
pensation could be a source of agency costs, requiring the adoption of
specific mechanisms to realign management pay with shareholder inter-
ests (Bebchuk et al. 2002; Bebchuk and Fried 2004). Based on a similar
view, we expect European companies with more dispersed ownership to
limit the managerial opportunism related to incentive compensation
through better disclosure and a larger presence of independent control
mechanisms, such as remuneration committees and consultants. The
empirical evidence provided by Chizema (2008), even though limited to
a sample of German firms, provides support to this view by finding that
dispersed ownership is positively associated with disclosure of individual
compensation for management board members.
Concerning remuneration structure and levels (topics discussed in
Section 4) there is a large and growing number of papers focusing on pay
practices in individual European countries. A non-exhaustive list
includes: Conyon and Sadler (2010) for the UK; Drobetz et al. (2007)
for Switzerland; Rosenberg (2003) for Finland; Alcouffe and Alcouffe

(2000) for France; Elston and Goldberg (2003) and Chizema (2008) for
Germany; Ángel and Fumas (1997; 1998) for Spain; Brunello et al.
(2001), Barontini and Bozzi (2009) and Zattoni and Minichilli (2009)
for Italy; Vittanemi (1997), Ikäheimo et al. (2004) and Jones et al. (2006)
for Finland.
However, single-country studies offer little evidence on systematic pay
differences in executive compensation across different institutional settings.
Notable exceptions are Abowd and Bognanno (1995), who illustrate the
level and evolution of companies’ total compensation costs for CEOs, top
human resources directors and non-supervisory manufacturing employees
for twelve OECD countries over the 1984–92 period.1 Conyon and
Schwalbach (2000) compare UK and German levels and structures of
executive compensation. In an extensive work, Conyon et al. (2010) study
the determinants of the difference in the level of CEO pay in the US and
Europe, while Barontini and Bozzi (2011) and Croci et al. (2012) focus on
CEO pay across Continental European companies.
Section 4 offers further evidence on directors’ pay across Europe
focusing on the largest listed companies over the 2007–10 period. In
particular, we provide relevant information on the compensation poli-
cies adopted by listed companies as a response to the 2007 financial crisis
and to renewed pressure by regulators and the public opinion on firms’
pay structures and levels.
An issue that has attracted the attention of politicians and regulators in
recent years is the possibility that inappropriate incentives lead the execu-
tives of financial firms to take excessive risks in the management of their
institutions (Walker 2009). Indeed, the recent financial turmoil resurfaced
the question (already encountered in the corporate scandals which occurred
at the beginning of this century) as to whether high-powered incentives
could perversely induce the managers to inflate their firm’s share price in
the short term. The managers could, for instance, undertake high risk
projects, ultimately compromising their firm’s long-term value, for the
sole purpose of realising short-term gains. Coffee (2003; 2005) and
Bebchuk (2009) warn, in particular, against the risk that stock-based com-
pensation may create perverse incentives for the managers to engage in
short-term behaviour at the expense of long-term firm value.
1
The OECD countries included in the sample are: Belgium, Canada, France, Germany,
Italy, Japan, the Netherlands, Spain, Sweden, Switzerland, the United Kingdom, and the
United States.
This issue is particularly relevant for the financial sector, given the
systemic risk generated by the failure of a few large institutions; never-
theless, the empirical evidence in this regard is mixed. Chen et al. (2006)
analyse a sample of US banks over the 1992–2000 period finding that stock-
option-based wealth induces risk taking by bank managers. Kim et al.
(2011) provide empirical evidence that the sensitivity of a chief financial
officer’s option portfolio value to stock price is significantly and positively
related to the risk that the firm’s stock price will fall in the future, although
this relation does not hold for the CEO. However, as posited by Murphy
(2009), no systematic evidence is found that compensation structures have
been responsible for excessive risk-taking in the financial services industry.
Fahlenbrach and Stulz (2011) study the relation between CEO incentives
and bank performance during the crisis and find no evidence supporting
the view that larger risk exposures through option compensation are
responsible for the poor performance of banks.
Nonetheless, most post-crisis international reform initiatives primarily
target the limitation of risk in the banking sector, as we briefly show in
Section 2. Regulation considers flawed executive pay schemes as potentially
distortive with regard to managing the risk appetite and the overall stability
and long-term profitability of the company. The impact of remuneration
policies on risk is mainly addressed through requirements on the structure
of compensation, i.e. the relative weight of variable and fixed pay, as well as
that of stock and cash-based components within the pay package. The 2009
EC Recommendations and – specifically for the financial sector – the FSB
Principles and the Capital Requirements Directive (CRD III) explicitly call
for an adequate balance of these components within the compensation
package in order to avoid excessive risk taking.
2. Evolution of the EU regulatory framework

The main pre-crisis reforms on directors’ remuneration were adopted in
Europe in the years immediately following the Enron-type corporate scan-
dals. They were effected mainly through soft law, such as EU recommen-
dations and codes of best practice, seeking to enhance the ability of the firm
governance framework to produce appropriate remuneration and incentive
outcomes, while avoiding interference with compensation structures. The
European Commission identified remuneration as an area in which the
potential for conflict of interest was particularly high. Accordingly,
its 2004–05 Recommendations targeted internal firm governance, in
particular suggesting enhanced board independence, better disclosure
of remuneration policies (including explanation of how these align with

company performance), and mechanisms allowing shareholders to
express their voice on executive remuneration.2 Although enabled to
follow these Recommendations either through corporate law reform or
corporate governance codes subject to the ‘comply or explain’ principle,
Member States generally opted for the latter route.
The 2004–05 Recommendations did not engage with pay design,
although they implicitly supported performance-based pay. However,
following the outbreak of the financial crisis, EU reforms switched their
focus on pay structures, with particular emphasis on the financial sector.
The Commission 2009 Recommendations continued to promote the role
of governance and disclosure in the remuneration process, additionally
calling for appropriate pay structures to be aligned with corporate
sustainability.3 Moreover, the post-crisis policy discussion on remuner-
ation practices started to look at mandatory rules as a substitute for the
‘comply or explain’ approach to regulation. In particular, mandatory
disclosure and ‘say on pay’ were amongst the issues raised by the
European Commission in its 2011 Green Paper on the corporate govern-
ance framework.4 In the remainder of this section we focus on these new
features of executive pay, including the regulation of pay at financial
institutions, while referring to Chapter 1 for a more comprehensive
treatment of the whole subject.
2.1. From soft to mandatory regulation

Either anticipating or following the Commission’s proposals, Belgium,
Portugal, Spain, Italy and the UK were the first jurisdictions to consider
2
Commission Recommendation of 14 December 2004 fostering an appropriate regime for
the remuneration of directors of listed companies (2004/913/EC); Commission
Recommendation of 15 February 2005 on the role of non-executive or supervisory
directors of listed companies and on the committees of the (supervisory) board (2005/
162/EC).
3
Commission Recommendation on remuneration policies in the financial sector, C(2009)
3159, April 2009; Commission Recommendation of 30 April 2009 complementing
Recommendations 2004/913/EC and 2005/162/EC as regards the regime for the remu-
neration of directors of listed companies.
4
Commission Green Paper on the EU Corporate Governance Framework (COM(2011)
164). In fact, this followed a previous consultation in which the Commission had already
approached the issue of pay, calling for adjustments in financiers’ compensation.
Commission Green Paper on corporate governance in financial institutions and remu-
neration policies COM(2010) 284.
moving remuneration governance and disclosure into law. Belgium

adopted a law aimed primarily at reinforcing boards in listed compan-
ies,5 which lifted a number of the national Corporate Governance Code
provisions to the legislative level. As a result, the creation of
a remuneration committee became mandatory and the publication
of a corporate governance statement including a remuneration
report was required. The Portuguese market regulator issued a 2010
Regulation on Corporate Governance, which provides for mandatory
description of the remuneration policy and disclosure of individual
director remuneration.6 The regulation also requires firms to report
on the composition of the remuneration committee and the fact that
at least one of its members has knowledge and experience in remuner-
ation policy issues. In Spain, the Law on Sustainable Economy, in
effect since March 2011, delegated the Ministry of Economy and
Finance and the market supervisor (CNMV) to determine the structure
and content of companies’ remuneration report.7 The CNMV issued
a regulation requiring disclosure of remuneration in a standard
annual report format.8 Similarly, in 2011 in Italy the Securities
Commission (CONSOB) adopted new rules on transparency of remu-
neration,9 requiring uniform and detailed disclosure of compensation
practices and setting standard characteristics to be included in
the remuneration report. The regulation also makes provisions for
shareholder vote on both the previous year’s policy and the proposed
future policy.
The UK has traditionally had the most extensive set of governance
requirements in force with respect to executive compensation in Europe.
Listed companies have been required to prepare a Directors’
Remuneration Report since 2002, and to submit it to the advisory vote
of shareholders.10 Despite similar regulatory measures, during the recent
crisis, UK companies – banks in particular – raised serious concerns for
5
Law on the reinforcement of corporate governance in listed companies, April 2010,
Moniteur Belge 22709.
6
CMVM Regulation No. 1/2010 Corporate Governance.
7
Law No. 2/2011 of 4 March 2011 on Sustainable Economy (last amended by Law No.
2/2012 of 29 June 2012).
8
New section 61ter in the Securities Market Law (Ley del Mercado de Valores).
9
Article. 84-quater, Consob. Regulation 11971/1999 implementing Italian Legislative
Decree No. 58 of 24 February 1998, concerning the discipline of issuers.
10
The Directors’ Remuneration Report Regulations 2002, Schedule 7A of the Companies
Act 1985, re-enacted in Regulation 11 and Schedule 8 of the Large and Medium-Sized
Companies and Groups (Accounts and Reports) Regulations 2008.
what many observers considered as ‘excessive executive pay’ (Ferrarini

and Ungureanu 2010). This led the Government in 2011 and 2012 to
announce a reform directed to curb executive pay through greater
remuneration transparency, more shareholder powers and more diverse
board and remuneration committees.11
Other Member States have kept most of the requirements for remu-
neration governance and disclosure in corporate governance codes,
situation subject to change, however, given the swift trends at supra-
national levels.12
2.2. ‘Say on pay’

Pressure over dealing with ‘inappropriate’ executive compensation, be it
understood as either ‘excessive’ or misaligned with shareholder value,
have led to initiatives giving investors greater influence over executive
pay through a vote on companies’ remuneration policies and packages,
i.e. through the ‘say-on-pay’ process.
In the US, votes on pay are mandatory under the 2010 Dodd-Frank Wall
Street Reform and Consumer Protection Act, but the voting result is not
binding. Most European jurisdictions, in their governance codes, intro-
duced an advisory vote on the remuneration policy and a binding vote on
equity-based incentive schemes. Few regulators went further, enabling
binding votes on pay policy, in the hope that such votes would determine
corporations to be more conservative with respect to the total amount paid
to their executives and that this would be more driven by corporate
performance. The Netherlands, Sweden and Norway, however, had already
legislated binding say on pay before the crisis (Larcker et al. 2012).
11
The UK government proposals suggest, amongst others, a standardised form of remu-
neration report with the aim of making it simpler and easier to understand. Companies
should state a single figure for total pay of each director, for how much each executive
was paid in the previous year and what the maximum is that they could be paid in the
following year. See Department for Business Innovation and Skills (BIS), Department for
Business Innovation and Skills, Executive Remuneration: Discussion Paper (2011):
‘Enterprise and Regulatory Reform Bill’, announcement, November 2012, available at
www.bis.org.uk.
12
For example, at the time of writing, France and Germany are also considering new rules
on compensation for executives; for France, see ‘Conférence Sociale – Salaires : des
“convergences” entre partenaires sociaux, selon Moscovici (Source: AFP)’, Le Point,
10/07/2012, available at www.lepoint.fr; for Germany, see Mercer, Executive remuner-
ation podcast interview series, available at www.mercer.com.
Post-crisis reforms not only regard the nature of the vote (binding or
advisory), but also the possible shift of voting requirements from best
practice principles to legislation. Spain and Italy were among the first
countries to introduce a similar rule in their corporate laws during the
post-crisis reform initiatives,13 while France has extensively debated the
issue at government level.14 For a long time, in the UK the shareholder
advisory vote on executive compensation has been non-binding on
companies and their boards. Since spring 2012, however, the govern-
ment moved toward a binding regime through a range of proposals,
including: an annual binding vote on future remuneration policy; an
annual advisory vote on how the company’s pay policy was implemented
in the previous year; and a binding vote on ‘exit payments’ of more than
one year’s salary.
The effects of say on pay started to be felt soon after the launch of these
reforms. The case of UK companies failing to receive majority support
for their pay policies in 2011 could be considered representative for the
history of ‘say on pay’. For example, at the AGM of Barclays, over 25 per
cent of shareowners voted against the company’s pay plan at a very
tumultuous meeting.15 Shareholder discontent over pay also contributed
to the exit of insurer Aviva’s CEO, who resigned five days after 54 per
cent of shareowners voted against pay at the company’s annual meet-
ing.16 Over 40 per cent of WPP investors voted ‘no’ on pay, prompting
the company’s compensation committee chair to reach out to investors
prior to the upcoming AGM to defend a 30 per cent pay raise for the
company’s CEO.17 And there are several other examples.18
13
See supra notes 3 and 5, respectively.
14
Consultation sur la remuneration des dirigeants d’entreprise (2012), available at www.
tresor.economie.gouv.fr.
15
‘Barclays stunned by shareholder pay revolt’, 27 April 2012, BBC News, available at
www.bbc.co.uk/news/business.
16
‘Aviva rocked by shareholder rebellion over executive pay’, The Guardian, 3 May 2012,
available at www.guardian.co.uk/business.
17
‘WPP shareholders vote against £6.8m pay packet for Sir Martin Sorrell’, The Guardian,
13 June 2012, available at www.guardian.co.uk/media.
18
Just under half of shareowners at sports and online betting company William Hill voted
against the company’s plan to give CEO Ralph Topping a large pay increase and
retention bonus; see ‘William Hill wins pay vote by “short head”’, Reuters, 8 May
2012, available at www.uk.reuters.com/article. Investors at Cairn Energy set the bar
high for discontent over executive pay, with 67 per cent of shareowners voting
against the CEO’s pay package at the company’s annual meeting in mid-May; see
‘Cairn Energy faces shareholder rebellion over pay’, The Guardian, 17 May 2012,
available at www.guardian.co.uk/business. CEOs at AstraZeneca and Trinity Mirror
In the US, by comparison, the impact of the recent reform introducing

say on pay may be seen as modest. Amongst the Russell 3000 companies
with say-on-pay votes occurring between September 2011 and June 2012,
2.4 per cent failed to achieve shareholder support levels of 50 per cent or
higher. These results point to a slight rise in say-on-pay failure rates
compared to 2011, when 1.6 per cent of Russell 3000 companies failed
over the same time frame.19 In a sense, the 2011 proxy season could be
considered as successful in terms of the high number of companies having
their pay packages approved. A different view is that a non-negligible
number of companies experienced criticism of their remuneration policies.
Whilst shareholder activism through say on pay has no doubt influ-
enced the practice of remuneration, there is little that shareholders
realistically can do when companies facing a negative advisory vote
choose to disregard the same and keep their policy in place. In a similar
case, shareholders would have to resort to their power not to re-elect
directors, an option that is not available in companies where a control-
ling shareholder is present, however.
Table 6.1 presents the say-on-pay regulations in various jurisdictions.
2.3. Financial institutions

The international reform agenda for financial institutions has focused on
the link between executive pay and risk management and on how pay
can be used to align managers’ and stakeholders’ interests, including
those of governments as shareholders or creditors of state-supported
banks (Bhagat and Romano 2010; Ferrarini and Ungureanu 2010).
The FSB Principles for Sound Compensation Practices20 and their
Implementation Standards,21 endorsed by the G20 Leaders at their
Summits in London in April 2009 and Pittsburgh in September 2009,22
were asked to resign due to poor performance, but are included in the say-on-pay debate
because of generous exit packages which have been criticised by investors; see
‘AstraZeneca boss David Brennan quits under pressure from investors’, The Guardian,
26 April 2012, available at www.guardian.co.uk/business; see also ‘Trinity Mirror chief
executive Sly Bailey steps down’,The Guardian, 3 May 2012, available at: www.guardian.
co.uk/media.
19
Source: Institutional Shareholder Services publications, 2011 and 2012.
20
Financial Services Forum (FSF), Principles for Sound Compensation Practices (April
2009).
21
Financial Services Board (FSB), Principles for Sound Compensation Practices:
Implementation Standards (September 2009).
22
The 2009 meeting of G-20 leaders issued a report calling for ‘action to ensure that
governance of compensation is effective; that financial firms align their compensation
Table 6.1 Say-on-pay regulations in various jurisdictions
Remuneration
policy (ex ante)/ Changes
Remuneration report Binding/ Requirement (after 2010,
Country (ex post) Advisory (as at 2010) as at 2012)
Belgium Remuneration reportAdvisory Legislation
Denmark Remuneration policyBinding Legislation
France Remuneration policyAdvisory Code of
Corporate
Governance
Germany Remuneration report Advisory Voluntary Legislation
Italy Remuneration policy Advisory Code of Legislation
Corporate
Governance
Norway Remuneration policy Binding Legislation
Spain Remuneration report Advisory Code of Legislation
Corporate
Governance
Sweden Remuneration policy Binding Legislation Legislation
Switzerland Remuneration Report Advisory Voluntary Voluntary
Netherlands Remuneration policy Binding Legislation
UK Remuneration Report Advisory Legislation Binding/
Legislation
are addressed to ‘significant financial institutions’ which are considered,

more than others, in need of an internationally uniform regime. The FSB
principles are not new, to the extent that they address the issue of
balanced pay, the alignment of pay with performance, the independence
of the pay-setting process and the disclosure of remuneration policies.
The emphasis of the FSB standards on effective alignment of compensa-
tion with prudent risk taking is relatively new. This came as a reaction to
the debate against excessive bonuses, considered as one of the causes of
the excessive risk-taking by financial institutions which has contributed
to the market turmoil.
The EU implemented the international principles and standards
through specific amendments to CRD III, which took effect in
practices with prudent risk taking; and that compensation policies are subject to
effective supervisory oversight and engagement by stakeholders’.
January 2011.23 Implementation is currently monitored by the European

Banking Authority (EBA) and by the national supervisory authorities.
Across Europe, the CRD III has been implemented in distinctive ways,
either through amendments to existing laws and/or new regulations issued
by financial supervisors (Ferrarini and Ungureanu 2011a, b). Within each of
these approaches, there is variation in flexibility depending on whether
prescriptive rules or high-level principles have been adopted.
Table 6.2 outlines the main provisions regarding the structure of
remuneration, together with what we see as ‘opportunities’ and
‘threats’ that may be faced by banks complying with the same.
The proportionality principle is critical to the CRD III enforcement, to
the extent that it grants supervisors and institutions discretion in imple-
mentation/compliance with the new remuneration requirements.24 The
principle of proportionality has been applied differently in the Member
States, which may allow local differences to be recognised, but also make
the EU playing field unlevel (Ferrarini and Ungureanu, 2011c). For
example, the UK regulator applies the proportionality principle by
dividing institutions into four tiers, subject to decreasing minimum
expectations of compliance.25 In Luxembourg, credit institutions that
have total assets not exceeding €5 billion are allowed to neutralise the
requirements related to deferral, proportion of share-linked instru-
ments, retention policy and remuneration committee.26 In Italy, the
rules issued by the Bank of Italy require that deferral and share-based
instruments criteria be applied only by significant financial institutions
23
Directive 2010/76/EU of the European Parliament and of the Council of 24 November
2010 Amending Directives 2006/48/EC and 2006/49/EC As Regards Capital
Requirements for the Trading Book and for Re-Securitisations, and the Supervisory
Review of Remuneration Policies, Official Journal of the European Union L329/3.
24
CRD III, 4° Considerandum: ‘Principles should recognize that credit institutions and
investment firms may apply the provisions in different ways according to size, internal
organization and the nature, scope and complexity of their activities [. . .]’.
25
Tiers one and two contain credit institutions and broker dealers that engage in significant
banking activities; tier three includes small firms that may occasionally take risks; tier four
includes no-risk firms. Tier three and four firms may neutralise the rules on deferral,
performance adjustment and retained shares: Para. 1.14 and Chapter 3, Revised Code.
26
See Circular CSSF 11/505, Details relating to the application of the principle of propor-
tionality when establishing and applying remuneration policies that are consistent with
sound and effective risk management as laid down in Circulars CSSF 10/496 and CSSF
10/497 (‘CRD III Circulars’), transposing Directive 2010/76/EU of the European
Parliament and of the Council of 24 November 2010 amending Directives 2006/48/EC
and 2006/49/EC as regards capital requirements for the trading book and for re-
securitisations, and the supervisory review of remuneration policies (‘CRD III’).
Table 6.2 Remuneration characteristics and expected effect on

European firms
Mechanism Opportunities Threats

Proportionality principle Opportunity for To be sufficiently justified
institutions to neutraliseby each financial
some requirements institution
Appropriate balance Remunerate both the Moving from variable pay
between fixed and professional towards fixed salaries may
variable components responsibilities and the imply a potential
rendered performance reduction in the link
between compensation
and risk-adjusted
performance
40% to 60% (for Strengthen link between Could adversely affect
particularly high risk time horizon and the market competitiveness:
remuneration) of variable timeline for deferred more difficult to compete
compensation should be payouts for talent on a global basis
deferred over a period of
three-to-five years
50% of the variable Link with shareholder May generate incremental
remuneration component value compliance costs for
consists of shares or firms: makes the
share-linked instruments requirements operate in a
more complicated way.
Could adversely affect EU
market competitiveness:
more difficult to compete
for talent on a global basis
Malus and clawback Ensure link between Interaction of these
arrangements deferred compensation requirements with
and future performance existing principles of
Risk alignment process national law is often
complex
Guaranteed bonuses Avoid pay for failure Difficult hiring dynamics
banned (not >1 y,
occurring only for hiring
new staff)
with consolidated assets above €40 billion.27 Similar disparities may have
an impact on the competition amongst European banks and the market
for bank managers, to the extent that the latter is gaining an EU
dimension.
3. Pre- and post-crisis remuneration practices

Until recently, few papers explored European compensation practices
from an empirical perspective, mainly due to the difficulties in collecting
data, given the relevant differences amongst countries in terms of dis-
closure of individual pay to board members. Under the impulse provided
by the 2004–05 Commission Recommendations, European companies
started to reduce the information gap, offering researchers the oppor-
tunity to extend the analysis of compensation practices also to Europe.
In this section, we analyse firms’ remuneration practices before and
after the crisis, with particular emphasis on the governance process and
the quality of disclosure. Our analysis compares data for the years 2007
and 2010, thus providing evidence on the evolution of pay practices in
response to the recent financial crisis and to the remuneration reforms
adopted by European policymakers.
After describing our dataset and methodology (Section 3.1), we show
some descriptive statistics on the firms included in our sample
(Section 3.2). We then analyse the data concerning remuneration
governance and disclosure (Section 3.3) and focus specifically on
financial institutions (Section 3.4). We subsequently conduct a
multivariate analysis aimed at investigating the evolution of compliance
with the governance and disclosure criteria over the 2007–10 period, in
response to increased pressure from international regulators.
Observations are conducted on the evolution of the criteria for the
sample as a whole and on the level of compliance across countries
(Section 3.5).
3.1. Data and methodology

Our dataset is composed of the FTSE Eurofirst 300 Index constituents as
of 2007, i.e. the 300 largest companies ranked by market capitalisation in
27
See Bank of Italy, ‘Disposizioni in materia di politiche e prassi di remunerazione e
incentivazione, nelle banche e nei gruppi bancari’, Gazzetta n. 80 of 7 April 2011 (in
Italian).
the FTSE Developed Europe Index. Due to the existence of dual class
shares, the FTSE Eurofirst 300 Index for 2007 comprises 313 shares.
After removing 16 ‘B class’ shares, the sample is reduced to 297 firms.
Given that our analysis focuses on the comparison of compensation
practices between 2007 and 2010, we further adjust the sample in order
to have the same firms both in 2007 and in 2010. As a consequence, we
remove 18 firms that were delisted after 2007 due to mergers and
acquisitions, thus reducing the sample to 279 firms. According to the
criterion adopted for the selection of the sample, not all of them are still
included in the FTSE Eurofirst 300 Index in 2010.28
Our analysis of remuneration governance and disclosure therefore
covers a sample of 279 firms for both 2007 and 2010. However, our
analysis on the level and structure of executive compensation in
Section 4, excludes from our sample 34 firms that did not disclose
individual compensation of the CEO and each member of the board in
2007 or in 2010. As a result, the sub-sample used in Section 4 consists of
245 firms.29
Companies included in the sample are distributed across 16 European
countries, of which 14 are EU and 2 non-EU countries.30 Since the
number of available firms in seven countries is lower than 10, we
group them under the following acronyms: continental countries
(CONT) = Austria, Belgium, Denmark and the Netherlands; Nordic
countries (NORD) = Finland, Norway and Sweden; and (PIG) =
Portugal, Ireland and Greece.
With respect to the governance and disclosure of remuneration practices,
our analysis covers 15 criteria reflecting three areas: remuneration govern-
ance, disclosure of remuneration policy and individual disclosure of
director compensation. In setting these criteria we followed the provisions
28
Imposing the condition that a company should be included in the index in 2010 could
induce a sort of ‘survivor bias’ in our results, since it is more likely that badly performing
firms over the 2007–10 period have been dropped by the index. In our sample, the number of
firms included in the FTSE 300 Index in 2007 and excluded in 2010 is equal to 39.
29
It is not surprising that the percentage of firms deleted due to the lack of information on
individual compensation is higher in Portugal, Ireland, Greece (i.e. PIG) and Spain,
namely, those countries that, according to our analysis in the next section, show a lower
degree of compliance in terms of disclosure. It is worth noting that the number of firms
excluded from our sample is quite balanced between financial and non-financial.
30
EU: UK, France, Italy, Germany, Netherlands, Belgium, Spain, Sweden, Ireland, Austria,
Denmark, Portugal, Greece, Finland; non-EU: Switzerland, Norway. Swiss and
Norwegian firms have similar investor requirements to EU firms.
included in the 2004, 2005 and 2009 Commission Recommendations,

which are currently found in most international best practice guidelines.
Table 6.3 provides explanations for the evaluation of each criterion.
For each criterion we ascertain whether a firm attains a minimum level
of implementation. We assign a value of 1 to each criterion that a firm
complies with and 0 otherwise. If the firm does not provide information
on a criterion, we assign a missing value.31 We also introduce the variable
Y_all, which captures the average compliance with all of the governance
and disclosure of remuneration practices (computed as the simple aver-
age of all the dummy variables with non-missing values).
Data on directors’ characteristics and their compensation are hand-
collected from annual reports, or corporate governance reports if pub-
lished separately, for the two years. For each director we record the date
he/she entered (and, where applicable, left) the board, the type of his/her
relationship with the company (executive, non-executive, independent)
and membership of any board committee.
Compensation data include the single components of cash-based pay
(board fee, fixed pay, benefits, variable compensation, other compensa-
tion),32 and the characteristics of the director’s annual grants of stock
31
In order to consider the proportion of missing information for variables Y2, Y3 and Y4,
we conduct a separate analysis by introducing the additional dummy variables Y2bis,
Y3bis and Y4bis, defined as follows:
* Y2bis shows that ‘no information on the independence of remuneration committee is
disclosed’; it takes the values: 1 if no information on independence is present, but the

committee exists; 0 if information on independence (of any type) is present;
* Y3bis captures ‘lack of information on the existence of remuneration consultants’,
and takes the values: 1 if no information on remuneration consultants is disclosed; 0 if

information on the existence of remuneration consultants (of any type) is present;
* Y4bis captures ‘lack of information on the independence of remuneration consultants’
and takes the values: 1 if no information on independence of consultants is disclosed, but

the consultants exist; 0 if information on independence (of any type) is present.
For other variables with missing values (i.e. Y9, Y13, Y14 and Y15), we define similar
variables, which are not examined here in detail, for brevity.
32
This item includes forms of compensation with a very different nature, such as the
executive committee participation fee, the indemnity corresponded when the executive
leaves the firm, compensation granted by the participated companies, compensation
that the executives return to another company (generally the holding of the group) that
appointed him in the board, reimbursement of anticipated expenses, compensation for
consulting services provided to the firm by the member of the board and, in a smaller
number of observations, the compensation the executive perceives as an employee of the
company. In order to obtain a more consistent and comparable remuneration data, we
consider data on ‘other compensation’ net of termination payments and indemnities, as
well as of compensation received for salary or consulting services provided to the firm.
Table 6.3 Criteria describing the governance and disclosure of remuneration practices
Area Category Variable Criteria to research Explanations

Y1 Existence The existence of a remuneration committee set up within
the board, either separate or joined with other committee
Governance Remuneration Y2 Independence If RemCom is made up of all non-executive, majority
governance independent director
Y3 Existence Company making use of remuneration consultant
(should state if not)
Y4 Independence Remuneration consultant is independent of
management, i.e. does not work for management
Y5 Policy overview Description of the remuneration policy implemented in
the financial year in review
Y6 Forward-looking policy Overview of the remuneration policy for the following
financial year/subsequent years
Remuneration Y7 Fixed-variable Proportion between fixed and variable components
policy (state if no variable compensation)
Y8 Performance criteria for Financial/non-financial performance criteria applied
bonus the annual bonus (state if bonus is not awarded)
Y9 Performance criteria for Financial/non-financial performance criteria applied for
share plans the share-based remuneration (state if no share-based
pay is awarded)
Table 6.3 (cont.)
Area Category Variable Criteria to research Explanations
Disclosure Y10 Termination payments Information on the policy regarding termination

payments (state if no policy or no such payments)
Y11 Executive directors For each executive director, breakdown of each
component of annual compensation
Y12 Non-executive directors For each non-executive director, breakdown of each
component of annual compensation
Individual Y13 Granted Number of shares granted during the year in review
disclosure
Y14 Exercised Number of shares excercised during the year in review
Y15 Unexercised Number of shares unexercised/outstanding
and stock options (number, vesting period and vesting conditions, strike
price, any selling restrictions of shares granted, etc.) as well as the
portfolio – at the beginning and at the end of the two periods – of
stock and options granted to the director within incentive plans.
Data on firms’ financial characteristics are obtained mainly from
Datastream. When considering the ownership structure, we identify
the ultimate shareholder and the size of its voting rights according to
the standard methodology developed by La Porta et al. (1999).33
3.2. Summary statistics for firms included in the sample

The mean size of the companies in the sample, measured in terms of
Total Assets, is €130.6 billion in 2007. As shown in Table 6.4(a), country-
specific data reveal that company size differs, ranging from a mean of €49
billion for PIG to almost €169 billion for Switzerland. Data for 2010
confirm similar variability of firm size among countries. This circum-
stance is of particular interest for the aim of our study, given the
significant impact of size on executive compensation, as pointed out by
previous papers (Rosen 1982; Baker et al. 1988; Conyon et al. 2010).
Looking at ownership characteristics, we observe that they vary
among countries, with CONT, NORD and Italy having the larger pro-
portion of firms with concentrated ownership, compared with the UK,
PIG and Switzerland, where ownership is generally diffuse. Moreover,
financial institutions have a more dispersed ownership structure than
non-financial companies, as revealed by the comparison of the last
column of Tables 6.4(b) and 6.4(c).34 In the ‘industrial’ sample, only
the UK shows a proportion of firms with concentrated ownership well
below 50 per cent.
Data on leverage reveal that in countries like PIG, Italy and Spain
(which were most severely hit by the recent sovereign debt crisis), non-
financial companies’ debt is, on average, higher.
33
We detect the voting rights and the cash-flow rights held by the largest direct share-
holders, then trace the map of the ownership of the stakes, in order to identify the
ultimate shareholders and their ownership of voting and cash-flow rights. We use 20 per
cent as the cut-off point for the existence of a control chain; consequently, a listed
company with the largest shareholder holding a stake larger than 20 per cent is
considered closely held (CH = 1), otherwise widely held.
34
Only in France, CONT and NORD do more than 50 per cent of the financial firms in the
sample have a largest shareholder with a voting stake over the threshold of 20 per cent.
Table 6.4(a) Characteristics of the firms included in the sample for 2007 and 2010: whole sample
2007
Market Cap Total Assets Tobin’s 1-Year stock Dev. Std. stock
Country No. (mean in €.000) (mean in €.000) Leverage Q returns returns ROA CH
Switzerland 18 35,300,007 168,708,565 54.4% 2.16 4.0% 0.230 11.3% 44.4%
CONT 29 17,283,543 124,401,485 63.1% 1.70 10.9% 0.250 8.6% 62.1%
Germany 35 29,973,014 159,306,699 68.5% 1.64 31.0% 0.275 6.1% 54.3%
Spain 21 25,000,490 104,429,264 75.3% 2.06 6.2% 0.266 7.2% 57.1%
France 52 27,496,779 139,741,464 65.9% 1.51 9.9% 0.271 7.2% 57.7%
UK 61 32,981,655 154,493,393 64.7% 1.92 16.6% 0.268 10.6% 23.0%
Italy 18 28,284,929 160,075,927 77.1% 1.31 3.0% 0.211 5.7% 66.7%
NORD 27 20,964,263 58,071,630 59.9% 2.10 17.9% 0.292 10.3% 70.4%
PIG 18 11,515,167 49,306,489 81.2% 2.19 22.2% 0.286 7.6% 38.9%
Whole 279 26,648,168 130,611,530 66.9% 1.81 14.6% 0.265 8.4% 49.8%
sample
2010
Market Cap (mean Total Assets (mean Tobin’s 1-Year stock Dev. Std. stock
Country No. in €.000) in €.000) Leverage Q returns returns ROA CH
Switzerland 18 35,718,486 150,840,441 56.0% 2.02 12.9% 0.236 11.4% 44.4%
CONT 29 13,989,050 124,713,598 65.0% 1.45 18.8% 0.275 7.5% 62.1%
Germany 35 22,216,349 149,588,718 68.4% 1.27 14.7% 0.280 5.5% 54.3%
Spain 21 15,929,798 126,397,124 75.7% 1.46 −17.1% 0.333 5.9% 57.1%
France 52 20,763,117 155,992,161 64.9% 1.34 12.9% 0.292 5.7% 57.7%
UK 61 29,207,615 162,883,604 63.9% 1.60 30.9% 0.263 8.7% 23.0%
Italy 18 16,108,568 173,845,995 76.3% 1.12 −6.2% 0.279 4.5% 66.7%
NORD 27 17,703,361 71,449,232 59.7% 1.59 23.8% 0.285 8.0% 70.4%
PIG 18 4,890,472 52,685,488 78.6% 1.40 −19.3% 0.454 6.6% 38.9%
Whole 279 21,068,220 137,189,804 66.6% 1.46 13.2% 0.291 7.1% 49.8%
sample
Note: PIG=Portugal, Ireland, and Greece; CONT = Austria, Belgium, Denmark and Netherlands; NORD = Finland, Norway, and
Sweden. Market Cap and Total Asset are proxies for the size of the firm; Leverage is the ratio between total financial debt and book
value of equity; Tobin’s Q is the ratio between market and book value of the assets of the firm; 1-Year stock returns is the average
stock returns of the firm over the previous year; Dev. Std. stock returns is the standard deviation of stock returs over the previous
year; ROA is the ratio between the net income of the firm and total assets; CH is a dummy variable that takes the value 1 if the largest
shareholder owns more than 20% of voting rights.
Table 6.4(b) Characteristics of the firms included in the sample for 2007 and 2010: sample of financial firms
2007
Switzerland 4 39,891,147 682,750,206 95.2% 1.03 −14.7% 0.251 1.1% 0.0%
CONT 7 25,097,653 471,189,642 95.2% 1.04 −10.7% 0.239 1.0% 85.7%
Germany 6 32,009,685 692,303,733 95.9% 1.07 14.2% 0.263 1.1% 33.3%
Spain 6 32,385,363 288,084,365 92.7% 1.06 −6.9% 0.226 1.8% 16.7%
France 6 38,776,218 945,222,250 96.3% 1.01 −14.0% 0.275 0.5% 66.7%
UK 12 34,197,766 675,553,313 93.6% 1.06 −3.2% 0.291 1.3% 16.7%
Italy 7 32,971,401 346,690,194 91.7% 1.03 −6.2% 0.209 1.0% 14.3%
NORD 6 14,951,840 202,696,743 91.5% 1.04 −6.2% 0.233 1.4% 50.0%
PIG 8 12,643,254 97,679,939 93.6% 1.09 10.0% 0.287 1.4% 25.0%
Whole 62 28,811,384 485,710,992 93.8% 1.05 −3.4% 0.257 1.2% 33.9%
sample
2010
Market Cap (mean Total Assets (mean Tobin’s 1-Year stock Dev. Std.
Country No. in €.000) in €.000) Leverage Q returns stock returns ROA CH
Switzerland 4 31,957,523 582,547,451 94.4% 1.01 −1.9% 0.286 1.0% 0.0%
CONT 7 11,823,296 459,111,757 94.9% 0.99 2.7% 0.393 0.6% 85.7%
Germany 6 19,541,315 640,876,133 95.4% 1.00 −2.8% 0.266 0.7% 33.3%
Spain 6 19,789,985 356,194,714 91.9% 0.99 −24.6% 0.380 1.2% 16.7%
France 6 25,937,712 1,031,307,117 95.7% 0.99 −12.7% 0.412 0.5% 66.7%
UK 12 29,822,504 690,590,355 88.6% 1.04 14.1% 0.327 1.4% 16.7%
Italy 7 13,567,277 366,047,458 92.1% 0.96 −29.0% 0.342 0.4% 14.3%
NORD 6 17,003,032 254,044,293 91.3% 1.03 23.3% 0.258 1.1% 50.0%
PIG 8 2,455,245 102,808,662 93.6% 0.97 −52.6% 0.619 −0.3% 25.0%
Whole 62 18,979,167 498,554,700 92.6% 1.00 −8.8% 0.372 0.7% 33.9%
sample
Note: PIG = Portugal, Ireland, and Greece; CONT = Austria, Belgium, Denmark and Netherlands; NORD = Finland, Norway, and
value of equity; Tobin’s Q is the ratio between market and book value of the assets of the firm; 1-Year stock returns is the average stock
returns of the firm over the previous year; Dev. Std. stock returns is the standard deviation of stock returs over the previous year; ROA
is the ratio between the income generated by total assets of the firm and total assets; CH is the percentage of firms whose largest
shareholder owns more than 20% of voting rights.
Table 6.4(c) Characteristics of the firms included in the sample for 2007 and 2010: sample of non-financial firms
2007
Switzerland 14 33,988,252 21,839,525 45.0% 2.43 9.3% 0.224 13.7% 57.1%
CONT 22 14,797,235 14,059,799 52.9% 1.91 17.7% 0.253 11.0% 54.5%
Germany 29 29,551,634 49,031,451 62.9% 1.76 34.4% 0.277 7.2% 58.6%
Spain 15 22,046,540 30,967,224 68.3% 2.46 11.4% 0.282 9.4% 73.3%
France 46 26,025,548 34,678,753 61.9% 1.58 13.0% 0.271 8.0% 56.5%
UK 49 32,683,832 26,886,882 57.6% 2.13 21.5% 0.263 12.8% 24.5%
Italy 11 25,302,628 41,321,394 67.9% 1.49 8.8% 0.212 8.6% 100.0%
NORD 21 22,682,097 16,750,170 50.5% 2.42 24.8% 0.309 13.0% 76.2%
PIG 10 10,612,697 10,607,729 71.4% 3.06 32.0% 0.286 12.6% 50.0%
Whole 217 26,030,106 29,154,541 59.3% 2.02 19.7% 0.267 10.5% 54.4%
sample
2010
Market cap Total Assets Tobin’s 1-Year stock Dev. Std.
Country No. (mean in €.000) (mean in €.000) Leverage Q returns stock returns ROA CH
Switzerland 14 36,793,047 27,495,581 47.1% 2.25 17.2% 0.221 13.8% 57.1%
CONT 22 14,678,154 18,314,183 55.5% 1.59 23.9% 0.237 9.8% 54.5%
Germany 29 22,769,804 47,943,045 62.8% 1.32 18.3% 0.283 6.5% 58.6%
Spain 15 14,385,722 34,478,088 69.3% 1.65 −14.1% 0.314 7.8% 73.3%
France 46 20,088,170 41,820,645 60.9% 1.39 16.2% 0.276 6.4% 56.5%
UK 49 29,057,030 33,649,298 57.8% 1.74 35.0% 0.247 10.5% 24.5%
Italy 11 17,725,753 51,535,972 66.2% 1.23 9.7% 0.238 7.1% 100.0%
NORD 21 17,903,455 19,279,215 50.2% 1.75 24.0% 0.293 10.1% 76.2%
PIG 10 6,838,654 12,586,949 66.6% 1.73 7.2% 0.322 12.1% 50.0%
Whole sample 217 21,665,092 33,942,690 59.2% 1.59 19.5% 0.268 8.9% 54.4%
value of equity; Tobin’s Q is the ratio between market and book value of the assets of the firm; 1-Year stock returns is the average
stock returns of the firm over the previous year; Dev. Std. stock returns is the standard deviation of stock returs over the previous
year; ROA is the ratio between the income generated by total assets of the firm and total assets; CH is the percentage of firms whose
largest shareholder owns more than 20% of voting rights.
Some effects of the financial crisis are displayed through the compari-
son of the mean returns between 2007 and 2010: 1-year stock returns
decreased slightly on average from 14.6 per cent to 13.2 per cent, while
accounting returns (ROA) shrank to 7.1 per cent in 2010 from 8.4 per
cent in 2007. However, the impact of the crisis was not uniform across
countries and industries. In PIG, Italy and Spain, the effects of the crisis
were quite severe, with returns well below the mean of the sample, as
shown by the 2010 1-year stock returns, ranging from −6.2 per cent in
Italy to −19.3 per cent in PIG. Moreover, the financial sector was most
affected by the global crisis: Table 6.4(b) shows that, for financial com-
panies, mean 1-year stock returns decreased from −3.4 per cent in 2007
to −8.8 per cent in 2010, while for non-financial companies the returns
were about 19 per cent for both 2007 and 2010 (Table 6.4(c)).
The effects of the financial crisis are even more evident looking at
market capitalisation, which decreased on average by more than 20 per
cent (−34 per cent for the financial sample), and Tobin’s Q. Again, these
effects were much more severe for Spain and Italy (with a reduction in
market capitalisation of about 40 per cent) and PIG (close to 60 per cent),
in particular for financial companies.
3.3. Remuneration governance and disclosure

In this section we analyse the evolution of firms’ remuneration govern-
ance and disclosure characteristics. The following section 4 focuses on
the level and structure of directors’ compensation packages.
3.3.1. Country analysis

We first investigate the country-specific evolution of the 15 criteria on
remuneration governance and disclosure, as captured by the variables
Y1–Y15 described above. Next, we examine the evolution of each vari-
able in a multivariate setting, also controlling for certain company
characteristics (section 3.4).
Table 6.5 displays the breakdown by country of compliance with
individual criteria. The last column exhibits the variable Y_all, showing
the overall degree of compliance with governance and disclosure
requirements, measured as the simple average of all 15 criteria.
We observe that compliance with governance and disclosure charac-
teristics generally improved over time, although with clear variations
across jurisdictions. Germany, Italy and France show a statistically sig-
nificant increase; for Italy, we also register the largest improvement in
Table 6.5 Country-specific evolution of the 15 criteria on remuneration and governance characteristics
Year Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12 Y13 Y14 Y15 Y_All
Switzerland −2010 100.0% 18 100.0% 15 100.0% 14 100.0% 11 100.0% 18 11.1% 18 88.9% 18 76.5% 17 72.2% 18 88.9% 18 88.9% 18 94.4% 18 55.6% 18 44.4% 18 61.1% 18 73.3% 18
−2007 100.0% 18 100.0% 14 100.0% 14 100.0% 13 100.0% 18 5.6% 18 66.7% 18 61.1% 18 52.9% 17 72.2% 18 94.4% 18 94.4% 18 47.1% 17 16.7% 18 55.6% 18 65.9% 18
diff. 0.0% 0.0% 0.0% 0.0% 0.0% 5.6% 22.2% 15.4% 19.3% 16.7% −5.6% 0.0% 8.5% 27.8%* 5.6% 7.4%
CONT −2010 89.7% 29 84.6% 26 100.0% 10 100.0% 8 93.1% 29 48.3% 29 69.0% 29 62.1% 29 67.9% 28 86.2% 29 79.3% 29 82.8% 29 75.0% 28 71.4% 28 67.9% 28 68.0% 29
−2007 86.2% 29 68.0% 25 100.0% 6 100.0% 4 86.2% 29 37.9% 29 62.1% 29 62.1% 29 48.3% 29 51.7% 29 69.0% 29 69.0% 29 65.5% 29 48.3% 29 48.3% 29 55.2% 29
diff. 3.4% 16.6% 0.0% 0.0% 6.9% 10.3% 6.9% 0.0% 19.6% 34.5%*** 10.3% 13.8% 9.5% 23.2%* 19.6% 12.9%*
Germany −2010 57.1% 35 18.8% 16 100.0% 11 100.0% 9 100.0% 35 14.3% 35 45.7% 35 74.3% 35 63.6% 33 94.3% 35 94.3% 35 97.1% 35 51.5% 33 18.2% 33 21.2% 33 52.6% 35
−2007 40.0% 35 27.3% 11 100.0% 3 100.0% 2 100.0% 35 2.9% 35 20.0% 35 80.0% 35 58.8% 34 74.3% 35 91.4% 35 88.6% 35 50.0% 34 11.8% 34 8.8% 34 43.0% 35
diff. 17.1% −8.5% 0.0% 0.0% 0.0% 11.4%* 25.7%** −5.7% 4.8% 20.0%** 2.9% 8.6% 1.5% 6.4% 12.4% 9.5%***
Spain −2010 95.2% 21 70.0% 20 73.3% 15 100.0% 2 85.7% 21 19.0% 21 14.3% 21 14.3% 21 35.0% 20 47.6% 21 28.6% 21 57.1% 21 25.0% 20 15.0% 20 15.0% 20 38.4% 21
−2007 90.5% 21 47.4% 19 42.9% 14 0 76.2% 21 9.5% 21 0.0% 21 9.5% 21 15.0% 20 52.4% 21 23.8% 21 52.4% 21 15.0% 20 5.0% 20 5.0% 20 28.3% 21
diff. 4.8% 22.6% 30.5% 100.0% 9.5% 9.5% 14.3%* 4.8% 20.0% −4.8% 4.8% 4.8% 10.0% 10.0% 10.0% 10.2%
France −2010 96.2% 52 73.5% 49 100.0% 4 100.0% 2 96.2% 52 5.8% 52 70.6% 51 74.5% 51 52.2% 46 94.2% 52 100.0% 52 100.0% 52 97.8% 46 87.0% 46 68.9% 45 65.6% 52
−2007 96.2% 52 72.3% 47 100.0% 1 100.0% 2 94.2% 52 7.7% 52 51.9% 52 70.6% 51 43.5% 46 55.8% 52 98.1% 52 98.1% 52 93.5% 46 56.5% 46 15.2% 45 55.1% 52
diff. 0.0% 1.1% 0.0% 0.0% 1.9% −1.9% 18.7%* 3.9% 8.7% 38.5%*** 1.9% 1.9% 4.3% 30.4%*** 53.7%*** 10.5%***
Italy −2010 100.0% 18 83.3% 18 100.0% 4 100.0% 2 94.4% 18 33.3% 18 22.2% 18 44.4% 18 66.7% 15 77.8% 18 77.8% 18 83.3% 18 80.0% 15 80.0% 15 80.0% 15 60.4% 18
−2007 88.9% 18 81.3% 16 0 0 88.9% 18 5.6% 18 16.7% 18 27.8% 18 46.7% 15 22.2% 18 77.8% 18 77.8% 18 53.3% 15 40.0% 15 40.0% 15 41.9% 18
diff. 11.1% 2.1% 100.0% 100.0% 5.6% 27.8%** 5.6% 16.7% 20.0% 55.6%*** 0.0% 5.6% 26.7% 40.0%** 40.0%** 18.5%***
Table 6.5 (cont.)
Year Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12 Y13 Y14 Y15 Y_All
NORD −2010 96.3% 27 84.0% 25 100.0% 4 100.0% 4 96.3% 27 40.7% 27 77.8% 27 51.9% 27 77.8% 27 96.3% 27 74.1% 27 88.9% 27 29.6% 27 18.5% 27 22.2% 27 58.5% 27
−2007 96.3% 27 75.0% 24 100.0% 3 100.0% 2 96.3% 27 14.8% 27 70.4% 27 38.5% 26 64.0% 25 92.6% 27 70.4% 27 88.9% 27 20.0% 25 12.0% 25 20.0% 25 50.6% 27
diff. 0.0% 9.0% 0.0% 0.0% 0.0% 25.9%** 7.4% 13.4% 13.8% 3.7% 3.7% 0.0% 9.6% 6.5% 2.2% 7.9%
PIG −2010 88.9% 18 68.8% 16 100.0% 3 66.7% 3 72.2% 18 0.0% 18 33.3% 18 33.3% 15 30.8% 13 50.0% 18 50.0% 18 55.6% 18 28.6% 14 28.6% 14 28.6% 14 37.0% 18
−2007 88.9% 18 56.3% 16 100.0% 1 100.0% 1 55.6% 18 0.0% 18 22.2% 18 33.3% 18 35.7% 14 33.3% 18 27.8% 18 38.9% 18 28.6% 14 28.6% 14 30.4% 14 30.4% 18
diff. 0.0% 12.5% 0.0% −33.3% 16.7% 0.0% 11.1% 0.0% −4.9% 16.7% 22.2% 16.7% 0.0% 0.0% 0.0% 6.7%
UK −2010 100.0% 61 100.0% 61 100.0% 61 100.0% 60 100.0% 61 88.5% 61 96.7% 61 98.4% 61 100.0% 61 100.0% 61 100.0% 61 100.0% 61 100.0% 61 100.0% 61 100.0% 61 98.8% 61
−2007 100.0% 61 100.0% 61 100.0% 61 100.0% 60 100.0% 61 72.1% 61 98.4% 61 95.1% 61 95.1% 61 93.4% 61 98.4% 61 98.4% 61 95.1% 61 93.4% 61 93.4% 61 95.4% 61
diff. 0.0% 0.0% 0.0% 0.0% 0.0% 16.4%** −1.6% 3.3% 4.9%* 6.6%** 1.6% 1.6% 4.9%* 6.6%** 6.6%** 3.4%**
ALL −2010 91.4% 279 80.5% 246 96.8% 126 99.0% 101 95.0% 279 35.5% 279 65.1% 278 67.5% 274 69.0% 261 87.1% 279 83.9% 279 89.2% 279 69.8% 262 60.7% 262 59.0% 261 67.1%
−2007 87.8% 279 76.4% 233 92.2% 103 100.0% 84 91.8% 279 24.4% 279 53.8% 279 62.8% 277 58.2% 261 66.7% 279 79.9% 279 84.2% 279 63.2% 261 45.0% 262 40.8% 262 58.2%
diff. 3.6% 4.1% 4.6% −1.0% 3.2% 11.1%*** 11.3%*** 4.7% 10.7%** 20.4%*** 3.9% 5.0%* 6.6% 15.6%*** 18.2%*** 8.9%***
Note: PIG = Portugal, Ireland, and Greece; CONT = Austria, Belgium, Denmark and Netherlands;. NORD = Finland, Norway, and Sweden.Remuneration governance (variables Y1–Y4): Y1=Existence of a remuneration committee; Y2=If RemCom
is made up of all non-executive, majority independent director; Y3=Company making use of remuneration consultant (should state if not); Y4=Remuneration consultant is independent of management; Disclosure of remuneration policy (variables
Y5–Y10): Y5=Description of the remuneration policy implemented in the financial year in review. Y6=Overview of the remuneration policy for the following financial year/subsequent years; Y7=Proportion between fixed and variable components;
Y8=Financial/non-financial performance criteria applied the annual bonus; Y9=Financial/non-financial performance criteria applied for the share-based remuneration; Y10=Information on the policy regarding termination payments. Disclosure
of individual remuneration (variables Y11–Y15): Y11=For each executive director, breakdown of each component of annual compensation; Y12=For each non-executive director, breakdown of each component of annual compensation;
Y13=Number of shares granted during the year in review; Y14=Number of shares exercised during the year in review; Y15=Number of shares unexercised/outstanding; Y_All is the average of all the variables Y1–Y15 with no missing values. Column
Y_All contains the results of OLS regression analysis between overall compliance with remuneration and governance criteria and firms characteristics.
*,**, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
overall compliance. Moreover, as was the case before the crisis, the
highest level of overall compliance is found in the UK, Switzerland and
some of the CONT.35 Countries with low levels of compliance in 2007
(Spain and PIG) did not significantly increase their average score.
3.3.2. Descriptive statistics on remuneration governance

Following the 2005 Recommendation, most corporate governance codes
already endorsed the setting-up of a remuneration committee with a
majority of independent directors before the financial crisis. Only
the German Corporate Governance Code did not specifically
recommend the formation of a remuneration committee. Moreover,
the 2009 German Act on the Appropriateness of Management Board
Remuneration marked a departure from European corporate law and
practice by requiring that the full supervisory board decide on individual
management board pay (including salary and incentive-based pay).36
Reflecting this regulatory framework, our analysis, summarised in
Table 6.5, finds a widespread recourse to the remuneration committee
(proxied by variable Y1), both before and after the crisis, in all countries
except Germany, where only about half of the companies in our sample
established this committee, with a slight increase in 2010. Again with the
exception of German firms, the independence criterion (variable Y2) is
fulfilled by most of the companies having a remuneration committee
(the average value is about 80 per cent, with a small increase from 2007).
However, several compensation committees still do not fulfil the com-
position criteria established by either best practice or regulation (i.e. all
non-executive members, with a majority of independent directors).37
Requirements concerning the presence and role of compensation
consultants in continental European countries are weak compared to
35
CONT countries show a large average increase in compliance, equal to 13 per cent,
although statistically significant only at the 10 per cent level. On the other side, UK
companies present a limited but statistically highly significant increase, since in 2007 the
compliance levels were already close to maximum.
36
Act on the Appropriateness of Management Board Remuneration (‘VorstAG’),
September 2009.
37
In Spain, France, and PIG, about 30 per cent of the board does not fulfil independence
requirements, although this proportion is lower than in 2007. In contrast, for Germany,
committee independence reduces in 2010 (from 27 per cent to 19 per cent). In the UK
and Switzerland, where the requirement, according to the Corporate Governance Code,
is for all non-executive members of the remuneration committee to be independent,
there is full compliance.
the UK (Conyon 2011).38 In the UK, the Directors’ Remuneration

Report Regulations of 2002 require that firms disclose consultant infor-
mation. Although the Commission supports the presence and
independence of remuneration consultants in its 2004 and 2009
Recommendations, strong disclosure requirements are not found in
Continental Europe, where the relevant provisions are rather patchy.
Our analysis shows that all UK companies have used a third-party
consultant to advise them on compensation levels and design since
before the crisis (Y3). Furthermore, all UK firms in our sample make a
statement regarding their independence, i.e. non-engagement in other
consulting services for the management (Y4). In the other jurisdictions,
the presence of an external consultant is usually not disclosed.39 Similar
disclosure gaps bar an accurate review of the remuneration governance
process.
Overall, compliance with the remuneration governance variables
(Y1–Y4) improved slightly after the crisis, with a notable variation
amongst countries and sometimes amongst the variables examined.
3.3.3. Descriptive statistics on disclosure

Current disclosure criteria require remuneration statements to be clear
and easily understandable, to provide detail on the alignment between
pay and performance and to be transparent about the individual
directors’ compensation packages. However, significant differences
existed amongst national jurisdictions as to pay disclosure before the
financial crisis. In this section we find that disclosure of remuneration
generally improved post-crisis in all jurisdictions, even though the levels
of compliance with the variables describing the companies’ approach
to disclosure (Y5–Y15) vary greatly across countries. We show that
compliance with the remuneration statement requirement (variable
Y5) was quite strong across jurisdictions before the crisis and improved
38
Furthermore, results of the variable Y2bis show that in only two countries (Germany
and Switzerland) 20 per cent of the firms do not disclose information on the independ-
ence of remuneration committee, even where the committee exists.
39
Variable Y3bis, not reported in Table 6.5, shows that in 55 per cent of the sample no
information on remuneration consultants is disclosed in 2010 (63 per cent in 2007); if we
exclude the UK firms from the analysis, these figures increase to 70 per cent and 81 per
cent, respectively. It is worth noting that, in 2010, Germany and Italy significantly
reduced the proportion of firms not disclosing information on remuneration consul-
tants, showing some convergence towards best practices. Furthermore, even where such
information exists, in more than a third of the cases no information on the independence
status (Y4bis) is disclosed.
post-crisis at the few non-compliant firms. However, these high levels of

compliance may be explained by the generic nature of the relevant
requirement.
As previously noted, UK companies already followed high standards
of disclosure and governance of the remuneration process prior to the
crisis, reflecting the mandatory regime in place since 2002. As a result,
the UK levels of compliance with the relevant criteria for both reference
periods are generally close to maximum. In 2007, the ‘forward-looking
policy’ requirement (Y6) was the only area getting lower support at UK
firms. However, the situation significantly improved afterwards, also
in anticipation of new legislation announced by the government.40 A
significant increase is also found for Italy and NORD countries.
Nevertheless, the average compliance with the requirement at issue for
the whole sample is still limited (close to 35 per cent in 2010, or about 21
per cent if the UK is excluded).
All of the remaining variables for the remuneration policy show a
significant increase in compliance for the whole sample, with the excep-
tion of Y8 (performance criteria for bonuses). Looking at individual
countries, German companies significantly improved their approach to
disclosing the proportion between fixed and variable components (Y7),
while Switzerland, France and NORD reached compliance levels close to
80 per cent. With regard to the disclosure of performance criteria for
share plans (Y9), our results show a widespread increase in compliance,
even though not particularly strong at the individual country level.41
Moreover, we detect a remarkable increase in disclosure of termination
payments (Y10) for Italian, French, German and CONT companies.42
Disclosure of termination payments shows the largest increase in com-
pliance following the crisis (20.4 per cent, from 66.7 per cent in 2007 to
40
The UK government proposed legislation for a binding vote on the future policy. See
previous section.
41
The largest increases, close to 20 per cent, are detected in Switzerland, continental
countries, Spain and Italy.
42
The rise in compliance levels for termination payments (Y10) experienced by the CONT
group is also related to the Austrian and Belgian firms. In the case of Austria, the rise is
primarily supported by better disclosure and improved individual disclosure of
directors’ compensation (also for variable Y11 and Y12 which will be discussed later),
in anticipation of further regulations on disclosure. The Belgian case is supported by the
introduction of the new law on corporate governance that has particular focus on the
issue of remuneration. Accordingly, Belgian firms are now obliged by law to set up a
remuneration committee, majority independent, and to provide clear disclosure of the
individual pay components, particularly for board members (see previous section).
87.1 per cent in 2010). Severance payments were heavily criticised by

shareholders during the crisis, who revolted against their often excessive
levels not sustained by performance. Reacting to public discontent,
international and national regulations or best practices established
caps on termination payments (normally one- or two- year compensa-
tion) and required higher transparency of the relevant policy.
As to the variables concerning individual disclosure (Y11–Y15),
increased compliance (even though not statistically significant) is
observed for the two variables capturing disclosure of the annual com-
pensation components for executive and non-executive directors (Y11
and Y12). On average, disclosure of individual share schemes (Y13–Y15)
is lower, although some countries (in particular the UK, Italy and
France) show significant improvements. However, the jurisdictions
where remuneration disclosure and governance standards were lower
prior to the crisis (as in the case of Greek and Portuguese financial
institutions) did not show substantial improvements post-crisis.
These findings show that the firms’ approach to compliance is strongly
dependent on their home country’s approach to regulation and govern-
ance culture. Firms generally tend to reflect the way in which EU
regulations are implemented at the national level – either through
mandatory legislative requirements or best practice guidelines – and
the level of detail in the formulation of the relevant standards.
An area where differences across EU countries are still relevant is the
role of investors in the remuneration process. As discussed in the
previous section, some EU countries recommend a vote (either advisory
or binding) on the remuneration policy (ex ante) and/or the remuner-
ation report (ex post), while other Member States have made the share-
holder vote on pay mandatory.
Overall, standards of governance and disclosure in the field of remu-
neration in large listed firms have improved all over Europe. However,
differences persist across countries, reflecting the different regulatory
approaches particularly concerning the description of performance-
related characteristics of the incentives (Y8 and Y9) and the disclosure
of individual share schemes (Y13–Y15).
3.4. Financial institutions

The importance of the financial sector and its regulatory overhaul in the
wake of the financial crisis makes a comparison between financial and non-
financial firms particularly relevant. In fact, our analysis on disclosure and
2007
1
0.8
0.6
0.4
0.2
0
Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12 Y13 Y14 Y15
Industrial Financial
Figure 6.1(a): Compliance in 2007. Financial vs non-financial companies
2010
1
0.8
0.6
0.4
0.2
0
Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12 Y13 Y14 Y15
Industrial Financial
Figure 6.1(b): Compliance in 2010. Financial vs non-financial companies
governance does not show major variations between financial and non-
financial firms in the pre-crisis period (see Figure 6.1(a)).43
Considering all of our 15 variables, both categories of firms show
an overall progress over time.44 However, progress at financial insti-
tutions is slightly more evident (see Figure 6.1(b)).45 In particular,
disclosure of the remuneration policy for subsequent years (Y6),
disclosure of the proportion of fixed versus variable pay (Y7) and
performance criteria for share plans (Y9) exhibit a significant increase
for financial firms.
These results may be driven largely by those financial firms which
received state aid during the crisis. In fact, the institutions that resorted
43
Among all the variables considered, only Y4bis (lack of information on the independ-
ence of remuneration consultants) is significantly higher in financial firms (25 per cent
vs. 9 per cent).
44
The average compliance to governance and disclosure criteria (variable Y_all) increases
significantly for both financial and non-financial firms. However, the average increase
for financial firms is larger (12.5 per cent vs. 7.9 per cent).
45
In 2010, financial firms also fill the gap in information on the independence of remu-
neration consultants, in contrast with non-financial firms.
to state rescue programmes were forced by the governments to change

their remuneration policies, substantially reducing their payout to execu-
tives (Ferrarini and Ungureanu 2010). Furthermore, different stakeholders
(investors, regulators and the general public) requested a better balance
between fixed and variable pay. Subsequent reforms introduced limits on
bankers’ variable compensation and a shift in focus from short-term to
long-term compensation. The increased disclosure of performance criteria
may also be interpreted as a reaction to heavy criticism against excessive pay
not clearly related to performance.
3.5 Multivariate analysis on remuneration

governance and disclosure
In this section we analyse the evolution of compliance with the 15 criteria
over the 2007–10 period in response to increased pressure from govern-
ments and regulators. We conduct observations on the evolution of the 15
criteria for the sample as a whole and for the level of compliance across
countries. We also consider the institutional settings and company charac-
teristics that could influence conformity with the criteria at issue.
For each variable Yn (with n = 1 . . . 15), we run multiple logistic
regressions, based on the following general model:
ð1Þ ProbðY n Þ ¼ α þ β1 Year2010 þ β2 ½Country þ β3 ½Control þ ε
where prob (Yn) is the conditional probability of firm compliance with

respect to the criterion Yn.
[Year 2010] is a dummy variable that takes the value 1 if the data refers
to the year 2010, and 0 otherwise; this variable captures the different
probability to be compliant with the criteria in 2010, with respect to
2007.
[Country] is a set of dummy variables used as proxies for differences in
the level of investor protection; these variables capture the different
probability to be compliant with the criteria with respect to the UK,
which is used as the reference point for other countries.
[Control] is a set of variables representing the firms’ financial charac-
teristics, such as:
– Financial, a dummy variable that takes the value 1 for financial firms,
and 0 otherwise;
– firm-specific variables for financial characteristics: Log Assets (the log
of Total Assets); P_BV, the price-to-book ratio, is the ratio between
market and book value of firms’ stock; Tobin’s Q is the ratio between
market and book value of the assets of the firm; 3Y_Ret is the average
stock returns over the last three years; 3Y_ROA is the average account-
ing returns over the last three years;
– CH, a dummy variable that takes the value 1 if the ownership of the
first shareholder is higher than 20 per cent, 0 otherwise; this variable is
thus a proxy for concentrated ownership.
A logistic regression is applied to the data gathered on the 15 criteria,
related to remuneration governance, disclosure of remuneration policy
and disclosure of individual director compensation.46 This approach
investigates whether the dependent variable is significantly related to
explanatory variables (such as the year, size and performance of the
company, the country of reference and ownership concentration) in a
multivariate setting. According to this framework, a positive and signifi-
cant coefficient related to an independent variable – for example, firm
size – means that a larger company has a higher probability of being
compliant with the criteria.
Table 6.6 reports the results for a set of regressions that relates each
governance or disclosure variable (Y1 . . . Y15) – as well as the variable
Y_all, showing the overall degree of compliance with governance and
disclosure requirements – to the independent variables explained above.
Year 2010, capturing the variations in compliance across countries for
each of the 15 criteria in the post-crisis period is positive and statistically
significant for several variables, showing an overall progress over time.
However, the degree of compliance varies with respect to each of the
criteria considered. Compliance with remuneration governance stand-
ards, particularly the remuneration committee presence and composi-
tion (variables Y1–Y4) does not portray a relevant progress. This may be
due to national regulations already being in place before the crisis, and
therefore showing no significant changes in the sample period. As for
disclosure of remuneration, our results show some improvements,
mostly related to share-based plans, therefore revealing an attention to
aspects that called for specific reforms.
46
This econometric analysis can be used when the target variable is categorical, as it may
assume only two values (compliance with a specific governance or disclosure character-
istic Y=1; no compliance Y=0).
Table 6.6 Governance, disclosure variables and firms’ characteristics
Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12 Y13 Y14 Y15 Y_All
Intercept 13.102 −2.914 0.371 −2.622 6.337 −8.236 −5.167** −5.278** −8.870*** −4.836** −4.197 −9.729*** −8.984*** −4.495* −4.283* −0.355***
Y 2010 0.238 0.127 1.080 −2.942 0.255 0.437** 0.418*** 0.056 0.411*** 0.798*** 0.102 −0.044 0.194 0.582*** 0.734*** 0.071***
Switzerland 6.998 7.327 3.400 −0.450 8.153 0.020 1.144*** 0.487 0.119 0.295 0.722 0.957 −0.705** −0.791** 0.571 −0.245***
CONT −5.488 −4.921 3.850 4.420 −4.213 2.883 0.618** 0.222 0.008 −0.323 −0.610* −0.972*** 0.619** 1.001*** 0.949*** −0.303***
Germany −7.960 −8.174 7.084 0.968 7.842 0.261 −1.053*** 0.717** −0.245 0.471 0.749 0.342 −0.820*** −1.917*** −1.955*** −0.476***
Spain −4.067 −5.866 −10.816 0.306 −4.847 0.852 −2.879*** −2.535*** −1.795*** −1.393*** −3.068*** −2.365*** −2.091*** −2.077*** −2.234*** −0.597***
France −4.153 −5.443 2.094 −5.006 −3.554 0.134 0.244 0.462* −0.843*** −0.167 2.814*** 2.348** 2.467*** 1.292*** −0.120 −0.346***
Italy −4.803 −4.599 −4.570 −2.405 −4.159 1.185 −1.733*** −1.084*** −0.389 −1.508*** −0.561 −1.069** 0.247 0.974** 1.048*** −0.424***
NORD −3.393 −4.560 2.780 8.050 −2.757 2.225 1.119*** −0.342 0.812** 2.096*** −0.547 0.417 −1.507*** −1.383* −0.986*** −0.356***
PIG −5.277 −5.214 −8.876 −8.698 −5.919 −11.895 −0.980** −0.929** −0.985** −1.675*** −2.394*** −2.480*** −1.305*** −0.804** −0.984** −0.586***
Log Assets −0.222 0.567*** 0.513 1.378 0.065 0.318** 0.339*** 0.366*** 0.588*** 0.401*** 0.393*** 0.780*** 0.594*** 0.274* 0.259* 0.037***
Financial −0.225 −2.017** 10.934 −11.031 −0.522 0.252 −0.621 −0.975** −1.002** −0.571 −0.863 −2.091*** −2.325*** −1.146** −0.416 −0.099***
P BV −0.111 −0.126 0.465 −0.023 −0.022 0.031 0.001 0.108 0.137 0.000 0.001 0.004 0.013 0.003 0.009 0.001
Tobin’s Q −0.239 0.376 −1.610 1.170 −0.153 0.121 −0.161 −0.170 −0.261 −0.132 0.192 0.193 0.405*** 0.393*** 0.333** 0.001
3Y Ret 0.692 −0.345 −2.420 −8.753 0.063 −0.549 0.300 −0.295 0.672 0.119 −0.409 −1.375* −0.582 −0.271 −0.202 −0.026
3Y ROA −9.214* −0.545 4.312 2.771 −5.621 −0.296 3.029 −2.846 −1.943 1.661 −4.334 −7.195** −3.082 −4.401 −2.885 −0.211
CH −0.803** −0.726** 2.613 −10.401 −0.419 −1.085*** −0.548** −0.500** −0.802*** −0.795*** −1.023*** −1.093*** −1.220** −1.334*** −1.635** −0.121***
Note: PIG = Portugal, Ireland, and Greece; CONT = Austria, Belgium, Denmark and Netherlands; NORD = Finland, Norway, and Sweden. Remuneration governance (variables Y1–Y4):
Y1=Existence of a remuneration committee; Y2=If RemCom is made up of all non-executive, majority independent director; Y3=Company making use of remuneration consultant (should state if
not); Y4=Remuneration consultant is independent of management; Disclosure of remuneration policy (variables Y5–Y10): Y5=Description of the remuneration policy implemented in the financial
year in review. Y6=Overview of the remuneration policy for the following financial year/subsequent years; Y7=Proportion between fixed and variable components; Y8=Financial/non-financial
performance criteria applied the annual bonus; Y9=Financial/non-financial performance criteria applied for the share-based remuneration; Y10=Information on the policy regarding termination
payments. Disclosure of individual remuneration (variables Y11–Y15): Y11=For each executive director, breakdown of each component of annual compensation; Y12=For each non-executive
director, breakdown of each component of annual compensation; Y13=Number of shares granted during the year in review; Y14=Number of shares excerised during the year in review;
Y15=Number of shares unexercised/outstanding; Y_All is the average of all the variables Y1–Y15 with no missing values; Log Assets is the Log of Total assets; Financial is a dummy variable equal to
1 if the ultimate shareholder (at a cut-off of 10%) is a financial institution, and 0 otherwise; P_BV is the ratio between market and book value of firms’ stock; Tobin’s q is the ratio between market and
book value of the assets of the firm; 3Y_Ret is the average stock returns over the last three years; 3Y_ROA is the average accounting returns over the last three years; CH is a dummy variable equal to
1 if the first largest shareholders owns more than 20% of voting rights and 0 otherwise. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
3.5.1 The evolution across jurisdictions

An independent and clearly defined decision-making process, including
the setting up of a remuneration committee with a majority of inde-
pendent directors and the recourse to independent external consultants,
was already required or recommended before the crisis, either by regu-
lation or by standards of best practice. As a result, a majority of firms had
established a remuneration committee (Y1).47 However, not all of them
met the composition requirement (all non-executive members, with a
majority of independent directors). As indicated by the positive coeffi-
cient on 2010 for variable Y2, compliance with the remuneration com-
mittee requirements increased after the crisis, yet not significantly. Our
results also reveal that independence of the remuneration committee is
positively related to firm size, as indicated by the significant coefficient
on Log Assets for variable Y2, while both the presence of the committee
and its independence are negatively related to the ownership character-
istics of the company (variable CH on Y1 and Y2 variables). In other
terms, both the presence and the independence of the remuneration
committee are more frequent in firms with dispersed ownership.
Companies were already required before the crisis to issue a remu-
neration policy statement covering their board members and key execu-
tives and explaining the relationship between remuneration and
performance (with an emphasis on the long-term interests of the com-
pany). In the wake of the financial crisis, pressure has increased on firms
to show that their compensation policies are sound, thereby facilitating
constructive engagement with stakeholders and diluting potentially
harmful ‘outrage’ effects. Our results show some progress in remuner-
ation disclosure, as well as in terms of overall compliance, as confirmed
by the positive and significant coefficient on variable Y_all.
Compliance with disclosure standards varies remarkably across coun-
tries. This outcome may be driven by the fact that similar requirements
are generally established by corporate governance codes, which are
adopted by firms on a ‘comply or explain’ basis. The emphasis on each
of these requirements varies across Member States, leading to variations
in compliance across jurisdictions, as argued in the previous section.
Variable Y5 (presence of a remuneration policy in the report) already
displays a high degree of compliance for 2007. A slight improvement can be
47
Either separate or joined with nomination committees; for a comprehensive description
of the regulations on remuneration committees adopted in various jurisdictions, see
Ferrarini et al. (2009).
observed across countries over the 2007–10 period: nearly all firms that had
not reported anything on their pay policy pre-crisis became compliant post-
crisis. A higher increase in compliance levels is observed with regard to the
following disclosure variables: Y6 (future remuneration policy); Y7 (propor-
tion between fixed and variable pay); Y9 (performance criteria for share
plans); Y10 (policy for termination payments); Y14–Y15 (individual disclo-
sure of share-based compensation).
The recent reforms focus on a forward-looking approach to remuner-
ation, which should be reflected in the structure of pay and disclosed in
the firm’s remuneration policy. The policy should also balance the fixed
and variable components of pay, in addition to explaining the perform-
ance criteria chosen for short-term and long-term incentives. A similar
approach aims to align the firm’s compensation policy with its overall
performance, strategy and operational environment.
Our results show general progress in the disclosure of a forward-
looking pay policy (Y6), no substantial variation in the disclosure of
performance measures for bonuses (Y8) and some improvement in the
disclosure of the performance criteria for stock-based incentives (Y9).
The disclosure of termination arrangements (variable Y10) shows rele-
vant improvement in the reference period, possibly in response to
pressures from both investors and regulators. Disclosure of individual
compensation shows significant progress only for share-based plans,
particularly with regard to disclosure of exercised and unexercised
options (variables Y14–Y15). Similar results show that companies strive
to achieve a better alignment of remuneration with their long-term
objectives and shareholder interests.
Disclosure of the remuneration of executives on an individual basis is
not significantly better in 2010, while that of non-executives’ pay is
slightly worse, albeit with no statistical significance (variables Y11 and
Y12, respectively). This result is mainly due to the majority of firms
already providing reasonable disclosure of individual amounts before the
crisis.
Disclosure of individual director compensation reflects, inter alia, the
applicable national regulations, as shown by the different coefficients
associated with country variables. Overall, our results show that individ-
ual pay disclosure improved mainly with regard to share-based plans, in
some jurisdictions (like Belgium, Spain and Italy) anticipating subse-
quent legislative reforms. Where individual disclosure is poor, firms do
not provide appropriate explanations for their approach, mainly
referring – when they explain – to privacy issues.
Our analysis shows that, after controlling for the variations in firm
characteristics over the sample period, the crisis and the subsequent
reforms did not have a uniform impact on the evolution of remuneration
practices: variables related to remuneration policy and disclosure were
affected more than variables related to remuneration governance. The
analysis in this section also highlights that jurisdictions react differently,
leading to significant variations in compliance across countries.
3.5.2. Company characteristic factors

Table 6.6 shows the influence of relevant firm characteristics on pay govern-
ance and disclosure. Firm characteristics are related to the overall degree of
compliance with the 15 criteria considered in the analysis. As to firm own-
ership, we observe some significant differences in approach between con-
centrated and widely-held firms, with the latter usually being more
compliant with almost all criteria, as highlighted by the negative and statisti-
cally significant coefficients on CH for almost all of the specifications.
Further insights are provided by Table 6.7, reporting the difference in
compliance between widely-held (WH) and closely-held (CH) firms for
each criterion.
These results indicate that firms with dispersed ownership provide
better disclosure both pre- and post-crisis. This is true for all variables
considered. As to the presence of remuneration committees and external
consultants, significant differences are observed only for the pre-crisis
period, when CH firms were less compliant. This gap tends to shrink in
2010, even though a remuneration committee is still significantly less
frequent in CH firms.
From Table 6.6 it also emerges that most remuneration governance
and disclosure variables are positively related to firm size, while market
valuation (variable P_BV) does not seem to affect the results. Where size
has an influence, the coefficient is always positive, implying that larger
firms are typically more compliant with most criteria. The business
sector (financial vs. non-financial) also has an influence on compliance
with some of the requirements.
These results suggest that, across Europe, companies with more dis-
persed ownership try to limit the agency costs deriving from the sepa-
ration of ownership and control through better compliance with best
practice remuneration governance and disclosure. These results confirm,
and extend to Europe, theoretical predictions and previous country-
specific empirical evidence about the impact of ownership concentration
on remuneration governance and disclosure.
Table 6.7 Governance, disclosure variables and firms’ ownership characteristics
Year Type Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12 Y13 Y14 Y15 Y_All
2010 WH 0.950 140 0.827 133 0.976 85 1.000 72 0.971 140 0.493 140 0.743 140 0.761 138 0.770 135 0.907 140 0.900 140 0.929 140 0.809 136 0.735 136 0.770 135 0.768 140
2007 WH 0.921 140 0.835 127 0.959 73 1.000 64 0.936 140 0.357 140 0.629 140 0.727 139 0.731 134 0.779 140 0.871 140 0.914 140 0.754 134 0.607 135 0.578 135 0.694 140
dff. 0.029 −0.008 0.018 0.000 0.036 0.136** 0.114** 0.034 0.039 0.129*** 0.029 0.014 0.055 0.128** 0.193*** 0.074**
2010 CH 0.878 139 0.779 113 0.951 41 0.966 29 0.928 139 0.216 139 0.558 138 0.588 136 0.603 126 0.835 139 0.777 139 0.856 139 0.579 126 0.468 126 0.397 126 0.573 139
2007 CH 0.835 139 0.679 106 0.833 30 1.000 20 0.899 139 0.129 139 0.446 139 0.529 138 0.425 127 0.554 139 0.727 139 0.770 139 0.504 127 0.283 127 0.228 127 0.470 139
dff. 0.043 0.100* 0.118 −0.034 0.029 0.086* 0.112* 0.059 0.178*** 0.281*** 0.050 0.086* 0.075 0.185*** 0.168*** 0.103***
2010 WH-CH 0.072** 0.048 0.025 0.034 0.043* 0.277*** 0.185*** 0.173*** 0.167*** 0.073* 0.123*** 0.072* 0.229*** 0.267*** 0.374*** 0.195***
2007 WH-CH 0.087** 0.155*** 0.126** − 0.036 0.228*** 0.183*** 0.198*** 0.306*** 0.225*** 0.145*** 0.145*** 0.250*** 0.324*** 0.349*** 0.224***
Note: CH indicates closely held firms (ownership of the largest shareholders larger than 20%); WH indicates widely held firms (ownership of the largest shareholders smaller than 20%). Remuneration governance (variables Y1–Y4):
Y1=Existence of a remuneration committee; Y2=If RemCom is made up of all non-executive, majority independent director; Y3=Company making use of remuneration consultant (should state if not); Y4=Remuneration consultant is
independent of management; Y5=Description of the remuneration policy implemented in the financial year in review. Disclosure of remuneration policy (variables Y5–Y10): Y6=Overview of the remuneration policy for the following financial
year/subsequent years; Y7=Proportion between fixed and variable components; Y8=Financial/non-financial performance criteria applied the annual bonus; Y9=Financial/non-financial performance criteria applied for the share-based
remuneration; Y10=Information on the policy regarding termination payments. Disclosure of individual remuneration (variables Y11–Y15): Y11=For each executive director, breakdown of each component of annual compensation; Y12=For
each non-executive director, breakdown of each component of annual compensation, Y13=Number of shares granted during the year in review; Y14=Number of shares exercised during the year in review; Y15=Number of shares unexercised/
outstanding; Y_All is the average of all the variables Y1–Y15 with no missing values. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
4. Pay structure and levels

In this section, we measure the level and analyse the structure of
directors’ and CEO compensation. As anticipated in Section 3, the
sample used here consists of 245 firms and is obtained by subtracting
from the whole sample of 279 firms analysed in the previous Section, 34
firms that did not disclose individual compensation of the CEO and each
member of the board, either in 2007 and/nor in 2010.
4.1. Summary statistics on the level and structure of executive

compensation at EU firms
We measure the level of total compensation as the sum of base salary,
benefits, cash bonuses and other types of cash-based compensation, as well
as the estimated value of annual stock grants and stock options. In order to
improve comparability, we exclude from ‘other cash pay’ occasional pay-
ments, such as pension annuities, termination payments and consulting
fees. Stock grants at year t are valued as the product of the stock price in t
multiplied by the number of stocks granted.48 Stock options are valued by
applying the Black and Scholes model to the option’s relevant characteristics
(exercise price; time to maturity; volatility, market price and dividend yield
of the underlying stock; risk free rate) at the valuation date. The remuner-
ation structure is defined as the proportion of the different components of
the remuneration package, as explained in detail below.
Tables 6.8(a) and 6.8(b) summarise the average and median amount
of total compensation for the CEO and the full board in our sample.
The level of board compensation varies across countries, with the UK
and German firms paying their boards significantly more, although
arguably for different reasons. While in the UK the high levels of
board compensation are determined by higher individual average pay
(in particular for CEOs), in Germany the two-tier system is characterised
by a relatively high number of supervisory directors and an addi-
tional number of management board members. In fact, as reported in
48
For performance-based stock grants we use the fair value, if provided by the firm. Otherwise,
we value performance-contingent awards at 100 per cent of their face value. This method
leads to an over-valuation of performance-based stock grants; however, a more thorough
evaluation is often precluded either by the lack of sufficient information about contingency
terms or by the difficulty of calculating the discount related to certain conditional perform-
ance, such as accounting performance. Previous papers, e.g. Muslu (2010) and Conyon et al.
(2011), adopt the same convention, while some other studies use a lower, although arbitrary,
valuation rate (i.e. 80 per cent as in Conyon and Murphy (2000)).
Table 6.8(a) Mean (median) total compensation of the board of directors
Whole sample Non-Financial Financial

Delta Delta Delta
Countries n 2007 2010 (%) n 2007 2010 (%) n 2007 2010 (%)
Switzerland 17 8,246,703 12,039,883 46.0 13 6,926,873 11,987,744 73.1 4 12,536,150 12,209,336 −2.6
5,356,642 8,342,879 55.7* 4,774,811 7,898,956 65.4* 10,046,059 14,667,530 46.0
CONT 23 7,908,795 7,692,860 −2.7 18 7,196,877 8,474,474 17.8 5 10,471,699 4,879,052 −53.4
7,786,200 7,466,675 −4.1 7,892,931 8,738,962 10.7 7,078,623 3,756,620 −46.9
Germany 32 18,454,757 17,602,244 −4.6 26 16,961,471 17,583,634 3.7 6 24,925,666 17,682,886 −29.1
16,762,879 16,143,038 −3.7 15,808,800 16,297,122 3.1 18,795,779 13,685,441 −27.2
Spain 12 6,420,661 7,327,803 14.1 7 3,170,205 4,317,414 36.2 5 10,971,300 11,542,348 5.2
3,880,500 4,793,063 23.5 3,380,665 4,522,538 33.8* 7,935,500 8,846,958 11.5
France 49 6,784,722 5,425,671 −20.0 43 7,038,186 5,619,516 −20.2 6 4,968,234 4,036,449 −18.8
5,139,626 3,946,294 −23.2 5,139,626 4,315,512 −16.0 5,046,436 3,136,425 −37.8
UK 61 15,654,013 15,645,729 −0.1 49 14,222,731 15,109,479 6.2 12 21,498,417 17,835,414 −17.0
12,389,949 13,861,763 11.9 11,833,451 13,117,894 10.9 23,753,056 16,442,263 −30.8
Italy 18 9,347,950 7,861,918 −15.9 11 9,279,477 7,567,501 −18.4 7 9,455,549 8,324,572 −12.0
6,351,791 6,472,000 1.9 6,187,581 5,564,342 −10.1 6,516,000 7,252,000 11.3
NORD 27 3,568,703 3,050,778 −14.5 21 3,855,751 3,279,669 −14.9 6 2,564,034 2,249,658 −12.3
1,961,327 1,937,722 −12.0 1,961,327 1,863,190 −5.0 2,135,566 2,083,516 −2.4
PIG 6 8,691,645 5,144,873 −40.8 4 8,065,063 6,365,309 −21.1 2 9,944,809 2,704,000 −72.8**
9,483,503 5,327,665 −43.8** 8,993,633 6,872,012 −23.6 9,944,809 2,704,000 −72.8*
Total 245 10,586,982 10,236,006 −3.3 192 9,883,562 10,184,157 3.0 53 13,135,220 10,423,835 −20.6
7,786,200 7,066,146 −9.20 7,117,884 7,029,041 −1.2 9,972,401 7,252,000 −27.3
Sweden. Total compensation is measured as the sum of base salary, benefits, cash bonuses and other types of cash pay, plus the
estimated value of annual stock grants and stock options; *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels,
respectively.
Table 6.8(b) Mean (median) total compensation of the CEO
Whole sample Non-Financial Financial

Countries n 2007 2010 Delta (%) n 2007 2010 Delta (%) n 2007 2010 Delta (%)
Switzerland 17 4,004,414 5,113,417 27.7 13 3,329,907 5,238,218 57.3 4 6,196,562 4,707,813 −24.0
3,165,816 5,078,844 60.4* 3,058,466 4,808,943 57.2* 4,724,064 5,258,717 11.3
CONT 23 3,079,837 3,120,057 1.3 18 3,003,776 3,516,572 17.1 5 3,353,656 1,692,602 −49.5
2,955,411 2,347,000 −20.6 7,AQ 2,650,667 −6.3 4,058,637 1,610,000 −60.3
Germany 32* 4,316,769 4,148,762 −3.9 26 4,092,236 4,418,206 8.0 6 5,289,744 2,981,172 −43.6
3,527,580 3,647,281 3.4 3,673,791 3,795,031 3.3 3,274,264 2,578,100 −21.3
Spain 12 1,826,847 1,973,399 8.0 7 575,380 758,001 31.7** 5 3,578,900 3,674,955 2.7
839,836 972,500 15.8 386,667 344,574 −10.9* 1,549,500 2,124,500 37.1
France 49 3,277,435 2,962,373 −9.6 43 3,370,271 3,132,490 −7.1 6 2,612,110 1,743,204 −33.3
2,850,056 2,641,635 −7.5 2,850,056 2,679,600 −6.0 2,910,161 957,449 −67.1
UK 61 5,939,228 6,400,478 7.8 49 5,808,493 6,503,662 12.0 12 6,473,062 5,979,144 −7.6
5,018,280 5,903,814 17.6 4,911,832 5,809,655 18.3 5,632,922 6,134,589 8.9
Italy 18 4,854,126 3,571,968 −26.4 11 5,058,572 4,326,113 −14.5 7 4,532,854 2,386,881 −47.3
3,040,500 2,991,500 −7.6* 3,042,000 2,960,000 −2.7 3,039,000 3,023,000 −0.5
NORD 27 1,847,393 1,593,561 −13.7 21 1,963,651 1,667,598 −15.1 6 1,440,492 1,334,431 −7.4
1,372,711 1,249,237 −9.0 1,372,711 1,338,803 −2.5 1,260,331 1,057,390 −16.1
PIG 6 3,095,233 1,627,009 −47.4 4 2,910,599 2,147,513 −26.2 2 3,464,501 586,000 −83.1**
2,859,521 927,566 −67.6 1,979,514 2,014,922 1.8 3,464,501 586,000 −83.1*
Total 245 3,990,541 3,950,203 −1.0 192 3,884,603 4,146,604 6.7 53 4,374,317 3,238,713 −26.0
3,128,027 3,008,052 −3.8 3,041,158 3,117,724 2.5 3,544,349 2,337,475 −34.1
Sweden. Total compensation is measured as the sum of base salary, benefits, cash bonuses and other types of cash pay, plus the
estimated value of annual stock grants and stock options; *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels,
respectively.
Table 6.8(b), the average total pay of a UK CEO is consistently higher

than that of the German counterpart, which – in turn – is in line with the
remuneration found for most other countries.
The evolution of total compensation between 2007 and 2010 reveals
that pay practices have been affected by the financial crisis. For the whole
board, total compensation decreases in most European countries. While
the mean variation for the whole sample is equal to −3.3 per cent
(median −9.2 per cent), for PIG countries, more exposed to the early
wave of the sovereign debt crisis, the impact is more severe (average
−40.8 per cent, median −43.8 per cent).49 Differential effects emerge for
financial and non-financial companies. Over the whole period, board
compensation at financial firms decreases by 20.6 per cent (median
−27.3 per cent), while non-financial firms experience less significant
changes (average +3.0 per cent, median −1.2 per cent). A similar trend
is detected for the CEO, whose total compensation decreases by 1.0 per cent
over the whole sample (median −3.8 per cent), with significant differences
between financial and non-financial firms. In contrast with the slight
increase detected for non-financial firms (average 6.7 per cent, median 2.5
per cent), CEO compensation in financial firms is reduced by 26.0 per cent
(median −34.1 per cent). Moreover, the drop in total CEO compensation in
financial firms is different in magnitude from that found for the board as a
whole, with the CEOs experiencing much stronger decreases than other
board members.
We further examine CEO and board remuneration through the analysis
of the composition of the compensation package. Tables 6.9 and 6.10
summarise the following main components of CEO and board compensa-
tion: (1) fixed cash pay, (2) variable cash pay, (4) value of annual grants of
stock and option-based incentive plans, and (6) value at the end of the year
of the portfolio of stocks and options granted through incentive plans over
the previous years. The last four columns of the Tables indicate the pro-
portion between the various aggregates of compensation: (7) cash-based
compensation over total compensation; (8) incentive compensation (vari-
able cash and stock-based) over total compensation; (9) cash-based variable
compensation over total variable compensation; (10) annual stock grants
over annual stock and option grants.50
49
However, the very small number of observations and the high volatility of the results
makes this trend inconsistent.
50
Given that the last four columns are the mean of individual ratios, they do not match
with the ratios of the average values reported in columns (1), (2), (4) and (5).
Table 6.9(a) Composition of CEO mean and median pay and stock-based incentive portfolio: whole sample
(9) BONUS
(6) Value of (8) PROP (10) SG
the (7) CASH INCENTIVES Bonus PROP Stock
(1) Fixed (2) Variable (3) = (1)+(2) (4) Stock- (5) = (3)+(4) portfolio of PROP Cash/ Compensation (Bonus Grant/
No. Pay & Cash Cash based Total SG & SO at Total (Bonus+Stock- +Stock- (Stock-
Year of s. Benefits Compensation Compensation Compensation Compensation 31/12 Compensation based)/Total based) based)
2007 (mean) 245 1,103,367 1,342,611 2,445,978 1,544,563 3,990,541 6,770,201 74.7% 59.9% 61.5% 61.7%
2010 (mean) 245 1,156,408 1,096,816 2,253,224 1,696,979 3,950,203 6,334,242 73.5% 54.3% 57.0% 67.9%
Difference 53,041 −245,795** −192,754 152,415 −40,338 −435,960 −1.14% −5.5%** −4.6% 6.2%
2007 (median) 245 993,000 1,089,782 2,202,000 665,062 3,128,027 1,114,300 79.5% 68.6% 60.3% 79.7%
2010 (median) 245 1,036,000 834,680 2,059,836 628,943 3,008,052 1,348,389 77.4% 62.2% 52.9% 100.0%
Difference 43,000 −255,102* −142,164 −36,119 −119,975 234,089 −2.1% −6.4%*** −7.3% 20.3%
Note: *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 6.9(b) Composition of CEO mean and median pay and stock-based incentive portfolio: sample of non-financial firms
(8)
(6) Value of (7) CASH INCENTIVES
(1) Fixed (2) Variable (3) = (1)+(2) (4) Stock- (5) = (3)+(4) the portfolio PROP Cash/ Compensation (9) BONUS PROP (10) SG PROP
No. Pay & Cash Cash based Total of SG & SO at Total (Bonus+Stock- Bonus (Bonus Stock Grant/
Year of s. Benefits Compensation Compensation Compensation Compensation 31/12 Compensation based)/Total +Stock-based) (Stock-based)
2007 (mean) 192 1,051,262 1,207,662 2,258,925 1,625,678 3,884,603 7,183,093 72.9% 60.8% 60.1% 59.3%
2010 (mean) 192 1,126,097 1,190,376 2,316,473 1,830,131 4,146,604 6,895,583 71.7% 58.4% 56.5% 65.2%
Difference 74,834 −17,286 57,548 204,453 262,001 −287,511 −1.24%** −2.4% −3.6% 5.9%**
2007 (median) 192 962,527 1,017,599 2,024,138 711,732 3,041,158 1,113,739 74.6% 69.0% 58.7% 68.6%
2010 (median) 192 1,030,269 920,996 2,097,355 806,206 3,117,724 1,872,210 74.9% 65.2% 52.8% 100.0%
Difference 67,742 −96,604 73,217 94,474 76,565 758,471 0.3% −3.8% −5.9% 31.4%
Note: *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 6.9(c) Composition of CEO mean and median pay and stock-based incentive portfolio: sample of financial firms
(2) Variable (3) = (1) (6) Value of (7) CASH (8) INCENTIVES (9) BONUS
(1) Fixed Cash +(2) Cash (4) Stock- (5) = (3)+(4) the portfolio PROP Cash/ Compensation PROP Bonus (10) SG PROP
No. Pay & Compensa Compensa based Total of SG & SO at Total (Bonus+Stock- (Bonus Stock Grant/
Year of s. Benefits tion tion Compensation Compensation 31/12 Compensation based)/Total +Stock-based) (Stock-based)
2007 (mean) 53 1,292,123 1,831,481 3,123,603 1,250,714 4,374,317 5,274,442 81.1 56.4% 67.1% 71.2%
2010 (mean) 53 1,266,216 757,880 2,024,096 1,214,617 3,238,713 4,300,705 80.3;1 39.5% 59.1% 84.4%
Difference −25,907 −1,073,601*** −1,099,508*** −36,097 −1,135,605* −973,737 −0.& −16.9%*** −8.1% 13.2%**
2007 (median) 53 1,100,000 1,571,591 2,853,543 252,000* 3,544,349 1,181,185 87.4;1 65.8% 70.9% 100.0%
2010 (median) 53 1,038,962 319,000** 1,961,085*** 2,337,475** 10,663* 100.0 39.4% 56.2% 100.0%
Difference −61,038 −1,252,591 −892,458 −252,000 −1,206,874 −1,170,522 12.61 −26.5% −14.6% 0.0%
Note: *** and *** denote statistical significance at the 10% 5% and 1% levels, respectively.
Table 6.10(a) Regression analysis of determinants of CEO total compensation
(1) (2) (3) (4) (5)

Intercept 11.106 1.257 0.256 0.940*** 0.158
17.76*** 6.57*** 1.31 3.04 0.34
Year 2010 0.027 −0.000 −0.000 0.006 0.077
0.34 −0.01 −0.01 0.16 1.32
Financial 1.048 −0.266 −0.432 −0.803 −1.151
0.85 −0.71 −1.12 −1.22 −1.20
Financial*Year 2010 −0.334 0.012 −0.104 −0.003 0.033
−2.45** 0.28 −2.44** −0.05 0.29
Log Assets 0.308 −0.039 0.024 −0.034 0.036
8.70*** −3.63*** 2.13** −1.96* 1.39
Log Assets*Financial −0.081 0.022 0.017 0.050 0.059
−1.25 1.11 0.84 1.45 1.17
Tobin’s Q 0.104 −0.052 0.023 −0.045 −0.022
2.49** −4.09*** 1.78* −2.17** −0.62
3Y_Ret −0.096 0.002 0.077 0.086 −0.072
−0.56 0.04 1.46 1.00 −0.51
3Y_ROA 0.551 0.411 −0.107 0.524 0.954
0.84 2.04** −0.52 1.60 2.03**
CH −0.411 0.089 −0.071 0.132 −0.022
−6.76*** 4.81*** −3.73*** 4.26*** −0.47
Country dummies Yes Yes Yes Yes Yes
n. significant at 5% 8 8 7 7 4
Note: In specification (1), the dependent variable is CEO total compensation; in

specifications from (2) to (5) the dependent variables are proxies for CEO
compensation structure, Cash/Total compensation in spec. (2); (Bonus+Stock based)/
Total Compensation in spec. (3); Bonus/(Bonus+Stock based) in spec. (4); Stock
Grant/(Stock based) in spec. (5). Year 2010 is a dummy variable equal to 1 for year
2010 and 0 otherwise; Financial is a dummy variable equal to 1 if the ultimate
shareholder is a financial institution, and 0 otherwise; Financial*Year is the interaction
between Financial and Year 2010 variables; Log Assets is the Log of Total assets;
LogAssets*Financial is the interaction between Log Assets and Financial variables;
Tobin’s Q is the ratio between market and book value of the assets of the firm; 3Y_Ret
is the average stock returns over the last three years; 3Y_ROA is the average accounting
returns over the last three years; CH is a dummy variable equal to 1 if the first largest
shareholders owns more than 20% of voting rights and 0 otherwise; *, **, and ***
denote statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 6.10(b) Regression analysis of determinants of board total

compensation
(1) (2) (3) (4) (5)

Intercept 10.102 1.284 0.233 0.845 0.101
14.79*** 5.80*** 1.04 2.67*** 0.21
Year 2010 0.002 −0.004 0.001 0.016 0.077
0.02 −0.13 0.04 0.40 1.28
Financial 1.996 −0.331 −0.391 −0.485 −1.259
1.49 −0.76 −0.89 −0.71 −1.29
Financial*Year 2010 −0.444 0.006 −0.143 −0.036 0.089
−2.98*** 0.12 −2.92*** −0.48 0.76
Log Assets 0.305 −0.044 0.030 −0.030 0.041
7.88*** −3.54*** 2.36** −1.65* 1.54
Log Assets*Financial −0.132 0.027 0.015 0.033 0.061
−1.86* 1.16 0.64 0.93 1.19
Tobin’s Q 0.190 −0.054 0.018 −0.046 −0.020
4.14*** −3.66*** 1.22 −2.16** −0.56
3Y_Ret −0.238 0.031 0.084 0.147 0.059
−1.28 0.51 1.39 1.67* 0.39
3Y_ROA 0.064 0.438 −0.067 0.515 0.750
0.09 1.89* −0.29 1.54 1.51
CH −0.357 0.096 −0.076 0.121 −0.025
−5.38*** 4.47*** −3.50*** 3.81*** −0.51
Country dummies Yes Yes Yes Yes Yes
n. significant at 5% 8 8 8 7 4
Note: In specification (1), the dependent variable is board total compensation; in

specifications from (2) to (5) the dependent variables are proxies for CEO
compensation structure (Cash/Total compensation in spec. (2); (Bonus+Stock
based)/Total Compensation in spec. (3); Bonus/(Bonus+Stock based) in spec. (4);
Stock Grant/(Stock based) in spec. (5)). Year 2010 is a dummy variable equal to 1
for year 2010 and 0 otherwise; Financial is a dummy variable equal to 1 if the
ultimate shareholder is a financial institution, and 0 otherwise; Financial*Year is
the interaction between Financial and Year 2010 variables; Log Assets is the Log of
Total assets; LogAssets*Financial is the interaction between Log Assets and
Financial variables; Tobin’s Q is the ratio between market and book value of the
assets of the firm; 3Y_Ret is the average stock returns over the last three years;
3Y_ROA is the average accounting returns over the last three years; CH is a
dummy variable equal to 1 if the first largest shareholders owns more than 20% of
voting rights, and 0 otherwise; *, **, and *** denote statistical significance at the
10%, 5%, and 1% levels, respectively.
For the CEO, the slight decrease in total compensation over the 2007–10
period is due mainly to the reduction in variable cash compensation
(bonus), that offsets the average increase in the annual stock-based grants.
In terms of composition of the remuneration package, column (7) of
Table 6.9(a) indicates that the pay of European CEOs still relies heavily
on cash, accounting for about 74 per cent of the total compensation in both
years. This compares with a 50 per cent value in the average US S&P 500
firm (Conyon et al. 2010). Nonetheless, total incentives of European CEOs
are not trivial. Summing up variable compensation paid in cash and annual
stock-based grants, the ratio of performance-based compensation over total
compensation (column (8)) is about 60 per cent in 2007, and decreases to 54
per cent in 2010; the reduction is driven by cash-based variable pay.
A further source of incentives is provided by the impact of firm stock
price changes on the value of the CEO’s portfolio of stock and options
granted in previous years (column (6)). This amounts to €6.7 million on
average in 2007, reducing to €6.3 million in 2010. These results suggest
that the pay-performance sensitivity of the compensation package
granted to European CEOs is indeed non-trivial, although a note of
caution is needed, given that variable compensation is not necessarily
related to high firm performance.51
Another result, shown in column (10) of Table 6.9(a), is the progressive
substitution of stock options with stock grants. This is probably due to the
widespread criticism that stock options had a role in increasing firms’ risk.
We further refine our analysis by splitting the sample into financial and
non-financial firms. For non-financial firms only minor changes occurred
between 2007 and 2010, both for the level and the structure of CEO
compensation. Total compensation increased slightly over time, driven
mainly by the larger amount of stock-based compensation, while only a
small decrease is detected for variable cash-based pay. As a consequence, the
structure of compensation is not significantly affected, while it is worth
noting the increase in the proportion of stock grants over stock-based grants
over stock-based compensation (see Table 6.9(b)).
In contrast, for financial firms both the level and the structure of CEO
compensation changed substantially over the sample period. In fact, CEO
51
For example, in financial firms one of the main issues regading bonus payments is
that they are often related to the amount of loans granted, but not to their quality.
This can lead to higher bonus levels awarded to management as the amount of loans
increases, even though the value of the firm may decrease due to the lower quality of
the assets.
total compensation in financial firms in 2007 was, on average, €4.3 million

higher than in non-financial firms (€3.9 million). In 2010, however,
the picture is reversed: total CEO compensation in financial firms
dropped to €3.2 million as a result of the large decrease in variable cash
pay (−60 per cent), while in non-financial firms it increased to €4.1 million.
These changes also affected the composition of CEO pay at financial
firms, an issue that has dominated the debate since 2008, when national
and international regulators emphasised the need for proportionality
between variable and fixed pay, and between cash-based and stock-based
compensation. Table 6.9(c) shows that the proportion of incentive-based
compensation over total compensation dropped significantly, by an average
17 per cent from 2007 to 2010, as a consequence of a lower variable cash-
based component. Within the variable part of compensation, the weight of
cash-based incentives shows a mean (median) decrease of 8 per cent (14 per
cent) (column (9)). Nevertheless, cash compensation in financial firms is still
significantly higher than in non-financial firms.
These results could be driven by the regulatory pressure on the
financial sector to move focus from short-term (typically cash-based)
to long-term incentives (typically stock-based).52
Similar results are obtained for the remuneration of the full board of
directors.
4.2. Regression analysis of the level and structure of executive

compensation
The results outlined in the previous section suggest that the level and
structure of board and CEO compensation experienced substantial
changes over the 2007–10 period. However, as highlighted by previous
studies, multiple factors could jointly affect CEO and board compensa-
tion, such as firm size, performance and ownership concentration.53
These factors must therefore be taken into account when comparing
remuneration packages for 2007 and 2010.
With regard to firm size, Rosen (1982) predicts that larger firms require
more talented and costly management, as confirmed by Baker et al. (1988)
and, more recently, Gabaix and Landier (2008). Firm performance is
expected to affect CEO pay positively, as outlined by Kaplan (1994),
Murphy (1985) and Core et al. (1999). As to ownership concentration,
52
FSB Principles and Standards, EU CRD III, supra. note 26.
53
For a review of the literature on this issue, see Barontini and Bozzi (2009).
Bertrand and Mullainathan (2001) point out that large shareholders exert
strong control over CEO behaviour, thus curtailing her ability to capture the
pay process and extract excessive compensation. Some papers offer empir-
ical evidence consistent with this hypothesis.54
Therefore, in evaluating the dynamics of executive pay in 2007 and 2010,
we control for these factors through a multivariate regression analysis.
Our general model is the following:
ð1ÞExe Comp ¼ α þ β1 Year 2010 þ β2 Financial þ β3 ½Country
þ β4 ½Control þ ε
The dependent variable ExeComp is alternatively a proxy for either the level
or the structure of CEO compensation. The level is measured as total
compensation, while the structure is described through the four indexes
reported in Table 6.7, i.e. the proportion of cash-based compensation,
incentive compensation, cash bonus over total incentives and stock grants
over stock-based bonus. Year 2010 and Financial are dummy variables that
capture the year (2007 or 2010) and the sector (financial vs. non-financial).
[Country] is a set of dummy variables that refers to the country where the
firm has its primary listing (the UK is the reference for any other state), while
[Control] is a set of independent variables that captures characteristics that
could affect compensation, in particular Log Assets, Tobin’s Q, previous three
years market and accounting returns, ownership concentration (a dummy
that takes value 1 if the largest shareholder has more than 20 per cent of
voting rights). We limit the discussion to CEO pay; however, the same
conclusions may be drawn for the board as a whole. The results for the
CEO are summarised in Table 6.10(a).
In specification (1) the dependent variable is the level of CEO total
compensation. Our results suggest that, after appropriate controls, CEO
pay changed in 2010 only in financial firms (the coefficient
Financial*Year 2010 is negative and significant, while the coefficient
Year2010, referred to non-financial firms, is not significant). The trend
outlined in the previous section is thus confirmed, even after removing
54
Hartzell and Starks (2003) find that institutional investors’ ownership concentration is
negatively related to the level of compensation. Further support is provided for
Germany, where a negative effect of concentrated ownership on the average annual
salary of the management board has been detected (FitzRoy and Schwalbach 1990),
while bank influence and large ownership of stock by various groups are associated with
lower executive pay (Elston and Goldberg 2003). The same negative relationship
between ownership concentration and the level of CEO pay is found by Mertens and
Knop (2010) on a sample of Dutch firms, and by Sapp (2008) for Canadian firms.
the confounding effect on total compensation exerted by the control

variables. As expected, we find that Log Assets and growth opportunities
(Tobin’s Q) coefficients are positive and highly significant. The owner-
ship concentration variable CH shows a significant negative effect, in line
with previous empirical evidence. Furthermore, country dummies con-
firm relevant differences across jurisdictions.
Specifications (2) to (5) adopt the four indexes of the structure of CEO
pay as dependent variables. The structure of CEO pay is affected by firm
characteristics. In fact, we obtain significant coefficients for all control
variables, with the exception of three previous years stock returns
(3YReturn). Firms that are larger (Log Assets) and with high growth
opportunities (Tobin’s Q) are associated with: lower proportion of cash-
based pay, lower incidence of bonuses on variable compensation (spec-
ifications (2) and (4)), and higher proportion of variable pay (specifica-
tion (3)). In this respect, financial firms do not behave differently from
non-financial firms, as pointed out by the non-significant coefficient on
the interaction term Log Assets*Financial.
The structure of CEO pay is also affected by firm accounting perform-
ance, as suggested by the positive relation between the recourse to cash-
based compensation (specifications (2) and (4)) and the ROA of the
previous three years. Finally, the presence of a large shareholder (CH) is
negatively related to the percentage of incentive-based compensation
(specification (3)), with a preference for cash-based instead of stock-
based compensation (the positive coefficients of CH on specifications (2)
and (4)). This evidence seems to provide empirical support for the idea
that, in closely held firms, the monitoring activity of the main share-
holder is a substitute for the financial incentives to the CEO (Shleifer and
Vishny 1986; Barontini and Bozzi 2011).
Concerning the evolution in the structure of CEO pay over the
2007–10 years, no significant changes are detected for non-financial
firms (the coefficient on Year2010 is not significant in all the regression
specifications), as already emerged from the univariate analysis reported
above. On the contrary, the results reported in Table 6.10 show that, for
financial firms, the CEO compensation structure has changed signifi-
cantly in 2010 relative to 2007. In particular, this is true with respect to
the proportion of incentive pay, as indicated by the significant and
negative coefficient on the interaction term Financial*Year 2010
in specification (3). This implies that over time financial firms have
reduced the proportion of variable pay with respect to total compensa-
tion; moreover, as suggested by the negative, albeit not significant,
coefficient on Financial*Year 2010 in specification (4) and by the results

in Table 6.9(c), this is due mainly to the reduction in the variable cash
compensation.
The analysis of this trend in financial firms opens the field to various
interpretations. The reduction could be due to the negative performance
of the firms in 2010, leading to a lower amount of variable cash com-
pensation awarded to the CEO with respect to other components of the
pay package. It could also be due to a high sensitivity of CEO remuner-
ation to firms’ performance, as further confirmed by the positive and
significant coefficient on firms’ previous returns (3YReturn). However,
more subtle evidence emerges from the analysis of Table 6.10(a), given
that the results are already filtered for the effect of firm’s performance
over the 2007–10 period, due to the presence of the control variables for
accounting and stock market performance (3YRet and 3YROA) in the
regression specification. Consequently, the change in the proportion of
incentives detected in 2010 for financial firms seems to be related to
factors going far beyond the negative performance during the 2007–10
period.
While several factors not considered in the analysis could explain this
occurrence, regulatory pressure on financial firms in the same period
may have played a role. Changes in the pay structure of financial firms
follow the direction indicated by the regulators, i.e. better focus on the
risk implications of pay-packages, ‘adequate’ balance between variable
and fixed compensation and a substantial portion of variable compen-
sation awarded in share or share-linked instruments. Our results there-
fore suggest that several financial institutions have modified their
compensation policies in line with the global principles.55
In order to check the robustness of these results, we ran further
regressions (not displayed here) where independent variables interact
with the year 2010 dummy. In line with our previous results, country-
specific differences in the dynamics of the CEO and board pay over the
2007–10 period are not significant in regression (1) and not significant in
specifications (2)–(5), related to the structure of compensation, leaving
all previous results unchanged. The same outcome is obtained for the
interaction of ownership and performance variables with the year 2010
dummy.
55
In particular, EU ‘significant financial institutions’ started to adopt some of the interna-
tional FSB Principles after their adoption in 2009, also in anticipation of the CRD III
official publication. See Ferrarini and Ungureanu 2011a; b; c.
5. Conclusions
In this chapter we have analysed the evolution of director remuneration
structure, governance and disclosure in the EU and Member States’
legislation and in the practice of large European firms before and after
the 2008 financial crisis. To start with, we have shown that the imple-
mentation of EU recommendations concerning remuneration govern-
ance and disclosure in the Member States has improved between the two
reference years (2007 and 2010). Compliance with all applicable criteria
has improved across jurisdictions; however, relevant differences still
remain. Germany, France and Italy show the most significant improve-
ments, while the highest level of overall conformity may be found in the
UK, in Switzerland and in some continental countries.
We found a remuneration committee present in the majority of our
sample companies, both before and after the crisis, with the notable
exception of Germany, where only about half of the companies estab-
lished a similar committee. The independence criteria for the remuner-
ation committee were fulfilled by about 80 per cent of the companies
appointing this committee (a level slightly higher than in 2007). All UK
companies in our sample disclosed the presence and independence
of remuneration consultants, whereas most other companies did not
disclose whether similar consultants had assisted their boards or whether
the same, when present, were independent.
In addition, we found that disclosure of remuneration generally
improved after the crisis in all jurisdictions, however, with remarkable
variations across countries. Amongst the requirements for remuneration
policy disclosure, the generic one concerning the disclosure of a remu-
neration statement was already widely complied with before the crisis.
The others showed significant improvements over the sample period
(the largest increase occurring for the disclosure of termination pay-
ments). As to individual disclosure, high compliance was already
observed pre-crisis for both executive and non-executive directors,
whereas increases were observed in the disclosure of individual share
schemes.
The general picture shows that compliance is on the rise and is
significantly affected by firm size, industry, ownership concentration
and country. In particular, we have shown some significant differences
in approach between concentrated and WH firms, with the latter gen-
erally being more compliant with respect to almost all criteria consid-
ered, both before and after the crisis. Where size influences the result, the
coefficient is always positive, reflecting higher compliance with most

criteria in larger firms. The company sector (financial vs. non-financial)
is also a main factor for predicting compliance with some of the
requirements.
We have also analysed the dynamics of the level and structure of
remuneration packages (for the whole board and for CEOs), both before
and after the crisis. Our results show that remuneration packages vary
remarkably across countries, also reflecting different board structures
and separate markets for managerial talent. These differences may be
attributed, at least in part, to differences in institutional contexts, allow-
ing segmentation to persist.
Whilst previous studies have shown that directors’ pay in Europe is
lower than in the US and a lower percentage of it is stock-based (25 per
cent against 50 per cent in the US), we observe that the variable remu-
neration of European CEOs is definitely non-trivial, amounting to 60 per
cent of total compensation (decreasing to 54 per cent in 2010, due to the
substantial reduction in cash-based variable pay). The structure of CEO
pay (in terms of relative weight of fixed/variable components) is affected
by firm size, growth opportunities and past firm performance. The
recourse to stock-based compensation is lower in CH companies. This
is consistent with the hypothesis that stock-based incentives are less
important when there are controlling shareholders and also with the
fact that blockholders are more sensitive to the implicit cost (in terms of
dilution) of new share issues.
Board and CEO remuneration have decreased remarkably after the
crisis in financial institutions, while slightly increasing in non-financial
companies. The decrease of CEO pay is substantially driven by the cash
portion of variable compensation (bonuses), whereas other components
have remained more or less stable. This is true for both fixed salary and
stock-based compensation; furthermore, stock grants are apparently
becoming more popular than stock options. The reduction of variable
pay in financial firms may be explained by the negative performance in
the 2007–10 period, but also by the pressure of national and interna-
tional financial regulators. Indeed, the changes observed in the pay
structure of financial firms follow the guidelines indicated by regulators
to reach an ‘adequate’ balance between variable and fixed components;
moreover, variable compensation is deferred and awarded, at least in
part, in shares or share-linked instruments, as required under the recent
international criteria.
Also from a policy perspective, a distinction should be made between

financial institutions and non-financial companies. As to the former,
managerial remuneration has been seen as one of the main determinants
of excessive risk-taking and therefore as one of the possible causes of the
financial crisis. However, the available evidence shows that managerial
and shareholder incentives were already aligned in banks before the
crisis and that short-term incentives should not have produced an
adverse impact on bank performance during the crisis. The main prob-
lem may have been insufficient or ineffective prudential regulation,
rather than flawed corporate governance.
A number of new rules concerning the disclosure, governance, level
and structure of managerial remuneration have been introduced world-
wide in response to the crisis. The evidence produced in this volume
shows that the level and structure of managerial compensation at
European financial institutions have, indeed, changed since the crisis.
In particular, CEOs have experienced a reduction of their cash bonuses,
while other components of their remuneration remained substantially
unchanged. Furthermore, stock grants have apparently become more
popular than stock options. It is, however, difficult to understand if pay-
performance sensitivity has increased or decreased as a result.
In non-financial companies leverage is generally much lower and
excessive risk-taking is regarded as a less important issue. The main
concerns focus on managers possibly using their power to extract rents
from the company at the expense of shareholders. These problems can be
remedied, in part at least, through governance and disclosure mecha-
nisms, as seen throughout this chapter.
Our evidence shows that compliance with the EU recommendations
concerning the governance of the remuneration process is satisfactory. We
see no need to change the current approach in any fundamental way. As to
disclosure issues, our evidence shows that the implementation of EU rec-
ommendations in this field has been diverse. Disclosure of individual remu-
neration has increased remarkably over the last few years; however,
transparency lags behind in respect of for example, forward-looking policy,
performance parameters for the variable component and dynamics of stock-
based compensation. The implementation of recommendations requires
time; nonetheless, should this situation persist, harmonisation measures
might also be advisable for non-financial firms.
However, we do not envisage the need for EU harmonisation and
reform as to remuneration structures in non-financial firms. Indeed, our
evidence has shown that remuneration is related to firm fundamentals,
such as size, sector, growth opportunities and corporate results.

Moreover, the incentive structures largely depend on ownership types
and are less relevant for concentrated ownership companies, while
diffuse ownership ones are generally more compliant with the remuner-
ation governance and disclosure requirements. All of this seems to be
consistent with the hypothesis that the market for managerial services is,
at least to some extent, efficient, and that no regulatory reforms are
needed to constrain the parties’ freedom to conform the structure of
executive pay arrangements.
References
Abowd, J.M. and M. Bognanno, 1995. ‘International Differences in Executive and
Managerial Compensation’, in Freeman, R. B. and L. F. Katz (eds.),
Differences and Changes in Wage Structures, pp. 67–104, University of
Chicago Press.
Ahn, S. and W. Choi 2009. ‘The role of bank monitoring in corporate governance:
Evidence from borrowers’ earnings management behavior’, Journal of
Banking & Finance 33: 425–34.
Alcouffe, A. and C. Alcouffe 2000. ‘Executive compensation-setting practices in
France’, Long Range Planning 33: 527–43.
Andres, P. and E. Vallelado 2008. ‘Corporate governance in banking: The role of
the board of directors’, Journal of Banking & Finance 32: 2570–80.
Ángel, P.O. and V. S. Fumás 1997. ‘The compensation of Spanish executives: A
test of a managerial talent allocation model’, International Journal of
Industrial Organization 15: 511–31.
1998. ‘Agency theory and internal labour market explanations of bonus pay-
ments: empirical evidence from Spanish forms’, Journal of Economics &
Management Strategy 7: 573–613.
Baker, G.P., M. Jensen and K. J. Murphy 1988. ‘Compensation and incentives:
Practice vs. theory’, Journal of Finance 43: 593–616.
Barontini, R. and S. Bozzi 2009. ‘Board compensation and ownership structure:
Empirical evidence for Italian listed companies’, Journal of Management
and Governance 15: 59–89.
2011. ‘Ownership structure, family control, and CEO compensation: Evidence
from Continental Europe’, Working Paper.
Bebchuk, L. 2009. ‘Compensation Structures and Systemic Risk, Committee on
Financial Services, written testimony to the Committee on Financial Services,
US House of Representatives, available at http://blogs.law.harvard.edu.
and J. Fried 2003. ‘Executive compensation as an agency problem’, Journal of
Economic Perspectives 17: 71–92.
2004. Pay without Performance: The Unfulfilled Promise of Executive
Compensation, Harvard University Press, Oxford.
and D. Walker 2002. ‘Managerial power and rent extraction in the design of
executive compensation’, University of Chicago Law Review 69: 751.
Berle, A. and G. Means 1932. The Modern Corporation and Private Property,
Harcourt, Brace & World, New York.
Bertrand, M. and S. Mullainathan 2001. ‘Are CEOS rewarded for luck? The ones
without principals are’, The Quarterly Journal of Economics 116: 901–32.
Bhagat, S. and R. Romano 2010. ‘Reforming executive compensation: Simplicity,
transparency and committing to the long-term’, European Company and
Financial Law Review 7: 273–96.
Bizjak, J.M., M.L. Lemmon and L. Naveen 2008. ‘Does the use of peer groups
contribute to higher pay and less efficient compensation?’, Journal of
Brunello, G., C. Graziano and B. Parigi 2001. ‘Executive compensation and firm
performance in Italy’, International Journal of Industrial Organization 19:
133–61.
Chen, C.R., T.L. Steiner and A.M. Whyte 2006. ‘Does stock option-based execu-
tive compensation induce risk-taking? An analysis of the banking industry’,
Journal of Banking and Finance 30: 915–45.
Chizema, A. 2008. ‘Institutions and voluntary compliance: The disclosure of
individual executive pay in Germany’, Corporate Governance: An
Coffee Jr., J. 2003. ‘What caused Enron? A capsule social and economic history of
the 1990s’, Cornell Law Review 89: 269–91.
2005. ‘A theory of corporate scandals: Why the USA and Europe differ’, Oxford
Review of Economic Policy 21: 198.
Conyon, M.J. and K.J. Murphy 2000. ‘The prince and the pauper? CEO pay in the
United States and United Kingdom’, The Economic Journal 110: 640–71.
Conyon, M.J. and J. Schwalbach 2000. ‘Executive compensation: Evidence from
the UK and Germany’, Long Range Planning 33: 504–26.
Conyon, M.J., N. Fernandes, M. Ferreira, P. Matos, and K. Murphy 2010. ‘The
Executive Compensation Controversy: A Transatlantic Analysis’,
Fondazione Rodolfo De Benedetti, available at www.frdb.org.
Conyon, M.J. and G. Sadler 2010. ‘Shareholder voting and directors’ remuner-
ation report legislation: Say on pay in the UK’, Corporate Governance: An
Conyon, M.J. 2011. ‘Executive compensation consultants and CEO pay’,
Vanderbilt Law Review 64: 397.
Conyon, M.J., J. Core and W. Guay 2011. ‘Are US CEOS paid more than UK
CEOS? Inferences from risk-adjusted pay’, Review of Financial Studies 24:
402–38.
Core, J., R. Holthausen and D. Larcker 1999. ‘Corporate governance, Chief
Executive Officer compensation, and firm performance’, Journal of
Croci, E., H. Gonenc and N. Ozkan 2012. ‘CEO compensation, family control, and
institutional investors in Continental Europe’, Journal of Banking &
Finance 36: 3318–35.
Drobetz, W., P. Pensa and M. Schmid 2007. ‘Estimating the cost of executive stock
options: Evidence from Switzerland’, Corporate Governance: An
Elston, J.A. and L. Goldberg 2003. ‘Executive compensation and agency costs in
Germany’, Journal of Banking and Finance 27: 1391–410.
Fahlenbrach, R. and R. Stulz 2011. ‘Bank CEO incentives and the credit crisis’,
Fama, E. 1980. ‘Agency problems and the theory of the firm’, Journal of Political
Economy 88: 288–307.
and M.C. Jensen 1983. ‘Separation of ownership and control’, Journal of Law &
Ferrarini G., N. Moloney and M.-C. Ungureanu 2009. ‘Understanding Directors’
Pay in Europe: A Comparative and Empirical Analysis’, ECGI Law Working
Paper No. 126.
Ferrarini, G., N. Moloney and M.-C. Ungureanu 2010. ‘Executive remuneration in
crisis: A critical assessment of reforms in Europe’, Journal of Corporate Law
Studies 10: 73–118.
Ferrarini, G. and M.-C. Ungureanu 2010. ‘Executive pay at ailing banks and
beyond: A European perspective’, Capital Markets Law Journal
5: 197–217.
Ferrarini, G. and M.-C. Ungureanu 2011a. ‘Economics, politics, and the
International principles for sound compensation practices: An analysis of
executive pay at European banks’, Vanderbilt Law Review 62: 431–502.
Ferrarini, G. and M.-C. Ungureanu 2011b. ‘Lost in implementation: The flexibility
of the FSB Principles for sound compensation practices at financial
institutions’, Revue Trimestrelle de Droit Financier 1–2: 60–5.
Ferrarini, G. and M.-C. Ungureanu 2011c. ‘Bankers’ pay after the 2008 crisis:
Regulatory reforms in the US and the EU’, Zeitschrift fur Bankrecht, und
Bankwirtschaft 6: 418–30.
FitzRoy, F.R. and J. Schwalbach 1990. ‘Managerial Compensation and Firm
Performance: Some Evidence from West Germany’, WZB Discussion
Paper FS, IV 90–120.
Gabaix, X. and A. Landier 2008. ‘Why has CEO pay increased so much?’,
Quarterly Journal of Economics 123: 49–100.
Grinstein, Y., D. Weinbaum and N. Yehuda 2011. ‘The Economic Consequences
of Perk Disclosure’, Johnson School Research Paper Series No. 06.
Hartzell, J. and L. Starks 2003. ‘Institutional investors and executive
compensation’, The Journal of Finance 58: 2351–74.
Hayes, R. and S. Schaefer 2009. ‘CEO pay and the Lake Wobegon effect’, Journal of
Ikäheimo, S., A. Kjellman, J. Holmberg and S. Jussila 2004. ‘Employee Stock
Option Plans and stock market reaction: Evidence from Finland’, The
European Journal of Finance 10: 105–22.
Jensen, M.C. and W. Meckling 1976. ‘Theory of the firm: Managerial behavior, agency
costs and ownership structure’, Journal of Financial Economics 3: 305–60.
Jones, D.C., P. Kalmi and M. Mäkinen 2006. ‘The determinants of stock option
compensation: Evidence from Finland’, Industrial Relations: A Journal of
Economy and Society 45: 437–68.
Kaplan, S. 1994. ‘Top executive rewards and firm performance: A comparison of
Japan and the United States’, Journal of Political Economy 102: 510–46.
Kim, J.B., Y. Li and L. Zhang 2011. ‘CFOs versus CEOs equity incentives and
crashes’, Journal of Financial Economics 101: 713–30.
La Porta, R., F.L. de Silanes and A. Shleifer 1999. ‘Corporate ownership around the
world’, Journal of Finance 54: 471–517.
Lang, M. and R. Lundholm 1993. ‘Cross-sectional determinants of analyst ratings
of corporate disclosures’, Journal of Accounting Research 31: 246–71.
Larcker, D., A. Mccall, G. Ormazabal and B. Tayan 2012. ‘Ten Myths of “Say on
Pay”, Rock Center for Corporate Governance at Stanford University Closer
Look Series: Topics, Issues and Controversies in Corporate Governance No.
CGRP-26, available at www.ssrn.com.
Mertens, G. and N. Knop 2010. ‘The Impact of Ownership and Board Structure on
CEO Compensation in the Netherlands’, Erasmus University of Rotterdam
Working Paper, available at www.erim.eur.nl.
Morey, M., A. Gottesman, E. Baker and B. Godridge 2009. ‘Does better corpo-
rate governance result in higher valuations in emerging markets? Another
examination using a new data set’, Journal of Banking & Finance
33: 254–62.
Murphy, K.J. 1985. ‘Corporate performance and managerial remuneration: An
empirical analysis’, Journal of Accounting and Economics 7: 11–42.
2009. ‘Compensation structure and systemic risk’, Marshall Research Paper
Series Working Paper FBE 34–09: 1–15.
Muslu, V. 2010. ‘Executive directors, pay disclosures, and incentive compensation
in large European companies’, Journal of Accounting, Auditing & Finance
25: 569–605.
Rosen, S. 1982. ‘Authority, control, and the distribution of earnings’, The Bell
Journal of Economics 13: 311–23.
Rosenberg, M. 2003. ‘Stock Option Compensation in Finland: An Analysis of
Economic Determinants, Contracting Frequency, and Design’,
Scandinavian Working Papers in Business Administration No. 496.
Sapp, S. 2008. ‘The impact of corporate governance on executive compensation’,
European Financial Management 14: 710–46.
Shleifer, A. and R. Vishny 1986. ‘Large shareholders and corporate control’, The
Journal of Political Economy 94: 461–88.
Sun, J., S. Cahan and D. Emanuel 2009. ‘Compensation Committee governance
quality, Chief Executive Officer stock option grants, and future firm
performance’, Journal of Banking & Finance 33: 1507–19.
Vittanemi, J. 1997. ‘Top executive compensation and company performance in
Finland’, Centre for Labour Market and Social Research Working Paper
97–10, available at hrrp:77pure.au.dk.
Walker, D. 2009, A Review of Corporate Governance in Banks and Other Financial
Entities: Final Recommendations (the Walker Review), available at www.
icaew.com.
Zattoni, A. and A. Minichilli 2009. ‘The diffusion of equity incentive plans in
Italian listed companies: What is the trigger?’, Corporate Governance: An
7
Shareholder engagement at European general

meetings
luc renneboog and peter szilagyi
1. Introduction
The European Commission has formally pursued modernising and
harmonising shareholder rights in the European Union (EU) for close
to a decade. Its May 2003 Action Plan stated that shareholder engage-
ment at company general meetings was a particular priority, and set out
to remove the obstacles that prevented cross-border shareholders in
particular from exercising their participation rights (European
Commission 2004). The Shareholder Rights Directive 2007/36/EC (the
Directive) was finally adopted in July 2007, introducing minimum stand-
ards in shareholder admission to meetings, the dissemination of meet-
ing-related information, proxy allocation and distance voting, and
participation rights in terms of shareholders asking questions and
tabling proposals of their own. This pro-shareholder tendency has
been further deepened with the onset of the Global Financial Crisis,
with the European Commission (2011) issuing a Green Paper on the
European corporate governance framework and the governance role of
institutional investors, and Member States updating their corporate
governance codes to better accommodate shareholder voice.
Empirical research on the role and benefits of shareholder engagement
in Europe nonetheless remains limited, largely due to data availability
constraints. There is ample evidence that in the US, shareholder activism
both at general meetings (e.g. Ertimur et al. 2010; Renneboog and
Szilagyi 2011) and behind the scenes (e.g. Greenwood and Schor 2009;
Bradley et al. 2010) plays a useful role in addressing managerial agency
problems. The European Commission (2006) and Hewitt (2011) provide
only descriptive analyses of shareholder participation at European gen-
eral meetings. European shareholder proposals are examined by Cziraki
315
316 luc renneboog and peter szilagyi
et al. (2011), and for the UK and the Netherlands by Buchanan et al.
(2012) and De Jong et al. (2006), respectively. Shareholder interventions
outside general meetings are investigated by Armour (2008), Becht et al.
(2009) and Girard (2009).
This chapter is the first to provide a comprehensive analysis of
shareholder voice at European general meetings. We examine 42,170
management proposals and 329 shareholder proposals submitted to
general meetings in 17 European countries during the period between
2005 and June 2010. We seek answers to the following questions:
(i) Why and under what conditions do shareholders voice governance
concerns by refusing to support management proposals?
(ii) Why and under what conditions are firms targeted by shareholder
proposals?
(iii) What drives the level of voting support attracted by these share-
holder proposals?
We answer these questions by investigating the impact of not only meeting,
proposal and firm characteristics, but the various regulatory conditions that
have been argued to affect shareholder participation at general meetings.
Our results indicate that the Directive points to the right direction in
enabling shareholder engagement. While shareholder dissent at European
general meetings remains limited, there is evidence that it tends to be well-
placed. The level of dissent over management proposals is predominantly
driven by the proposal characteristics, with shareholders mostly objecting to
the adoption of anti-takeover devices and executive compensation.
Shareholder proposals are most likely to be submitted to large and poorly
performing firms, which indicates that, as in the US, the sponsoring share-
holders have the ‘correct’ objective of disciplining management. The share-
holder proposals targeting anti-takeover devices are by far the most
successful, again showing that the voting shareholders seek to discipline
management, through exposure to the market for corporate control.
Shareholder dissent has increased somewhat over time, with manage-
ment proposals enjoying less and shareholder proposals enjoying more
support. However, management proposals still attracted an average 96.3
per cent of the votes cast in 2010, and there is no evidence that the
number of shareholder proposals tabled has increased at all, all else being
equal. This still-limited scope of shareholder participation can partly be
explained by the concentrated ownership structures of Continental
European firms in particular. The presence of controlling-interest share-
holders, as well as deviations from the one share-one vote principle, lead
shareholder engagement at general meetings 317
to ‘rational apathy’ among minority shareholders. We find that invest-

ment funds and other pressure-insensitive institutional investors are
prepared to vote against management proposals, and their own pro-
posals enjoy relatively strong support from other shareholders.
Nonetheless, they remain – and post-crisis have increasingly been –
criticised for not being sufficiently engaged by both regulators and
academics (McCahery et al. 2010; European Commission 2011).
We confirm that national regulation plays a very significant role in
galvanizing shareholders, lending strong support to the provisions of the
Directive. Dissent against management proposals is significantly greater
when shareholders can freely trade their shares and exercise their voting
rights, including when there is no share blocking, record date restrictions
are reduced, and electronic voting is permitted. Shareholder proposals
become more frequent when minimum ownership requirements are
reduced, and shareholders are better able to access information and
communicate with other shareholders. The voting support for share-
holder proposals is also increased by the abolition of share blocking.
Finally, we find critical evidence that shareholders ultimately use their
votes to address governance concerns at the firm level when concerned
about the general governance and institutional environment. All else
being equal, management proposals are actually met with less dissent in
countries that are ranked highly in the composite index we construct
from the World Bank’s Worldwide Governance Indicators. The proba-
bility of shareholder proposals increases in both the World Bank index
and the quality of minority investor protection as measured by Djankov
et al. (2008). The actual success of these proposals, however, is related
negatively to both these measures. This implies that allowing share-
holders to raise their voice at general meetings is a part of good govern-
ance, and the shareholders themselves are discerning enough that they
will not do so unless deemed necessary. We conclude that not only is the
Directive a move in the right direction, but that national regulators
should go beyond the minimum standards introduced by the Directive
to support shareholder rights.
The remainder of this chapter is outlined as follows. Section 2 reviews
the theoretical and empirical evidence on shareholder engagement at
general meetings, and discusses the provisions of the Directive. Section 3
provides a detailed discussion of our sample, and observes recent devel-
opments in shareholder participation at European meetings. The multi-
variate analysis of proposal submissions and their outcomes follows in
Section 4. Section 5 finally allows for some concluding remarks.
2. Background
2.1. The role of shareholder engagement in corporate governance
Shareholder interventions in corporate governance can be placed on a
continuum of responses that shareholders can give to concerns over
managerial performance and governance quality. At one extreme, share-
holders can simply vote with their feet by selling their shares. At the
other extreme is the market for corporate control, where investors
initiate takeovers and buyouts to bring about fundamental corporate
changes (Gillan and Starks 2007).
The role of shareholder interventions as a disciplinary mechanism has
historically been widely debated. Bebchuk (2005) argues that it has an
important role in mitigating the agency problems associated with man-
agerial decisions. Harris and Raviv’s (2010) theoretical paper agrees, by
showing that when agency concerns are exacerbated in the firm, it is
always optimal that shareholders seek control over corporate decisions.
Opposing arguments nonetheless remain, especially in the legal litera-
ture. Lipton (2002) and Stout (2007) argue that shareholders can be beset
with conflict of interest motivations, or simply be too uninformed to
make effective governance decisions. Bainbridge (2006) goes as far as
claiming that activist shareholders can damage the firm outright by
disrupting the authority of the board of directors, and infers that share-
holder voice should actually be restricted.
Despite these concerns, regulators have actively promoted share-
holder engagement at company general meetings since the onset of the
Global Financial Crisis. Indeed, Masouros (2010) argues that there is a
clear pro-shareholder tendency around the world, despite marked differ-
ences in national corporate governance regimes. The United States (US)
led the charge with the adoption of the Dodd-Frank Wall Street Reform
and Consumer Protection Act in July 2010, and the subsequent measures
taken by the Securities and Exchange Commission (SEC) to (i) introduce
say-on-pay and say-on-golden-parachutes provisions, (ii) permit share-
holder proposals on CEO succession, and (iii) allow certain shareholders
proxy access for director nominations, subject to majority consent.1 The
Directive was adopted in 2007, so it actually predates the crisis, but it also
promotes shareholder voice at general meetings. Many European
1
The SEC originally introduced a new Rule 14a-11 in 2009, which automatically allowed
qualifying shareholders to nominate directors. However, the rule was vacated by the
District of Columbia Court of Appeals by 2011.
countries have since updated their corporate governance codes not only
to transpose the Directive but to implement further reforms.2
Governance codes have also been updated in Nigeria, the Philippines
and the United Arab Emirates, among many others.
The recent US literature implies that efforts to promote shareholder
engagement at company general meetings point in the right direction.
Ertimur et al. (2010) and Renneboog and Szilagyi (2011) show that
shareholder proposals submitted to meetings tend to target firms that
underperform and have poor governance structures. The authors find no
evidence of systematic agenda-seeking by activists, as well as show that
the voting shareholders tend only to support proposals with discernible
control benefits. They also argue that the control benefits of shareholder
interventions are at least partly realised from the reputational pressure
imposed on management, rather than the interventions themselves.
Indeed, Buchanan et al. (2012) find that firms targeted by shareholder
proposals are subsequently more likely to replace their CEOs and elect
independent board chairmen. Each of these studies reports that share-
holder interventions with clear control benefits are met with positive
market reactions.
The empirical evidence on the benefits of behind-the-scenes inter-
ventions circumventing general meetings is decidedly more mixed.
In the US, pension funds shifted towards private negotiations with
management in the early 1990s, although their interventions are rela-
tively non-controversial despite some concerns (Woidtke 2002). Private
engagement by hedge funds and other investment funds has been a much
more contentious issue. Hedge funds are well-known to rely on con-
troversial activist strategies, whereby they take positions in underper-
forming firms and target management directly. A source of concern has
been that these interventions may push towards short-rather than long-
term gains, resulting in investment inefficiencies and excessive leverage
(Brav et al. 2008; Clifford 2008; Becht et al. 2009; Greenwood and Schor
2009; Klein and Zur 2009; Bradley et al. 2010).
Empirical research on the role and benefits of shareholder engagement
in Europe remains relatively rare. Buchanan et al. (2012) compare share-
holder proposals submitted to general meetings in the US and the UK,
and find that while there are systematic differences in the proposal
objectives, the sponsor identities as well as the voting outcomes, the
target firms tend to be poorly performing in both countries. This is
2
See www.ecgi.org/codes/all_codes.php.
confirmed by Cziraki et al. (2011), who examine shareholder proposals

submitted in both the UK and Continental Europe. The authors also
highlight, however, that shareholder interventions at European company
meetings are not met with positive market reactions and remain rela-
tively rare. Indeed, De Jong et al. (2006) find no evidence at all of
proposals submitted by shareholders to Dutch meetings, and even report
little shareholder opposition to submissions made by management.
A few other studies report evidence on shareholder interventions
outside general meetings. Girard (2009) studies activist strategies in
France, and finds that civil lawsuits are the preferred method of activists
engaging firms over governance concerns, and that this particularly
aggressive strategy is also the most likely to succeed. Armour (2008)
develops a taxonomy of shareholder activism in the UK, and conversely
finds that informal private and public enforcement is significantly more
prevalent than formal enforcement. Becht et al. (2009) examine the
strategies of a single UK investor, the Hermes UK Focus Fund, and
confirm that the fund predominantly pursues behind-the-scenes nego-
tiations with management. The authors attribute the success of this
strategy to the credible threat that if management refuses to negotiate,
the fund will call an extraordinary meeting, with the looming prospect of
a proxy fight. The credibility of this threat is underpinned by the fact that
unlike in the US, proposals that pass the shareholder vote are legally
binding in the UK, as in most of Europe, and shareholders can remove
directors by an ordinary proposal.
2.2. Participation at European general meetings

General meetings are the formal forum where firms present relevant
matters to shareholders, and where shareholders vote upon these matters
and put questions to management. However, shareholder absenteeism
remains significant in much of Europe. In the market-oriented corporate
governance regime of the UK, the turnout rate is 68 per cent on average
(Hewitt 2011), while the turnout of companies’ free float – shares not
held by managers, directors or controlling shareholders – is 40–52 per
cent (European Commission 2006). In the stakeholder-oriented gover-
nance regimes of Continental Europe, shareholders are far less engaged.
Turnout rates are less than 60 per cent on average and below 50 per cent
in Belgium, Denmark, Norway and Switzerland. The gap is even more
pronounced in the turnout of companies’ free float, which stands at only
17 per cent in France, 10 in Germany and 4 in Italy.
The obstacles that limit shareholder engagement at European com-

pany meetings have long been part of the dialogue on the future of
European corporate governance. Zetzsche (2008) argues that low turn-
out rates in Continental Europe are partly driven by concentrated own-
ership structures, which have historically remained in place due to poor
shareholder protection (La Porta et al. 1998; Martynova and Renneboog
2008). Dominant shareholders have strong incentives to participate and
vote at meetings, which should technically boost turnout levels.
However, their presence exacerbates ‘rational apathy’ among minority
investors, i.e. the perception that their vote would make little difference.
Indeed, while US firms tend to have widely dispersed ownership struc-
tures, their average turnout rate is 82 per cent. Ownership is slightly
more concentrated in the UK and significantly more concentrated in
Continental Europe. Of UK listed firms, 63 per cent are regarded as being
widely held, and the typical voting block is twice the size of that in the
US, at around ten per cent. In contrast, the largest voting blocks often
constitute controlling interest in Continental Europe, reaching 20 per
cent on average in France, 44 per cent in the Netherlands, and 57 per cent
in Germany (Becht and Mayer 2001; Goergen and Renneboog 2001;
Faccio and Lang 2002). The types of blockholders present are also
quite different. Blockholders in the US and the UK tend to be corporate
insiders and institutional investors (Becht 2001). In contrast, 50–60 per
cent of Continental European firms are effectively owned by families,
and many are controlled by banks that both sit on the board and extend
their voting power by voting the shares deposited with them (Nibler
1998; La Porta et al. 1999; Barca and Becht 2001; Franks and Mayer 2001;
Faccio and Lang 2002).3 Table 7.1 shows that Continental European
firms also often deviate from the one share-one vote principle by grant-
ing multiple voting rights, introducing voting right and ownership ceil-
ings, and creating pyramidal and cross-ownership structures.4
The more immediate concern of European regulators is that due to a
variety of reasons, minority shareholders in Continental Europe must
pay significant costs to exercise their participation rights, without enjoy-
ing the economies of scale that concentrated owners do. These costs
3
Goergen and Renneboog (2001) point out that in the UK, blank proxies are controlled
and can be voted by the board of directors.
4
Roe (2004) adds that major creditors and employees are often given board representation
in Continental Europe, implying a conflict of interest between the board and outside
minority shareholders.
Table 7.1 The use of control-enhancing mechanisms
Multiple Non-
voting Non- voting Voting
rights voting preference Pyramidal Priority Depository right Ownership Golden Cross- Shareholders
Country shares shares shares structures shares certificates ceilings ceilings shares shareholdings agreements
Belgium 0% 40% 0% 0% 0% 0% 0% 0% 25%
France 55% 0% 0% 25% 0% 20% 10% 5% 20% 15%
Germany 20% 15% 0% 5% 10% 0%
Greece 5% 15% 20% 0% 5%
Italy 0% 30% 45% Unclear 30% 20% 5% 40%
Netherlands 42% 11% 11% 21% 0% 0% 0% 11% 5%
Spain 0% 20% 35% 5% 15% 0% 5%
Sweden 80% 65% 0% Unclear 5% 25% 5%
UK 5% 0% 50% 0% 0% 10% 10% 0% 5%
Note: This table presents control-enhancing mechanisms used by European companies. The percentages show the percentage of listed firms
examined that use each mechanism; where percentages are not shown, the mechanism is not permitted. Multiple voting rights shares are shares
giving different voting rights based on an investment of equal value. Non-voting shares are shares that carry neither voting rights nor special cash
flow rights. Non-voting preference shares are shares that carry special cash flow rights but no voting rights. Pyramidal structures are chains of
companies, where an entity (a family or a company) controls a company that in turn controls another company. Priority shares are shares holding
powers of decision or veto rights, irrespective of the proportion of equity holding. Depository certificates are financial instruments issued to
represent underlying shares, which are held by a foundation that administers the voting rights. Voting right ceilings are restrictions prohibiting
shareholders from voting above a certain threshold. Ownership ceilings are restrictions prohibiting shareholders from taking ownership above a
certain threshold. Golden shares are priority shares issued for the benefit of governmental authorities. Cross-shareholdings are structures where
companies holds equity stakes in each other. Shareholders agreements are formal and/or informal shareholder alliances.
Source: Shearman & Sterling (2007).
include not only procedural, but information and decision-making costs.

The European Commission (2006), the OECD (2007) and Georgeson
(2008) report that the main impediments to shareholder engagement in
Europe have been (i) limited access to information about upcoming and
past meetings; (ii) limited access to meetings, e.g. through share block-
ing, which requires participating shareholders to deposit their shares;
(iii) restrictions on proxy allocation and distance voting; and (iv) restric-
tions on shareholder engagement, including the right to ask questions,
call general meetings, and submit shareholder proposals. A summary of
these impediments, as reported for the pre-Directive period in these
studies, is shown in Table 7.2.
Cross-border investors find the procedural and information costs of
meeting participation particularly burdensome. Cross-border invest-
ment has been actively stimulated by the EU to create integrated finan-
cial markets, and has now reached over 40 per cent of market
capitalisation on average (FESE 2008). However, foreign attendance at
general meetings remains poor. Foreign investors typically hold their
shares through accounts with securities intermediaries, which in turn
hold accounts with other intermediaries and central securities deposi-
tories. To vote in absentia, they must go through global custodian banks,
or their proxy vendors, which in turn must engage proxy-related services
from local market subcustodians (OECD 2011). The European
Commission (2006) argues that overall, the administrative costs of
cross-border voting are twice the costs of domestic voting, and are
largely prohibitive for foreign investors.
2.3. The Shareholder Rights Directive

The impediments to shareholder participation at general meetings were
widely recognised during the public consultation launched by the
European Commission in 2004, and subsequently formed the motiva-
tion behind the adoption of the Shareholder Rights Directive (Directive
2007/36/EC) in July 2007.
The Directive expressly states that effective shareholder control is a
prerequisite to sound corporate governance and should, therefore, be
facilitated. Its declared objective is ‘to strengthen shareholders’ rights, in
particular through the extension of the rules on transparency, proxy
voting rights, the possibility of participating in general meetings via
electronic means and ensuring that cross-border voting rights are able
to be exercised’. The key provisions of the Directive include:
Table 7.2 Statutory requirements with respect to general meetings
Notice Form of Share Submit Proposal Proxy Voting Electronic Post-GM
period notice blocking Record date proposals deadline Call EGM representation by post voting dissemination
Belgium 24 days; mailed gazette, local 3–6 workdays s.t.a.; 20%; 5% is 20% may yes on request
at 15 days paper, mail 5 workdays, advised limit to
max. 15 shareholder or
days spouse
France 35 days, ‘notice gazette, website no; shares 3 days 0.5–4%, depends 25 days; 5 5% to appoint spouse or yes s.t.a.
of call’ at 15 plus mail and immobilised on firm size days court shareholder; no
days; 15 if email at 5 days from representative permanent
takeover notice if to convene
takeover
Germany 1 month gazette, mail 21 days all on agenda; 10 days 5% permanent for instruction to register, on
5%/€500k from bearer shares proxy request
for new items notice representative
Greece 20 days gazette, national 5 days 5 days 5% 5% must submit 5 days
paper before GM
Italy 30 days; 20 if gazette or paper s.t.a., min. 2 s.t.a., 2 days 2.50% 5 days from 10%, lower s.t.a.; no permanent; s.t.a. s.t.a. Consob, on
GM called by days notice not on certain restricted at coop request
shareholders issues banks
Netherlands 15 days national paper or s.t.a., 7 days optional for 1% or €50m 30–60 days 10%, must apply individual yes website, on
letter if all AGM, min. at Amsterdam request
shareholders 7 days Court
known
Spain 15 days gazette and s.t.a., min. 5 5 days 5% 5 days from 5% can be restricted by yes yes website; no
provincial days notice articles vote count
paper
Sweden 28 days; 14 for gazette and 5 days all 7 weeks 10% up to 1 year yes yes
some EGMs; national
max. 6 weeks paper
UK 21 days; 14 days post, news 48 hrs 5%, or 100 6 weeks or 10% no obstacles yes yes website, LSE,
for EGM services shareholders when on request
through LSE with GBP100 notice
paid-up given
shares
Note: This table presents statutory requirements with respect to general meetings (GM). Notice period is the number of days that must pass between the (last) publication of a convocation and the day of the
meeting. Share blocking is the number of days before a meeting that shareholders must deposit their shares. Record date is the number of days before a meeting that the register of shareholders is closed. Submit
proposals and Call EGM are the minimum ownership required to place items on the agenda of a meeting and call an extraordinary meeting, respectively. Proposal deadline is the deadline before a meeting for
shareholders to submit proposals. Proxy representation shows provision on the appointment of proxies to vote. s.t.a. is subject to articles of association.
Source: European Commission (2006) and Georgeson (2008).
* a minimum notice period of 21 days; if shareholders agree in a public

vote, this can be reduced to 14 days if electronic voting is permitted;
* internet publication of the convocation and any documents submitted
to the GM at least 21 days before the GM;
* abolition of share blocking, and introduction of a record date that may
not be more than 30 days before the GM;
* abolition of obstacles to voting by post and electronic voting;
* right to ask questions and obligation on the part of management to
answer questions;
* abolition of constraints on eligibility to act as proxy holder and of
excessive requirements for the process of proxy appointment;
* the possibility that shareholders put items on the agenda and table
draft resolutions for items on the agenda, with a minimum ownership
requirement that does not exceed five per cent of the company’s share
capital;
* disclosure of voting results on the firm’s internet website.
While the Directive has generally been regarded as a move in the right
direction, it continues to be criticised for imposing only minimal stand-
ards that still fail to ensure a level playing field for all shareholders. For
example, Davies et al. (2011) argue that cross-border shareholders often
remain uninformed about future meetings, and their votes often
unexercised, or exercised by others, due to the intermediaries they go
through. Masouros (2010) adds that the provision on shareholder pro-
posals is ‘empty letter’, because the five per cent ownership threshold is
still highly prohibitive, and shareholders are unable to communicate and
form coalitions due to a lack of infrastructure for proxy solicitation and
even access to share registries. Many countries even dragged their feet
over the transposition of the Directive itself. Although the 27 EU mem-
bers were required to transpose by August 2009, 14 had not completed
the process by January 2010, and the European Commission threatened
action against nine of them in April 2010 by issuing reasoned opinions.
In some cases, this may well have reflected government concerns that the
Directive was unduly facilitating shareholder engagement and activism
at general meetings.
2.4. Shareholder activism at European general meetings

Shareholder activism in the form of tabling proposals has historically
been a prominent feature of US general meetings. Shareholders in the US
are not permitted to call extraordinary meetings unless the corporate

charter or bylaws allow otherwise. However, each year between 1996 and
2005, an average 14.1 per cent of S&P 1500 firms were targeted by
shareholder proposals, peaking at 21.3 per cent in 2003 (Renneboog
and Szilagyi 2011). This is largely due to the SEC’s fairly liberal Rule
14a-8 governing shareholder proposals, which allows submissions to be
made by any shareholder owning USD2,000, or one per cent of voting
shares.
While there is now ample evidence that shareholder proposals play a
useful and relevant role in US corporate governance (Ertimur et al. 2010;
Renneboog and Szilagyi 2011; Buchanan et al. 2012), the Directive
stopped short of truly encouraging proposal submissions in Europe.
The five per cent minimum ownership requirement is indeed quite
stringent and, as also shown in Table 7.2, had already been met by all
EU Member States except Belgium. This cautious approach may be due
to ongoing concerns that shareholder activism can come at a cost.
European policymakers also often argue that US lessons on the govern-
ance role of shareholder proposals may not be readily applicable in the
European context. First, proposals in the US are non-binding even if they
pass the shareholder vote, whereas they are legally binding in Europe,
with some exceptions (notably the Netherlands). Second, cross-country
variations in the regulation of proxy solicitation may affect the incentives
of and costs borne by activist shareholders. And third, the market-
oriented Anglo-American governance model is, indeed, quite different
from the stakeholder-oriented regimes of Continental Europe. As has
been discussed, minority shareholders enjoy much better protection
under US and UK common law, and are better incentivised and equip-
ped to challenge management in the absence of controlling-interest
shareholders.
Masouros (2010) describes the procedural barriers that are likely to
prevent shareholder proposals from becoming more prevalent in both
the UK and Continental Europe. Activists must build sufficient support
for their proposals to pass, and even form coalitions just to make sub-
missions if they do not meet the stringent ownership requirements.
However, potential allies are difficult to identify.5 For example, UK
5
Crespi and Renneboog (2010) point out that minority shareholders may even be reluctant
to build long-term coalitions, because they are subject to ‘acting in concert’ rules, and
regulators may end up regarding them as a single blockholder that has to comply with
regulations on disclosure, mandatory bids etc.
shareholders must hold ten per cent equity to order an inquiry into who
the ultimate shareholders are, German shareholders do not have the
right to inspect share registries at all, and registries do not even exist in
the Netherlands because all listed shares are bearer shares. As shown in
Table 7.2, the deadlines set for proposal submissions can also be fairly
tight for activists to reflect on the agenda and submit additions. Finally,
shareholder proposals in the UK, as in the US, can be included in the
firm’s proxy documents and distributed to shareholders at no major cost
to the activist. In other countries, however, proxy solicitation at the
firm’s expense is prohibited.
It is important to remember that in Continental Europe, the investor
base that is likely to submit and lend voting support to shareholder
proposals is relatively narrow. Foreign shareholders tend to be institu-
tional investors, but they often face prohibitively high voting costs. Of
domestic institutions, pension funds, insurance firms and investment
funds hold 41 per cent of equities in the UK, but only 29 per cent in
France, 14 per cent in Germany and Italy and 8 per cent in Spain (FESE
2008). Many of these investors also pursue predominantly passive
investment strategies, preferring to vote with their feet by selling their
shares. McCahery et al. (2010) find that 80 per cent of institutional
investors would consider selling rather than engaging, and while 66 per
cent would vote against management to address governance concerns,
only ten per cent would voice their concerns publicly. Indeed, institu-
tional investors have often been criticised post-crisis for their passivity
and not doing enough due diligence.
3. Analysis of management and shareholder proposals

In order to investigate the scale and scope of shareholder engagement
and dissent at European general meetings, we now examine both man-
agement and shareholder proposals submitted in 17 European countries
during the period between 2005 and June 2010. While comprehensive
data on meeting attendance rates are largely inaccessible, we can inves-
tigate (i) what drives the level of shareholder dissent over management
proposals; (ii) why firms are targeted by shareholder proposals; and (iii)
what drives the level of voting support attracted by shareholder pro-
posals. We relate these issues not only to proposal and firm character-
istics, but to country-level regulation potentially affecting shareholder
participation at company meetings.
Data from the meetings were gathered from the Manifest and
International Shareholder Services (ISS) databases for the periods
2005–07 and 2008–10, respectively. Each database covers, although not
exhaustively, firms that are members of the main market indices in each
sample country: ATX20 (Austria), BEL20 (Belgium), OMXC20 (Denmark),
OMX-H25 (Finland), SBF120 (France), DAX30 and MDAX50 (Germany),
ASE20 (Greece), ISEQ General (Ireland), FTSE MIB and MIDCAP (Italy),
LuxX (Luxembourg), AEX25 and AMX25 (Netherlands), OBX25
(Norway), PSI20 (Portugal), IBEX35 (Spain), OMXS30 (Sweden), SMI20
(Switzerland) and FTSE350 (UK). The total number of firms in the com-
bined sample is lower and increases over the sample period, for several
reasons. First, the coverage of Continental European firms by Manifest in
the early years of the sample period is limited.6 Second, some firms in the
national indices are incorporated in other jurisdictions. And third, we only
include proposals with available outcomes (either vote count or pass/fail) in
the analysis.7 Missing and ambiguous data on vote counts, the classification
of proposal objectives and the sponsors of shareholder proposals (not
reported in either database) were hand-collected and double-checked
using Factiva and company filings.
3.1. Number of proposals

Table 7.3 shows the number of proposals and general meetings covered
in our combined sample over the sample period between 2005 and June
2010. The Table shows that we have proposal data from 3,484 general
meetings, including 3,088 annual meetings and 396 extraordinary meet-
ings of 921 firms.
6
The two databases overlap for 2007 and do not in fact provide the same coverage. To
ensure consistent coverage, our combined dataset contains the set of companies that
appear in both databases for the overlapping year, and then tracks additions to/removals
from this set. The complete Manifest database actually contains 171,730 proposals
submitted between 1996 and 2008, at 19,055 general meetings of 2,885 firms. It also
covers a significantly higher number of UK firms than ISS. However, the proposal
outcomes are unavailable for 40 per cent of these proposals, and the database covers
the UK and Ireland only for the period before 2005.
7
In some countries, dissemination of the voting results is not compulsory. Manifest (2008)
reports that the dissemination of the voting results has historically been best practice in
the UK, with the disclosure rate at 96 per cent among the FTSE 250 firms. In Continental
Europe, it has only recently become common practice even for the largest firms, with the
disclosure rate increasing between 2005 and 2007 from 51 to 100 per cent for the CAC
100 firms in France, and from 68 to 88 per cent for the AEX 25 firms in the Netherlands.
Table 7.3 Number of management and shareholder proposals in Europe by country and year
Management proposals (firms) Shareholder proposals (firms)
All 2005 2006 2007 2008 2009 2010 All 2005 2006 2007 2008 2009 2010
All 42,170 (3,484) 3,875 (363) 4,754 (432) 6,118 (509) 9,706 (802) 8,534 (674) 9,183 (704) 329 (136) 1 (1) 23 (10) 79 (21) 88 (38) 57 (33) 81 (33)
Austria 623 (72) 37 (4) 40 (5) 85 (9) 164 (21) 129 (16) 168 (17) 5 (4) 2 (2) 3 (2)
Belgium 1,105 (90) 107 (11) 101 (11) 192 (16) 239 (21) 181 (15) 285 (16) 5 (1) 5 (1)
Denmark 644 (48) 163 (15) 159 (16) 322 (17) 41 (7) 3 (3) 4 (1) 34 (3)
Finland 929 (59) 1 (1) 241 (20) 336 (19) 351 (19) 27 (18) 11 (6) 8 (7) 8 (5)
France 7,505 (375) 306 (15) 604 (39) 1,062 (46) 1,797 (97) 1,931 (88) 1,805 (90) 57 (28) 1 (1) 21 (8) 11 (5) 15 (9) 9 (5)
Germany 4,328 (337) 389 (37) 486 (35) 632 (50) 1,051 (80) 864 (67) 906 (68) 82 (17) 1 (1) 38 (4) 26 (5) 9 (3) 8 (4)
Greece 519 (75) 6 (1) 23 (4) 46 (8) 138 (15) 160 (27) 146 (20)
Ireland 1,877 (175) 253 (24) 230 (23) 271 (28) 467 (43) 323 (25) 333 (32) 8 (6) 1 (1) 2 (1) 2 (2) 3 (2)
Italy 1,287 (318) 137 (29) 84 (18) 146 (33) 284 (73) 260 (68) 376 (97) 4 (4) 1 (1) 2 (2) 1 (1)
Luxembourg 378 (33) 6 (1) 27 (3) 73 (7) 105 (9) 78 (7) 89 (6)
Netherlands 2,240 (195) 147 (13) 249 (24) 389 (32) 522 (47) 432 (36) 501 (43) 5 (1) 5 (1)
Norway 739 (66) 16 (3) 23 (3) 238 (23) 203 (19) 259 (18) 8 (7) 3 (2) 1 (1) 4 (4)
Portugal 482 (51) 34 (2) 23 (3) 167 (17) 90 (11) 168 (18) 25 (11) 6 (3) 6 (2) 8 (3) 4 (2) 1 (1)
Spain 1,892 (161) 132 (15) 262 (27) 317 (25) 351 (29) 386 (33) 444 (32)
Sweden 1,213 (61) 4 (1) 423 (19) 373 (20) 413 (21) 22 (13) 12 (5) 3 (3) 7 (5)
Switzerland 1,127 (106) 62 (7) 166 (17) 162 (18) 312 (29) 203 (17) 222 (18) 6 (4) 3 (1) 1 (1) 1 (1) 1 (1)
UK 15,282 (1,262) 2,293 (206) 2,428 (220) 2,696 (230) 3,044 (244) 2,426 (190) 2,395 (172) 34 (15) 12 (4) 5 (3) 7 (3) 2 (1) 8 (4)
Note: This table shows the number of management and shareholder proposals submitted to firms in 17 European countries between 2005 and 2010. The number of firms the proposals were submitted
to is shown in brackets.
Source: Manifest, International Shareholder Services, own calculations.
At the sample meetings, 42,170 management proposals were put to

shareholder vote.8 There is significant variation in the number of pro-
posals per meeting across countries. In Italy, the average meeting had
just four management proposals, compared with 20 in France and
Sweden. The average number of proposals per meeting is 12.1 across
the whole sample, increasing over time from 10.7 in 2005 to 13.0 in 2010.
It is clear from Table 7.3 that shareholder proposals remain relatively
rare in Europe. For the 3,484 general meetings, we find only 329 share-
holder proposals, submitted at 136 meetings of 87 firms. Most of these
proposals were submitted in Germany (82), France (57), Denmark (41)
and the UK (34). However, the countries where firms were most likely to
be targeted are Portugal (22 per cent of general meetings) and the Nordic
countries of Finland (31 per cent), Sweden (21 per cent), Denmark (15
per cent) and Norway (11 per cent). We find no shareholder proposals
submitted to the general meetings held in Greece (75), Luxembourg (33)
and Spain (161). Table 7.3 shows that the frequency of shareholder
proposals increased over time, indicating a gradual rise in activist inter-
ventions in Europe, with 0.3 per cent of general meetings targeted in
2005, 2.3 in 2006, 4.1 in 2007 and 4.7 in 2010.
To put these findings into context, it is useful to revisit Renneboog and
Szilagyi’s (2011) analysis of shareholder proposals submitted in the US
to S&P1500 firms. The authors examine an earlier sample period of 1996
to 2005, and find that of 10,590 general meetings, 1,494 (or 14.1 per cent)
were targeted with 2,436 proposals. The percentage of meetings targeted
also increased in the US over time, from 11.2 per cent in 1996 to 16.1 per
cent in 2005, with a peak of 21.3 per cent in 2003. These findings clearly
show that, on the whole, the use of shareholder proposals to confront
management has historically been much more prevalent in the US.
3.2. Management proposals: characteristics and voting outcomes

Table 7.4 reports the number and voting success of both the manage-
ment and shareholder proposals stratified by a variety of proposal
characteristics. Voting success is defined as the number of votes cast in
8
The final sample excludes 234 proposals, because their three-way voting outcomes cannot
be interpreted like those of other proposals. Of these, 177 submissions were director or
auditor nominations submitted under Italy’s multiple-winner voting system (see Belcredi
et al. 2012). Another 49 of these proposals were submitted in France, mostly to elect a
representative of employee shareholders to the board.
Table 7.4 Votes for management and shareholder proposals in Europe
Proposal type Management proposals Shareholder proposals

N Mean Median N Mean Median
All 38,564 96.3 99.3 251 35.3 23.7
Panel A: Country
Austria 611 98.5 99.9 5 24.5 1.9
Belgium 1,082 96.6 99.8 4 70.6 74.1
Denmark 21 100.0 100.0
Finland 466 98.9 99.8 17 75.1 84.9
France 7,382 93.7 98.2 53 36.4 29.1
Germany 4,247 97.4 99.5 82 11.8 2.8
Greece 376 97.0 99.8
Ireland 1,656 95.8 99.5
Italy 1,204 98.1 99.8 4 48.6 49.4
Luxembourg 345 98.3 99.9
Netherlands 2,142 96.4 99.3 5 60.1 67.9
Norway 623 97.5 99.9 8 30.0 16.5
Portugal 409 97.5 99.9 25 86.3 91.0
Spain 1,727 97.8 99.6
Sweden 111 99.2 100.0 5 36.1 34.0
Switzerland 956 97.2 99.1 5 15.2 0.3
UK 14,955 97.6 99.4 30 36.5 41.7
Panel B: Management recommendation
For 38,296 96.7 99.4 34 91.3 91.6
None 14 88.4 95.3
Against 3 11.8 13.3 217 26.5 10.5
Panel C: Meeting type
Annual 37,226 96.7 99.4 201 35.5 24.1
Extraordinary 1,087 96.8 99.5 50 34.3 16.5
Panel D: Year
2005 3,696 97.6 99.5 1 7.3 7.3
2006 4,619 97.4 99.4 23 45.7 54.3
2007 5,947 96.7 99.4 79 26.7 2.9
2008 8,527 96.9 99.5 64 31.8 14.7
2009 7,312 96.0 99.2 44 47.5 34.6
2010 8,212 96.3 99.2 40 39.2 26.2
Table 7.4 (cont.)
Proposal type Management proposals Shareholder proposals

N Mean Median N Mean Median
All 38,564 96.3 99.3 251 35.3 23.7
Panel E: Shareholder proposal presented at the meeting
No 36,868 96.8 99.4
Yes 1,445 94.3 98.7
Panel F: Dissent at a previous meeting
No 35,189 97.0 99.4 235 31.9 16.9
Yes 3,124 93.5 98.7 16 84.9 86.0
Panel G: Proposal objectives
Operational 11,668 98.5 99.7
Elect directors 9,678 97.0 99.2 66 49.7 42.4
Discharge directors 2,203 97.1 99.4
Remove directors 21 27.6 15.1
Board governance 1,327 97.9 99.7 52 46.1 43.0
Adopt anti-takeover device 102 76.2 76.9
Repeal anti-takeover device 9 95.2 99.1 9 40.0 35.0
Voting and disclosure 497 96.8 98.5 18 26.9 3.8
Compensation 3,014 91.7 96.2 19 25.9 18.7
Capital 8,306 95.9 99.3 4 8.5 5.7
Restructuring 715 94.8 99.5 18 18.9 5.7
Dividends 7 6.6 1.9
Social 439 96.5 97.7
Other 606 37 23.3 4.1
Panel H: Proposal sponsors
Pension funds 7 30.8 27.4
Investment funds 76 51.3 46.6
Banks 3 94.2 93.8
Companies 23 65.9 83.8
Employees 20 20.1 17.5
Dissidents 2 26.7 26.7
Shareholder associations 6 37.3 35.0
State 18 82.3 90.9
Individuals/other 96 8.0 2.6
Note: This table shows the percentage votes cast in favour of proposals, taken from
the three-way voting results (for/against/abstain).
Source: Manifest, International Shareholder Services, own calculations.
favour divided by the total number of eligible votes. Eligible votes include
abstentions, because any vote not cast in favour can be interpreted as
shareholder dissent. The Table shows that the vote counts are available
for 38,564 management proposals and 251 shareholder proposals. For
the remaining proposals, which include most of the proposals submitted
in Denmark, Finland and Sweden, only the pass/fail outcomes are
known.
Panel A of Table 7.4 shows little objection to management proposals
in all 17 countries, with a mean 96.3 and median 99.3 per cent of the total
votes. In fact, only 255 of the sample proposals, submitted to 167 meet-
ings, failed to pass the shareholder vote. The voting outcomes were the
weakest in France (93.7 per cent), Ireland (95.8 per cent) and the
Netherlands (96.4 per cent). Interestingly, they were the strongest in
Denmark (100 per cent), Sweden (99.2 per cent) and Finland (98.9 per
cent), which may indicate that firms in these countries withhold vote
counts unfavourable to management. It is notable, however, that activist
interventions were among the most prevalent in these same countries.
Hewitt (2011) reports that in the US, management proposals achieve an
average 93.2 per cent of the votes.
Panel B of the Table provides further evidence that European share-
holders tend to vote in line with management. Voting support was 96.7
per cent on average when management recommended a vote in favour,
and a respective 88.4 and 11.8 per cent in the rare cases when it made no
recommendation on a proposal or recommended a rejection.9 Panel C
shows that the voting outcomes were comparable in annual and extra-
ordinary meetings.
There is some evidence in Panel D that voting dissent is on the rise at
European general meetings. The average voting support for management
proposals declined from 97.6 per cent in 2005 to 96.3 per cent in 2010 in
the sample – in fact, 215 of the 255 failed proposals were submitted in the
latter half of the sample period. Importantly, Panels E and F show that
public opposition by shareholders to management goes at least some way
in swinging voter sentiment on management proposals. The votes in
favour fell to 94.3 per cent on average when a shareholder proposal was
presented simultaneously, and 93.5 per cent when management had
actually already been defeated at a previous meeting, i.e. a management
9
The three proposals which management did not support were submitted in France;
management had to table these proposals due to regulatory requirements.
proposal had failed or a shareholder proposal contested by management

had passed.
Panel G of Table 7.4 shows that the voting outcomes on management
proposals are strongly affected by the proposal objectives. We classify
both management and shareholder proposals into mutually exclusive
categories: (i) operational and routine issues; the (ii) election, (iii) dis-
charge from liability or (iv) removal of directors; (v) board governance;
(vi) adoption or (vii) repeal of anti-takeover devices; (viii) voting and
disclosure issues; (ix) executive compensation; (x) capital authorisations;
(xi) corporate restructuring; (xii) dividend policy; and (xiii) social issues.
The proposals classified as being on director removals and dividend
policy were all shareholder proposals. Management proposals on divi-
dend and income allocation are part of the regular course of business and
therefore classified as operational proposals.
Unsurprisingly, the results show that of the various types of manage-
ment proposals, the routine operational proposals enjoyed the most
voting support, with an average 98.5 per cent of the votes. These include
proposals to approve annual accounts and audit reports, dividend and
income allocation, article amendments, and auditor appointments.
Proposals on board governance and the election and discharge of direc-
tors also received more than 97 per cent of the votes.
Evidence of shareholder dissent was strongest for proposals on execu-
tive compensation and anti-takeover devices. Compensation-related
submissions received only 91.7 per cent of the votes, while proposals to
adopt provisions blocking potential takeover attempts received 76.2 per
cent. This latter result is unsurprising. The market for corporate control
facilitates managerial accountability to minority shareholders, and take-
over bids generate high shareholder returns in the range of 25–35 per
cent even in Europe (Martynova and Renneboog 2008; 2011a).10
3.3. Shareholder proposals: characteristics and voting outcomes

Table 7.4 clearly shows that shareholder proposals submitted to
European firms do not attract a great deal of voting support. The average
proposal received 35.3 per cent of the votes, although Panel A shows
substantial variation across countries: the votes ranged from 11.8 per
10
Shareholders prefer that their firm is not entrenched against a possible takeover even in
Europe, as an acquisition may generate high returns, typically in the range of 25–35 per
cent (Martynova and Renneboog 2008; 2011a; b).
cent in Germany and 15.2 per cent in Switzerland to 75.1 per cent in
Finland and 86.3 per cent in Portugal. Some of these country outcomes
are driven by small sample sizes and limited diversity in the proposal
objectives and types of proposal sponsors. For example, 12 of the 17
Finnish proposals were submitted by the Finnish government, which has
a competitive advantage in proxy solicitation, and sought the establish-
ment of a nominating committee on the board. Similarly, while 24 of the
25 Portuguese proposals actually passed, 19 were sponsored by control-
ling owners such as firms, banks and wealthy individuals, and 18 were
actually supported by management. Possibly due to such issues, Panel D
of the table shows no discernible trend in the voting success of share-
holder proposals over time.
Table 7.4 confirms that the shareholder proposals supported by man-
agement enjoyed very strong voting success. All 34 management-
approved proposals passed the shareholder vote, whereas those opposed
by management received only 26.5 per cent support. Once again, we find
that shareholder dissent is greater if there is a history of public opposi-
tion to management: shareholder proposals attracted an average 84.9 per
cent of the votes when management had already been defeated at a
previous meeting. Panel C confirms that the voting outcomes were
comparable at annual and extraordinary meetings.
Finally, Panels G and H stratify the number and success of shareholder
proposals by proposal objective and the type of sponsoring shareholder.
Panel G reports that a third of the proposals nominated new directors or
sought to remove existing ones, and another fifth targeted the quality of
board governance. These submissions attracted substantial support, at a
respective 49.7, 27.6 and 46.1 per cent, respectively, of the votes on
average. Submissions targeting anti-takeover devices were relatively
rare in the sample, but they also enjoyed a significant 40 per cent
support. Conversely, proposals calling for restructuring of the target
firm and capital and dividend changes received only 18.9, 8.5 and 6.6
per cent of the votes, respectively.
Panel H shows that of the proposal sponsors, governments enjoyed
very significant voting support. Submissions were made by the Finnish,
French, German, Portuguese and Swedish governments, and received an
average 82.3 per cent of the votes. Affiliated companies and banks
similarly attracted 65.9 and 94.2 per cent, respectively, of the votes on
average. However, 14 of the 23 company proposals and two of the three
bank proposals were submitted by controlling owners and supported by
management.
A particularly important finding is that the submissions made by

institutional activists enjoyed considerable success. Aside from individ-
ual investors, investment funds were the most prolific proposal sponsors,
with 30 per cent of all submissions. They also attracted an average 51.3
per cent voting support, and 40 per cent of their proposals passed despite
opposition from management. Pension funds and shareholder associa-
tions received 30.8 and 37.3 per cent of the votes, respectively. These are
critical results, because they show that traditionally passive minority
shareholders are in fact receptive to institutional interventions in
Europe. Investment funds sponsored the most proposals in the three
biggest European markets, the UK (28), France (22) and Germany (12).
They mostly sponsored board-related proposals, often nominating or
seeking the removal of existing directors,11 but they also targeted anti-
takeover devices and sought restructuring of the target firm.
3.4. Shareholder proposals: comparison with the US

To provide a comparison of the characteristics and voting success of
European shareholder proposals, Table 7.5 reports details on the 2,436
US proposals examined in Renneboog and Szilagyi (2011) over the
period 1996 to 2005.
The Table suggests that shareholder proposals receive similar voting
support in Europe and the US, at an average 35.3 and 33.8 per cent of the
votes, respectively. However, the success of US submissions is in fact
greater. On the one hand, practically all US submissions were opposed by
management, with comparable proposals achieving only 26.5 per cent of
the vote in Europe; on the other, the US sample ends in 2005, the first
year of the European sample. The Table shows that the support attracted
by US proposals had actually increased over time, to 37.9 per cent by
2005.
Panel A of Table 7.5 shows that US proposals most frequently target
anti-takeover devices, followed by executive compensation and board
quality. Taken together, these constitute 75 per cent of all submissions
compared with 30 per cent in Europe. This is largely because proposals
seeking personal changes on the board, which are prevalent in Europe,
remain largely prohibited under the SEC’s Rule 14a-8. Panel A shows
11
Investment funds sought board seats mostly in the UK (20 proposals) and France (16).
Buchanan et al. (2012) discuss how UK shareholders can replace the board with their
own nominees by a simple majority vote.
Table 7.5 Number of shareholder proposals and votes for the proposals in the US
Repeal anti- Board Voting and

Proposal objectives All takeover governance disclosure Compensation Restructuring Other
N Mean N Mean N Mean N Mean N Mean N Mean N Mean
All 2,436 33.8 847 55.1 437 20.1 303 33.1 551 22.4 91 14.5 207 15.3
Panel A: Year
1996 181 29.4 72 43.8 54 20.8 25 25.7 20 12.5 3 17.6 7 13.8
1997 193 24.7 51 46.8 50 14.6 32 27.5 31 11.5 13 22.5 16 8.2
1998 188 27.7 58 49.7 35 20.1 39 29.4 23 7.9 17 10.3 16 7.9
1999 206 31.1 82 50.6 36 21.3 31 28.2 34 10.8 13 13.2 10 7.2
2000 185 33.3 77 52.7 35 20.2 20 35.1 18 10.6 21 17.0 14 10.4
2001 194 32.2 80 51.9 39 13.9 20 36.0 27 16.5 13 11.2 15 15.9
2002 214 38.6 98 57.9 36 19.1 23 35.7 23 18.1 34 19.3
2003 404 38.6 141 60.7 59 22.0 16 33.9 155 30.1 2 3.2 31 17.2
2004 362 35.0 105 61.4 54 23.7 31 28.8 129 25.2 5 20.8 38 18.9
2005 309 37.9 83 63.2 39 24.2 66 42.7 91 24.7 4 2.8 26 17.7
Panel B: Proposal sponsors
Pension funds 116 44.1 55 58.9 34 32.6 8 36.6 9 31.0 10 20.0
Investment funds 39 42.6 17 57.5 5 23.7 2 5.9 11 32.8 4 48.3
Companies 2 68.4 2 68.4
Coordinated investors 168 29.7 68 49.9 33 22.8 19 13.4 48 12.3
Unions 810 35.6 241 52.8 124 22.5 80 38.4 289 30.1 76 20.0
Socially responsible/religious 112 20.4 10 70.2 48 22.2 2 44.7 44 8.4 8 7.8
Individuals/other 1,189 33.1 454 56.2 193 15.2 213 30.9 188 14.7 32 11.4 109 11.0
Note: This table shows the number of shareholder proposals submitted to the S&P1500 firms in the US by year, issue and sponsor
type.
Source: Renneboog and Szilagyi (2011).
that anti-takeover proposals are by far the most successful in the US,
with an average 63.2 per cent of the votes in 2005. Such submissions,
mostly targeting classified boards, poison pills, golden parachutes and
supermajority provisions, remain relatively rare in Europe.
Panel B of Table 7.5 stratifies the US sample by sponsor type. The
results show that as in Europe, shareholder proposals are most fre-
quently submitted by individual investors. However, the similarities
are otherwise limited. In the US, the government and firms make neither
hostile nor friendly proposal submissions. Investment funds also rarely
submit, as they typically prefer to target management behind the scenes,
or they need to launch proxy fights if they seek a place on the board
(Szilagyi 2010). The panel reveals that in the US, the most prolific
institutional proposal sponsors are in fact unions and union pension
funds, engaging firms over a wide range of issues including anti-takeover
devices, executive compensation, voting issues, and board quality.
An important rule specific to the US market is that, in contrast with
European countries, firms have no obligation to implement shareholder
proposals even if they pass the shareholder vote. Nonetheless, proposals
passed are now implemented in most cases, with Renneboog and Szilagyi
(2011) reporting an implementation rate of 70.1 per cent for 2005.
Indeed, US firms ignoring proposals passed can suffer in a number of
ways, including by drawing negative press, receiving downgrades by
governance rating firms, or ending up on CalPERS’s ‘focus list’ of poor
performers. Ertimur et al. (2010) add that the directors of these firms are
also less likely to be re-elected and more likely to lose other directorships,
in many cases due to dissatisfied activists targeting director elections
with ‘just vote no’ campaigns (Del Guercio et al. 2008).
4. Multivariate analysis
To gain further insight into the drivers and success of shareholder
engagement at European shareholders’ meetings, we now use multi-
variate analysis to examine (i) what drives shareholder dissent over
management proposals (Section 4.2.); (ii) which firms are targeted by
shareholder proposals (Section 4.3.); and (iii) what drives the voting
success of these activist submissions (Section 4.4.). The analysis includes
extensive controls, defined in the Appendix, for meeting, proposal, firm
and country characteristics. Firm-level accounting and performance
data are taken from the Thomson ONE Banker and Datastream
databases, while data on firm ownership are collected from CapitalIQ

and company filings.
We use information reported by the European Commission (2006)
and Georgeson (2008) to capture country-level differences in share-
holder rights potentially relevant to shareholder participation at general
meetings. We control for (i) the notice period shareholders must be
given before a general meeting; (ii) the number of days before a meeting
that the register of shareholders must be closed; (iii) whether share-
holders must have their shares deposited i.e. blocked to attend a meeting;
(iv) whether firms can issue bearer shares; (v) whether shareholders have
the right to ask management questions prior to a meeting; whether
shareholders can vote (vi) by proxy and (vii) electronically; and (viii)
whether voting can be concluded by show of hands. For the analysis of
shareholder proposals we also control for (ix) minimum ownership
requirements that shareholders must meet to table proposals.
Two additional indices are included to capture governance quality at
the country level. We employ the anti-self-dealing index of Djankov et al.
(2008) to measure the quality of protection that minority shareholders
enjoy against expropriation by corporate insiders. We finally construct a
dynamic annual governance index for the general institutional and
governance environment, using the World Bank’s six Worldwide
Governance Indicators: (i) voice and accountability, (ii) political stability
and absence of violence, (iii) government effectiveness, (iv) regulatory
quality, (v) rule of law, and (vi) control of corruption (Kaufmann et al.
2010). The six indicators are totalled to form a single index for each year.
The country-level shareholder rights and corporate governance variables,
with the exception of the annual governance index, are cross-sectional and
predate the transposition of the Directive into national laws. While our
sample period ends in June 2010, EU Member States were required to
comply with the Directive by August 2009. However, only six of our sample
countries had completed the transposition process by January 2010, so the
Directive was largely not in force by the 2010 proxy season.12
4.1. Descriptive statistics

Descriptive statistics on the characteristics of the sample firms are
reported in Table 7.6. Panel A shows that the firms, all constituents of
12
See http://ec.europa.eu/internal_market/company/docs/official/1001041trans-play_en.
pdf.
Table 7.6. Financial, performance and ownership characteristics of the sample firms
Difference Difference
All Non-targets Targets in means in medians
N Mean Median Stdev N Mean Median N Mean Median
Panel A: Financial characteristics
Assets (€bn) 3543 61.01 4.04 220.10 3426 57.37 3.86 117 167.49 29.61 −110.12*** −25.76***
Market leverage 3543 20.5 17.8 15.7 3426 20.3 17.7 117 26.4 24.4 −6.1*** −6.7***
Book-to-market 3543 0.67 0.49 0.67 3426 0.66 0.48 117 0.86 0.63 −0.2*** −0.15***
Abnormal performance (%) 3543 6.79 0.60 43.30 3426 6.87 0.75 117 4.45 −3.78 2.42 4.53
Panel B: Ownership characteristics
Insiders 3543 4.60 0.12 11.70 3426 4.71 0.12 117 1.40 0.03 3.31*** 0.09***
Companies 3543 11.55 0 20.04 3426 11.58 0 117 10.52 0.64 1.06 −0.64*
State 3543 2.14 0 9.64 3426 1.94 0 117 8.02 0 −6.08*** 0***
Families 3543 0.02 0 0.17 3426 0.02 0 117 0.02 0 −0.001 0
Pressure-sensitive 3543 2.08 0.09 4.77 3426 2.06 0.09 117 2.85 0.53 −0.79* −0.44***
institutions
Pressure-insensitive 3543 32.16 29.39 20.90 3426 32.23 29.45 117 30.11 27.66 2.12 1.79**
institutions
Note: This table shows descriptive statistics on the financial, performance and ownership characteristics of the sample firms. Targets are
defined as those firms targeted with shareholder proposals. The variables are described in the Appendix. The difference in means t-test
assumes unequal variances when the test of equal variances is rejected at the 10% level. The significance of the difference in medians is based
on Wilcoxon ranksum tests. *, ** and *** denote significance at the 10%, 5% and 1% level, respectively.
Table 7.7 Country-level shareholder rights and corporate governance
Anti-
Sponsor Notice Record Share Bearer Pre- Proxy Electronic Show of self-
Governance index
block size period date blocking shares rights voting voting hands dealing
2005 2006 2007 2008 2009 2010
Austria 5 14 0 1 1 0 0 0 0 0.21 9.6 9.8 10.3 10.0 9.3 9.4
Belgium 20 24 5 1 1 0 1 1 0 0.54 7.8 7.8 7.8 7.4 8.0 8.0
Denmark 0 8 0 0 1 1 0 1 0 0.47 10.9 11.2 11.3 11.2 11.1 10.9
Finland 0 7 10 0 0 0 1 1 0 0.46 11.5 11.5 10.9 11.0 11.2 11.1
France 4 35 4 0 1 1 1 1 0 0.38 7.6 7.6 7.4 7.5 7.4 7.6
Germany 5 30 21 0 1 1 0 0 0 0.28 8.9 9.2 9.1 8.8 8.7 8.6
Greece 5 30 5 1 1 1 0 0 1 0.23 4.4 4.3 4.0 3.6 2.8 2.5
Ireland 0 21 0 0 1 1 0 0 1 0.79 9.3 9.5 9.6 9.7 8.9 8.7
Italy 2.5 30 3 1 1 1 1 0 1 0.39 3.7 3.6 3.3 3.4 3.1 3.1
Luxembourg 5 16 5 0 1 1 0 0 1 0.25 9.8 9.9 10.1 10.2 10.2 10.3
Netherlands 1 15 7 0 1 0 0 0 0 0.21 9.9 9.8 9.9 9.8 9.9 9.9
Norway 0 14 0 0 0 0 0 0 0 0.44 10.1 10.1 10.0 10.0 9.9 10.2
Portugal 5 30 5 1 1 0 1 1 1 0.49 6.9 5.9 5.9 6.3 6.3 5.7
Spain 5 30 10 1 1 1 1 1 1 0.17 6.6 5.3 5.2 5.3 5.2 5.3
Sweden 0 30 1 0 0 0 0 0 1 0.36 10.1 10.2 10.5 10.4 10.7 10.6
Switzerland 5 20 5 1 1 0 0 1 1 0.27 10.1 10.3 10.5 10.4 10.2 10.2
UK 5 21 3 0 1 0 1 1 1 0.93 8.4 9.1 8.8 8.5 7.9 8.3
Note: This table shows the country-level variables used in the analysis. The variables are described in the Appendix.
their home market indices, were very large, with total assets of €61.0
billion on average and €4.0 billion at the median. The mean (median)
market leverage, defined as the value of debt to the market value of assets,
was 20.5 (17.5) per cent in the sample. The mean (median) book-to-
market ratio was 0.67 (0.49), significantly higher than that in the US
sample of Renneboog and Szilagyi (2011), showing that European firms
are comparatively undervalued. This is somewhat surprising, because
the sample firms had actually outperformed their home market indices
in the year up to two months before their general meetings, by 6.79 per
cent on average and 0.60 at the median (both significant at the one per
cent level).
Ownership data for the sample firms are shown in Panel B of Table
7.6. More than 76 per cent of the firms reported shareholdings by
insiders, of 6.0 per cent on average but only 0.3 per cent at the median.
Holdings by affiliate companies, families and the government were
reported by 41, 39 and 8 per cent, respectively, of the sample firms,
with average stakes of 27.8, 28.3 and 0.4 per cent, respectively. Pressure-
sensitive institutional investors – which Brickley et al. (1988) call banks
and insurance firms due to their existing or potential business ties with
investee firms – held an average 2.7 per cent in 77 per cent of the sample
firms. Pressure-insensitive institutions – pension funds, investment
funds and investment advisors – held 32.2 per cent, significantly less
than the 49.2 per cent reported for the US by Renneboog and Szilagyi
(2011).
The country-level variables are summarised in Table 7.7 and show an
interesting picture. With the exception of Belgium, the sample countries
had met the Directive’s five per cent ownership requirement for share-
holder proposal submissions even before the Directive was transposed.
Interestingly, ownership restrictions have not existed at all in the Nordic
countries – and Ireland –, which possibly explains why activist inter-
ventions have been more prevalent in these countries. As has been
mentioned, proposals may be submitted by any shareholder with
USD2,000 worth of voting shares in the US.
The other variables show significant variation across countries. The
notice period required to be given before general meetings was 21 days
on average – the maximum prescribed by the Directive – but it ranged
from seven days in Finland to 35 days in France. The record date for the
register of shareholders was an average five days before meetings; none
of the sample countries exceeded the Directive’s maximum of 30 days,
but there was no record date requirement in four countries. Bearer
shares, not regulated by the Directive, are permitted in all countries

except Finland, Norway and Sweden. In terms of shareholder participa-
tion, the drawback of bearer shares is that the ultimate owners are very
difficult to identify. Share blocking existed in Southern European coun-
tries, as well as Austria, Belgium and Switzerland. Shareholders could be
prohibited from requesting information from management prior to
meetings in nine of the 17 countries. The voting process itself was also
liberalised to varying degrees. Voting by proxy and by electronic means
was fully permitted in seven and eight countries, respectively. Voting by
show of hands, also not regulated by the Directive, remains allowed in
nine countries including the UK – although in controversial cases, the
voting rights held by each shareholder can be counted.
Djankov et al. (2008) report that the anti-self-dealing index for the
sample countries ranges from 0.93 and 0.79 in the UK and Ireland, to
0.21 in Austria and the Netherlands and 0.17 in Spain. The authors
confirm the general observation that English common law countries
provide much better protection to minority shareholders. The govern-
ance index constructed from World Bank data is the highest for Finland,
Denmark and Sweden, and the lowest for Greece, Italy, Spain and
Portugal. There is some deterioration in the index over time for the
countries most affected by the European funding crisis, including
Ireland.
4.2. What determines the voting outcomes of management

proposals?
The multivariate pooled panel regressions explaining the voting success
of management proposals are shown in Table 7.8. As the dependent
variable – the percentage of votes in favour – is between 0 and 1, the
logistical transformation ln[votes for/(100-votes for)] is applied to create
a continuous variable with both negative and positive values.
The model statistics in Table 7.8 show that the success of management
proposals is predominantly determined by the proposal characteristics.
The recommendation of management is the single biggest driver of
proposal success. Importantly, however, we confirm that shareholder
dissent increases when management is simultaneously challenged with a
shareholder proposal or has been defeated at a previous meeting. Once
again we find that operational proposals are the most successful, while
voting dissent is the strongest over the adoption of anti-takeover devices
and executive compensation. The results now show that voting support
Table 7.8 Regressions explaining the votes for management proposals
Model 1 t-test Model 2 t-test Model 3 t-test Model 4 t-test Model 5 t-test
Coeff t-test Coeff t-test Coeff t-test Coeff t-test Coeff t-test
Meeting and proposal characteristics
Extraordinary meeting 0.286 1.97** 0.162 1.11 0.191 1.32 0.007 0.05
Shareholder proposal −0.566 −2.70*** −0.239 −1.07 −0.310 −1.51 −0.432 −2.36**
Management defeated before −0.851 −5.43*** −0.709 −4.57*** −0.687 −4.80*** −0.377 −2.78***
Recommendation – none −1.566 −2.82*** −1.750 −3.15*** −1.746 −3.34*** −2.452 −4.96***
Recommendation – against −8.159 −16.56*** −8.162 −16.01*** −7.937 −13.00*** −6.988 −9.73***
Proposal objectives
Operational issues 1.621 8.03*** 1.453 7.19*** 1.455 7.39*** 1.303 6.81***
Elect directors 0.271 1.32 0.126 0.62 0.177 0.89 0.067 0.35
Discharge directors 0.926 3.14*** 0.959 3.43*** 0.942 3.54*** 0.495 1.91*
Board governance 1.385 6.55*** 1.252 5.88*** 1.139 5.47*** 0.692 3.44***
Adopt anti-takeover device −2.898 −10.01*** −2.943 −9.52*** −3.060 −10.46*** −2.675 −10.23***
Repeal anti-takeover device 0.158 0.20 0.064 0.08 −0.071 −0.10 0.211 0.32
Voting and disclosure 0.231 1.03 −0.031 −0.14 0.120 0.55 0.004 0.02
Compensation −1.134 −5.42*** −1.329 −6.34*** −1.279 −6.26*** −1.386 −7.07***
Capital 0.491 2.40** 0.319 1.57 0.360 1.80* 0.322 1.66*
Restructuring 0.619 2.38** 0.615 2.28** 0.603 2.32** 0.740 2.99***
Social −0.703 −2.93*** −0.833 −3.55*** −0.641 −2.80*** −0.829 −3.79***
Table 7.8 (cont.)
Model 1 t-test Model 2 t-test Model 3 t-test Model 4 t-test Model 5 t-test
Financial characteristics
Log of assets −0.243 −11.01*** −0.244 −11.18*** −0.237 −11.73*** −0.213 −10.65***
Market leverage 0.004 1.24 0.003 1.18 0.000 0.00 0.000 0.10
Book-to-market 0.043 0.62 0.015 0.22 0.017 0.31 0.040 0.75
Abnormal performance 0.001 1.06 0.001 1.59 0.001 1.15 0.001 1.42
Ownership characteristics
Insiders 0.013 4.33*** 0.014 5.26*** 0.014 5.28***
Companies 0.014 5.84*** 0.014 6.11*** 0.014 6.23***
State 0.023 4.97*** 0.020 5.28*** 0.020 5.18***
Families −0.174 −0.67 −0.247 −1.39 −0.256 −1.55
Pressure-sensitive institutions 0.027 4.10*** 0.015 2.41** 0.014 2.38**
Pressure-insensitive institutions −0.004 −1.99** −0.005 −2.59*** −0.005 −2.47**
Shareholder rights and corporate governance
Notice period 0.003 0.13 0.015 0.64
Record date 0.040 4.69*** 0.031 3.86***
Share blocking 1.034 4.42*** 0.874 4.05***
Bearer shares −1.498 −6.22*** −1.403 −6.11***
Pre-rights −0.375 −1.43 −0.522 −2.09**
Proxy voting 1.034 2.50** 1.016 2.49**
Electronic voting −1.407 −3.72*** −1.412 −3.86***
Show of hands 0.409 1.61 0.423 1.76*
Anti-self-dealing −0.026 −0.05 0.096 0.21
Governance index 0.159 1.58 0.166 1.67*
2006 −0.158 −1.72* −0.113 −1.34 −0.141 −1.68* −0.163 −1.86* −0.121 −1.35
2007 −0.078 −0.74 −0.043 −0.44 −0.108 −1.07 −0.079 −0.81 −0.041 −0.43
2008 0.018 0.19 −0.013 −0.15 −0.147 −1.67* −0.112 −1.26 −0.044 −0.51
2009 −0.281 −2.94*** −0.317 −3.30*** −0.513 −5.20*** −0.485 −4.89*** −0.351 −3.70***
2010 −0.134 −1.36 −0.191 −2.01** −0.392 −4.08*** −0.506 −5.38*** −0.355 −3.88***
Industry dummies Yes Yes Yes Yes Yes
Constant 4.219 7.55*** 9.227 12.78*** 9.252 13.37*** 9.649 7.32*** 8.315 6.38***
No. of obs 38,313 38,313 38,313 38,313 38,313
No. of firms 845 845 845 845 845
F-test 101.23*** 97.87*** 86.71*** 25.95*** 82.35***
R2 0.132 0.164 0.189 0.131 0.232
Note: The table reports pooled panel regressions. The dependent variable is defined as ln(votes for)/(100-votes for), where the percentage votes for are
calculated from the three-way voting outcome. The variables are described in the Appendix. Log of assets is the natural logarithm of the book value of
assets. T-statistics use robust standard errors with White (1980) correction for heteroskedasticity and adjusted for clustering of observations on each
firm. *, ** and *** denote significance at the 10, 5 and 1% level, respectively.
is limited for social proposals, mostly related to charitable donations and

political expenditures. The year dummies in the regressions confirm that
shareholder dissent increased somewhat after 2008.
Surprisingly, we find no evidence that the voting outcomes on man-
agement proposals are affected by poor firm performance in the form of
a high book-to-market ratio or underperformance relative to the home
market index. However, they are strongly determined by the size of the
firm and the composition of the voting shareholders. Management
proposals are generally supported by insiders, affiliate firms, govern-
ments, and pressure-sensitive institutional investors. However, they are
significantly less successful in large, widely held firms with diverse
shareholder bases, as well as in firms held by pressure-insensitive institu-
tional owners. This latter result is particularly important. It confirms that
investment funds and other pressure-insensitive institutions are pre-
pared to use their vote to publicly challenge management. It also prom-
ises that as institutional ownership increases further in Europe, minority
shareholders will become increasingly discerning at general meetings.
Most importantly for European regulators, Table 7.8 confirms that
country-level regulation plays a very significant role in galvanising
shareholders. The voting success of management proposals is signifi-
cantly lower when shareholders can freely trade their shares and exercise
their voting rights, including when (i) record date restrictions are
reduced, (ii) bearer shares permit at least some level of anonymity, (iii)
there is no share blocking, (iv) electronic voting is permitted, and (v)
vote counts cannot be distorted by a show of hands. Interestingly, there is
evidence that proxy voting increases rather than decreases support for
management proposals – presumably due to the ultimate beneficial
owners not giving specific voting instructions. Finally, we find some
indication that the general governance environment matters, with man-
agement proposals seeing less dissent in countries with a high govern-
ance index. On the whole, these results critically demonstrate that the
Directive’s provisions are headed in the right direction in terms of
enabling shareholder voice.
4.3. Why are firms targeted by shareholder proposals?

To examine why activist shareholders resort to submitting their own
proposals against European firms, we now analyse the probability that a
shareholder proposal contested by management is tabled. Table 7.6 has
already provided univariate statistics on the financial and ownership
characteristics of target versus non-target firms. These statistics show

that target firms tend to be much larger and more levered than non-
targets, with a mean (median) asset value of €167.5 billion (€29.6 billion)
and market leverage of 26.4 (24.4) per cent. Importantly, activists also
tend to target firms that underperform, with a mean (median) book-to-
market ratio of 0.86 (0.63). For the US, Renneboog and Szilagyi (2011)
demonstrate that target firms also have generally poor governance
structures, including anti-takeover devices, ineffective boards and ill-
incentivised CEOs. While we cannot replicate their analysis due to a lack
of data, these findings uniformly imply that shareholder proposal sub-
missions, in both Europe and the US, are motivated by the ‘correct’
incentive of disciplining management rather than self-serving interests.
Table 7.6 has also shown some evidence that the activists submitting
shareholder proposals first examine the target firm’s shareholder base to
see the level of voting support they can potentially attract. Firms sig-
nificantly owned by insiders and other companies are less likely to be
targeted. This is unsurprising, because investors with major control
benefits in the firm rarely have the incentive to support a hostile activist.
Proposals are more likely to be submitted against state-owned firms, but
we have found that the proposal sponsors are often the governments
themselves. The univariate results surprisingly show that target firms
tend to have more of their equity held by pressure-sensitive and less by
pressure-insensitive institutional investors.
The multivariate pooled probit models explaining the probability of
management-contested shareholder submissions are shown in Table
7.9. We find that firms are significantly more likely to be targeted if
they have been targeted in a previous year, or management has pre-
viously been defeated in a shareholder vote. Importantly, the results
confirm our univariate findings that target firms tend to be large and
poorly performing. Controlling for firm size, we find no evidence that
targets tend to be more levered. This is expected, because large firms
have more debt capacity and therefore tend to employ somewhat more
leverage.
The multivariate regressions fail to confirm that, all else being equal,
shareholder proposals have become more prevalent in Europe over time.
This implies that the greater frequency of proposal submissions in the
latter part of the sample period is driven by other time-varying factors
such as poor firm performance. The regressions show little evidence for
the relevance of shareholder composition in the target selection process.
There is some indication that state-owned firms are more likely to be
Table 7.9 Determinants of shareholder proposal submissions
Model 1 Model 2 Model 3 Model 4

Coeff Z-test Coeff Z-test Coeff Z-test Coeff Z-test
Meeting characteristics
Previously targeted 1.257 6.92*** 1.251 6.80*** 1.227 6.67*** 0.956 5.05***
Management defeated before 0.351 2.35** 0.310 2.04** 0.540 3.42*** 0.477 2.89***
Log of assets 0.165 5.98*** 0.158 5.42*** 0.220 5.49***
Market leverage 0.003 1.07 0.005 1.48 0.005 1.41
Book-to-market 0.135 2.20** 0.136 2.20** 0.146 2.30**
Abnormal performance 0.001 0.43 0.001 0.53 0.001 0.65
Insiders −0.003 −0.44 −0.001 −0.12
Companies 0.000 0.07 0.003 0.96
State 0.012 3.03*** 0.006 1.53
Families −0.018 −0.12 0.114 0.62
Pressure-sensitive institutions 0.012 1.56 −0.001 −0.09
Pressure-insensitive institutions 0.003 0.81 0.002 0.56
Notice period 0.039 2.64*** 0.034 2.05**
Record date 0.010 0.79 0.009 0.58
Share blocking 1.116 2.65*** 1.381 2.87***
Bearer shares −0.911 −4.25*** −0.857 −3.53***
Pre-rights 0.384 1.88* 0.529 2.19**
Proxy voting 0.094 0.27 0.301 0.79
Electronic voting 0.039 0.15 −0.118 −0.41
Show of hands −0.624 −2.41** −0.625 −2.17**
Sponsor block size −0.085 −2.39** −0.099 −2.28**
Anti-self-dealing 1.571 2.18** 2.218 2.65***
Governance index 0.262 2.96*** 0.353 3.53***
2006 0.621 1.60 0.612 1.55 0.554 1.44 0.538 1.34
2007 0.880 2.41** 0.851 2.28** 0.833 2.30** 0.843 2.21**
2008 0.950 2.72*** 0.903 2.52** 0.660 1.88* 0.738 1.98**
2009 0.634 1.79* 0.580 1.61 0.516 1.43 0.482 1.26
2010 0.677 1.91* 0.636 1.74* 0.464 1.29 0.495 1.29
Industry dummies Yes Yes Yes Yes
Constant −10.384 −10.323 −10.000 −16.327
No. of obs 3,450 3,450 3,450 3,450
No. of firms 866 866 866 866
Wald χ2 233.11 247.20 277.40 330.54
Pseudo R2 0.228 0.242 0.272 0.324
Log pseudolikelihood −394.36 −387.32 −372.22 −345.65
Note: The table reports pooled probit models, where the dependent variable is a dummy equal to 1 if a shareholder proposal is submitted
and 0 otherwise. The variables are described in the Appendix. Log of assets is the natural logarithm of the book value of assets. Z-statistics use
robust standard errors with White (1980) correction for heteroskedasticity and adjusted for clustering of observations on each firm. *, ** and ***
denote significance at the 10, 5 and 1% level, respectively.
targeted, but this is not robust to the inclusion of the country-level

variables in Model 4.
Table 7.9 provides conclusive evidence that like the success of man-
agement proposals, the probability of activist interventions is heavily
affected by country-level regulation. Proposal submissions become more
frequent (i) when entry costs are reduced through lower minimum
ownership requirements; (ii) when shareholders have better access to
information, including longer notice periods and the ability to request
information from management; and (iii) when activists have a better
chance of identifying and communicating with the ultimate sharehold-
ers, because bearer shares are not permitted. To some extent, this latter
assertion is also supported by the positive relation between the proba-
bility of proposal submissions and share blocking. Share blocking is a
major impediment to shareholder participation at general meetings
because it prevents investors from trading their shares. However, the
firm’s shareholders should reveal themselves in the process, enabling
activists to communicate with them.
Finally, we find strong indication that shareholder activism at general
meetings is largely a function of the corporate governance environment.
Table 7.9 shows that the probability of proposal submissions increases
substantially in both the anti-self-dealing index and the general World
Bank governance index. This implies that with the protection and
empowerment of minority shareholders, which fundamentally encour-
ages equity investment itself, comes the more active involvement of these
shareholders in the corporate governance process.
4.4. What determines the voting outcomes of shareholder

proposals?
The final Table, Table 7.10, shows the multivariate pooled panel regres-
sions explaining the voting success of shareholder proposals opposed by
management. As before, we apply the logistical transformation ln[votes
for/(100-votes for)] to the voting outcomes. The regressions contain
only 217 observations, as the full set of explanatory variables is only
available for 117 target firms.
The results show that the voting success of shareholder proposals is
also predominantly determined by the proposal characteristics. We find
that all else being equal, the voting shareholders attribute by far the
greatest benefits to takeover-related proposals. This is very much in line
with the findings of Renneboog and Szilagyi (2011) for the US, and
Table 7.10 Regressions explaining the votes for shareholder proposals
Model 1 Model 2 Model 3 Model 4 Model 5

Meeting and proposal characteristics
Extraordinary meeting 0.457 0.97 0.157 0.34 0.103 0.30 −0.119 −0.30
Proposal passed before 2.519 5.16*** 2.120 3.64*** 1.405 2.23** 0.808 0.89
Proposal objectives
Elect directors 0.714 1.47 0.517 0.97 0.721 1.45 0.611 1.12
Remove directors 1.362 1.47 1.276 1.25 1.530 1.42 0.616 1.11
Board governance 0.024 0.05 0.013 0.02 0.126 0.25 −0.188 −0.30
Repeal anti-takeover device 1.968 2.55** 1.986 2.93*** 2.058 3.51*** 1.827 3.04***
Voting and disclosure 0.050 0.12 0.144 0.37 0.132 0.36 −0.067 −0.17
Compensation 0.370 0.68 0.333 0.64 0.386 0.74 −0.003 0.00
Capital −0.318 −0.35 −1.394 −1.25 −1.171 −1.06 −2.761 −1.58
Restructuring 0.313 0.81 0.402 0.89 0.406 0.98 −0.049 −0.12
Dividends −0.640 −1.38 −0.732 −1.60 0.243 0.45 −0.414 −0.54
Proposal sponsors
Pension funds 1.917 2.26** 2.075 2.43** 2.717 2.80*** 0.722 0.42
Investment funds 2.774 5.30*** 2.233 2.82*** 2.247 2.98*** 1.510 3.03***
Banks 3.978 2.48** 3.116 1.41 −1.525 −0.48 −7.571 −1.75*
Companies 0.869 0.84 0.448 0.34 1.224 0.83 1.470 1.42
Employees 1.646 2.38** 1.800 2.55** 1.968 3.26*** 1.753 3.43***
Dissidents 2.102 1.75* 1.138 0.68 3.424 3.34*** 2.825 1.63
Shareholder associations 1.634 1.40 1.160 0.90 1.549 2.13** −0.098 −0.09
State 2.998 3.05*** 2.779 2.57** 2.433 2.39** 2.811 3.56***
Table 7.10 (cont.)
Model 1 Model 2 Model 3 Model 4 Model 5

Log of assets −0.306 −1.70* −0.386 −2.41** −0.601 −3.81*** −0.385 −2.49**
Market leverage −0.001 −0.07 −0.009 −0.48 0.032 1.94* 0.001 0.03
Book-to-market −0.169 −0.49 −0.704 −1.61 −0.697 −2.03** −0.637 −1.82*
Abnormal performance 0.001 0.16 −0.002 −0.61 −0.002 −0.56 −0.001 −0.15
Insiders −0.045 −1.96** −0.079 −3.49*** −0.060 −2.79***
Companies −0.041 −4.53*** −0.022 −2.08** −0.026 −2.54**
State −0.019 −1.87* −0.010 −1.09 −0.012 −1.30
Families −9.413 −2.51** −6.573 −1.24 −5.441 −1.00
Pressure-sensitive institutions 0.104 1.83* 0.012 0.17 0.103 1.69*
Pressure-insensitive institutions −0.014 −2.83*** 0.001 0.13 0.000 −0.05
Notice period −0.168 −1.28 −0.227 −1.77*
Record date 0.016 0.54 −0.021 −0.41
Share blocking −5.619 −2.89*** −4.990 −2.10**
Bearer shares 0.270 0.20 0.128 0.10
Pre-rights −1.308 −1.13 −0.259 −0.21
Proxy voting 3.184 1.67* 3.663 1.92*
Electronic voting −0.783 −0.46 −2.624 −1.51
Show of hands 4.160 1.98** 4.666 2.04**
Sponsor block size 0.396 2.17** 0.483 1.88*
Anti-self-dealing −11.002 −2.48** −11.844 −2.36**
Governance index −0.751 −1.17 −0.968 −1.67*
2006 0.177 0.16 0.427 0.46 1.237 1.21 −0.960 −0.75 0.636 0.50
2007 0.572 0.76 1.177 1.65* 1.740 2.17** 0.796 1.21 1.496 1.90*
2008 1.100 1.50 1.420 2.20** 2.133 3.13*** 0.699 0.98 1.481 2.02**
2009 1.148 1.47 1.661 2.27** 3.068 3.69*** 1.837 2.03** 2.352 2.44**
2010 1.564 2.28** 1.982 2.80*** 3.347 4.26*** 2.077 2.77* 2.383 2.63**
Industry dummies Yes Yes Yes Yes Yes
Constant −5.584 −5.11*** 3.898 0.76 4.060 1.09 23.761 2.56** 21.576 2.00**
No. of obs 217 217 217 217 217
No. of firms 58 58 58 58 58
F-test 9.26*** 8.26*** 8.95*** 12.10*** 11.48***
R2 0.593 0.607 0.670 0.678 0.780
Note: The table reports pooled panel regressions. The dependent variable is defined as ln(votes for)/(100-votes for), where the percentage votes
for are calculated from the three-way voting outcome. The variables are described in the Appendix. Log of assets is the natural logarithm of the
book value of assets. T-statistics use robust standard errors with White (1980) correction for heteroskedasticity and adjusted for clustering of
observations on each firm. *, ** and *** denote significance at the 10, 5 and 1% level, respectively.
confirms that minority shareholders are keen to expose management to

takeover threat and reap potentially significant takeover premia. We also
confirm that submissions made by investment funds attract particularly
strong voting support, along with those made by the government and
employees. There is no robust evidence that proposals achieve more
support when another proposal has previously passed. However, the
votes cast in favour of shareholder proposals clearly increase over
time, implying that shareholder dissent at European general meetings
is on the rise.
The model statistics show that the firm characteristics have relatively
limited explanatory power, and their impact on the voting outcomes is
quite sensitive to alternative specifications. As with management pro-
posals, shareholder submissions are less successful when made against
large, widely held firms where voting coalitions are more difficult to
build. Surprisingly, however, there is evidence that shareholder pro-
posals also attract less, rather than more, voting support when the target
firm is underperforming. It is similarly interesting that proposal success
increases somewhat in ownership by pressure-sensitive but not pressure-
insensitive institutions. As expected, insider and company owners tend
not to support shareholder proposals.
Table 7.10 finally confirms that country-level regulation has a major
impact on the proposal outcomes. It is expected that voting support
increases in the stringency of minimum ownership requirements: pro-
posal sponsors are certain to support their own proposals, and more
powerful sponsors should better be able to build voting coalitions. We
confirm that the voting shareholders are more likely to support submis-
sions when their shares are not blocked and can therefore be freely
traded. It is interesting, however, that proposal success declines in the
notice period and increases when proxy voting is allowed. As with
management proposals, vote count distortions due to voting by a show
of hands increases the voting support recorded.
The most important result is that while good governance increases
the probability of proposal submissions, it actually reduces the voting
support they achieve. Table 7.10 shows that the success of shareholder
proposals is related negatively to both the protection of minority
shareholders and the general governance environment. This latter
finding is fully consistent with the negative relationship between gov-
ernance quality and voting dissent in Table 7.8. Fundamentally, these
results show that while allowing shareholders to raise their voice at
general meetings is a part of good governance, they will not feel the
need to do so, or indeed support any such initiatives, unless they deem
it necessary.
5. Conclusion and policy implications

There is now considerable evidence in the US academic literature that
shareholders participating at company general meetings are valuable moni-
toring agents. In Europe, the empirical investigation of this issue has been
complicated by data availability, as well as the fact that European countries
are very diverse in terms of ownership structures, legal provisions governing
shareholder rights, as well as the monitoring incentives of and costs borne
by shareholders. Shareholder absenteeism remains frequent in Continental
Europe, in particular, due to ‘rational apathy’, and voting dissent at general
meetings has increased only marginally in the last decade.
Whether shareholder participation in corporate governance should be
facilitated has been subject to heated policy debate around the world. With
the onset of the global financial crisis a clear pro-shareholder tendency
emerged, and corporate governance codes have been updated accordingly.
Nonetheless, regulators continue to drag their feet about truly enabling
shareholder voice. The European analysis presented in this chapter has
confirmed earlier US evidence that shareholder engagement at general
meetings is actually a part of good governance. Shareholders tend not to
have self-serving agendas and are discerning enough to intervene only at the
‘correct’ firms and when deemed necessary. In fact, there is evidence that
they use their voice not simply to discipline underperforming managers, but
also to make up for inefficiencies in the broader governance and institu-
tional environment that potentially lead to managerial agency problems and
underperformance in the first place.
Ultimately, our results indicate that national regulators in the EU
should go beyond the minimum standards introduced by the
Shareholder Rights Directive to support shareholder participation in
corporate governance. The Directive’s provisions still fail to ensure a
level playing field for all shareholders. The procedural and information
costs of cross-border voting remain largely prohibitive and must be
further reduced. Communication between atomistic minority share-
holders should be enabled, including by promoting registered rather
than bearer shares while easing registration rules, and by reducing and
harmonising ownership disclosure thresholds, perhaps at the 3 per cent
already in place in the UK, or 2 per cent in place in Italy. Shareholders
should also have access to company proxies and face less stringent
minimum ownership requirements to table their own proposals. Some

proposals may even be put routinely to shareholder vote to reduce the
need for shareholder intervention, as has been the case for some form of
say-on-pay not only in the US and the UK, but in Belgium, Denmark,
Italy, the Netherlands and Sweden, among others. Of course, these are
only some of the issues that regulators must consider, and a transition
from soft to hard law may be advisable.
Our bottom-line conclusion is that the rules governing shareholder
engagement at European general meetings should be further relaxed and
harmonised. Minority shareholders are useful monitoring agents, and
we have found that criticism that they might abuse their rights is
exaggerated. It is critical to point out that beyond helping to address
the managerial agency concerns highlighted by the global financial crisis,
the harmonisation of shareholder voice would also aid the European
Commission’s declared objective of deepening equity market integration
within the EU. The fundamental purpose of integration is to create liquid
markets that bring down financing costs for European firms. However,
shares will always trade at a discount if investors cannot freely exercise
the voting rights attached to them. Market liquidity will also continue to
be hindered by the persistence of concentrated ownership structures,
which have historically remained in place in countries where share-
holders have been hesitant to diversify due to restrictions on the rights
of minority shareholders.
APPENDIX
VARIABLE DESCRIPTIONS
Variable name Description and source

Panel A: Meeting and proposal characteristics
Extraordinary meeting A dummy variable equal to 1 if the proposal is presented
at an extraordinary meeting, and 0 if it is presented at
an annual meeting.
Shareholder proposal A dummy variable equal to 1 if a shareholder proposal is
presented at the meeting, and 0 otherwise.
Management defeated A dummy variable equal to 1 if a management proposal
before has previously failed or a management-contested
shareholder proposal has previously passed, and 0
otherwise.

Recommendation – A dummy variable equal to 1 if management has made
none no voting recommendation on the proposal, and 0
otherwise.
Recommendation – A dummy variable equal to 1 if management has
against recommended a vote against the proposal, and 0
otherwise.
Previously targeted A dummy variable equal to 1 if the firm has previously been
targeted by a shareholder proposal, and 0 otherwise.
Proposal passed before A dummy variable equal to 1 if a shareholder proposal
submitted to the firm has previously passed the
shareholder vote.
Panel A: Financial and ownership characteristics
Assets (€ millions) The book value of total assets. Source: Thomson ONE
Banker.
Market leverage Total debt divided by the book of liabilities plus the
market value of equity. Source: Thomson ONE
Banker.
Book-to-market ratio The book value of equity divided by the market value of
equity. Source: Thomson ONE Banker.
Abnormal The dividend-adjusted stock price return minus the
performance (%) return on the home market index, in the year up to two
months before the meeting date. Source: Datastream.
Ownership (%, by type The number of shares held by each type of owner divided
of owner) by the total number of shares outstanding. Pressure-
sensitive institutional investors are banks and
insurance companies. Pressure-insensitive
institutional investors are pension and labour union
funds, investment funds and their managers, and
independent investment advisers. Source: CapitalIQ.
Panel B: Shareholder rights and corporate governance (country level)
Notice period (days) The number of days that must pass between the day of
the (last) publication of a convocation to a general
meeting and the day of the meeting. Source: European
Commission (2006) and Georgeson (2008).
Record date (days) The minimum number of days between the day the
register of shareholders is closed before a general
meeting and the day of the meeting. Source: European

Share blocking A dummy variable equal to one if shareholders must
deposit their shares for a general meeting. Source:
European Commission (2006) and Georgeson (2008).
Bearer shares A dummy variable equal to one if companies are
permitted to issue bearer shares. Source: European
Pre-rights A dummy variable equal to one if shareholders have the
right to ask questions before a general meeting. Source:
Proxy voting A dummy variable equal to one if shareholders may be
fully permitted to vote by proxy. Source: European
Electronic voting A dummy variable equal to one if shareholders may be
permitted to vote electronically. Source: European
Show of hands A dummy variable equal to one if shareholders have to
right to vote on show of hands. Source: European
Sponsor block size (%) The percentage shareholding required to place items on
the agenda and table shareholder proposals. Source:
Anti-self-dealing index A measure of legal protection of minority shareholders
against expropriation by corporate insiders. Ranges
from 0 to 1. Source: Djankov et al. (2008).
Governance index The sum of the World Bank’s six Worldwide
Governance Indicators (voice and accountability,
political stability and absence of violence, government
effectiveness, regulatory quality, rule of law, and
control of corruption). Each indicator ranges between
–2.5 and 2.5. The methodology is presented in
Kaufmann et al. (2010). Source: http://info.worldbank.
org/governance/wgi/index.asp.
References
Armour, J. 2008. ‘Enforcement Strategies in UK Corporate Governance: A
Roadmap and Empirical Assessment’, ECGI Law Working Paper 106/2008.
Bainbridge, S. M. 2006. ‘Director primacy and shareholder disempowerment’,
Harvard Law Review 119: 1735–58.
Barca, F. and M. Becht (eds.) 2001. The Control of Corporate Europe. Oxford
University Press.
Bebchuk, L. A. 2005. ‘The case for increasing shareholder power’, Harvard Law
Review 118: 835–914.
Becht, M. 2001. ‘Beneficial ownership in the United States’, in Barca and Becht
(eds.): pp. 285–99.
Becht, M. and C. Mayer 2001. ‘Introduction’, in Barca and Becht (eds.): pp. 1–45.
Becht, M., J. Franks, C. Mayer and S. Rossi. 2009. ‘Returns to shareholder activism:
Evidence from a clinical study of the Hermes U.K. Focus Fund’, Review of
Belcredi, M., S. Bozzi and C. Di Noia 2012. ‘Board elections and shareholder
activism: the Italian experiment’, Chapter 8, this volume.
Bradley, M., A. Brav, I. Goldstein and W. Jiang 2010. ‘Activist arbitrage: A study of
open-ending attempts of closed-end funds’, Journal of Financial Economics
95: 1–19.
Brav, A., W. Jiang, F. Partnoy and R. Thomas 2008. ‘Hedge fund activism, corporate
governance, and firm performance’, Journal of Finance 63: 1729–75.
Brickley, J., R. Lease and C. Smith 1988. ‘Ownership structure and voting on anti-
takeover amendments’, Journal of Financial Economics 20: 267–91.
Buchanan, B., J. Netter, A. Poulsen and T. Yang 2012. ‘Shareholder proposal rules
and practice: Evidence from a comparison of the US and the UK’, American
Business Law Journal 49.
Clifford, C.P. 2008. ‘Value creation or destruction? Hedge funds as shareholder
activists’, Journal of Corporate Finance 14: 323–36.
Crespi, R. and L. Renneboog 2010. ‘Is (institutional) shareholder activism new?
Evidence from UK shareholder coalitions in the pre-Cadbury era’,
Corporate Governance International Review 18: 274–95.
Cziraki, P., L. Renneboog and P. Szilagyi 2011. ‘Shareholder activism through
proxy proposals: the European perspective’, European Financial
Davies, P., G. Ferrarini, K. Hopt, A. Pietrancosta, R. Skog, S. Soltysinski, J. Winter
and E. Wymeersch 2011. ‘Response to the European Commission’s green
paper “The EU Corporate Governance Framework” – European Company
Law Experts’, Working paper, available at www.ssrn.com.
De Jong, A., G. Mertens and P. Roosenboom 2006. ‘Shareholders’ voting at general
meetings: Evidence from the Netherlands’, Journal of Management and
Governance 10: 353–80.
Del Guercio, D., L. Seery and T. Woidtke 2008. ‘Do boards pay attention when
institutional investor activists “just vote no”?’, Journal of Financial
Djankov, S., R. La Porta, F. Lopez-de-Silanes and A. Shleifer 2008. ‘The law and
economics of self dealing’, Journal of Financial Economics 88: 430–65.
Ertimur, Y., F. Ferri and S. Stubben 2010. ‘Board of directors’ responsiveness to
shareholders: Evidence from shareholder proposals’, Journal of Corporate
Finance 16: 53–72.
European Commission, 2004. ‘Fostering an appropriate regime for shareholders’
rights’, Consultation Document of the Services of the Internal Market
Directorate General.
2006. ‘Annex to the proposal for a Directive of the European Parliament and of
the Council on the exercise of voting rights by shareholders of companies
having their registered office in a Member State and whose shares are
admitted to trading on a regulated market and amending Directive 2004/
109/EC – Impact assessment’, Commission staff working document.
2011. ‘The EU Corporate Governance Framework’, Green Paper.
Faccio, M. and L. Lang 2002. ‘The ultimate ownership of Western European
Fama, E. and K. French 2001. ‘Disappearing dividends: Changing firm character-
istics or lower propensity to pay?’, Journal of Financial Economics 60: 3–43.
FESE 2008. ‘Share Ownership Structure in Europe’, Federation of European
Securities Exchanges (FESE) Economics and Statistics Committee.
Franks, J. and C. Mayer 2001. ‘Ownership and control of German corporations’,
Georgeson 2008. ‘Shareholders meetings in Europe’, manual.
Gillan, S. and L. Starks 2007. ‘The evolution of shareholder activism in the United
States’, Journal of Applied Corporate Finance 19: 55–73.
Girard, C. 2009. ‘Comparative study of successful French and Anglo-Saxon share-
holder activism’, Working Paper.
Goergen, M. and L. Renneboog 2001. ‘Strong managers and passive institutional
investors in the UK’, in Barca and Becht (eds.): pp. 259–84.
Greenwood, R. and M. Schor 2009. ‘Investor activism and takeovers’, Journal of
Harris, M. and A. Raviv 2010. ‘Control of corporate decisions: shareholders vs.
management’, Review of Financial Studies 23: 4115–47.
Hewitt, P. 2011. ‘The Exercise of Shareholder Rights: Country Comparison of
Turnout and Dissent’, OECD Corporate Governance Working Paper 3.
Kaufmann, D., A. Kraay, and M. Mastruzzi, M. 2010. ‘The Worldwide
Governance Indicators: Methodology and Analytical Issues’, World Bank
Research Working Paper 5430.
Klein, A. and E. Zur 2009. ‘Entrepreneurial shareholder activism: Hedge funds
and other private investors’, Journal of Finance 64: 187–229.
and R. Vishny 1998. ‘Law and finance’, Journal of Political Economy 106: 1113–55.
Lipton, M. 2002. ‘Pills, polls, and professors redux’, University of Chicago Law
Review 69: 1037–65.
Manifest 2008. ‘Proxy voting 2007 – A pan-European perspective’, Manifest
Information Services.
Martin, K. and R. Thomas 1999. ‘The effect of shareholder proposals on executive
compensation’, University of Cincinnati Law Review 67: 1021–81.
Martynova, M. and L. Renneboog 2008. ‘Spillover of corporate governance standards in
cross-border mergers and acquisitions’, Journal of Corporate Finance 14: 200–23.
2011a. ‘The performance of the European market for corporate control: Evidence
from the 5th takeover wave’, European Financial Management 17: 208–60.
2011b. ‘Evidence on the international evolution and convergence of corporate
governance regulations’, Journal of Corporate Finance 17: 1531–57.
Masouros, P. 2010. ‘Is the EU taking shareholder rights seriously? An essay on the
impotence of shareholdership in corporate Europe’, European Company
Law 7: 195–203.
McCahery, J., Z. Sautner and L. Starks 2010. ‘Behind the Scenes: The Corporate
Governance Preferences of Institutional Investors’, Tilburg Law School
Research Paper 010/2010.
Nibler, M. 1998. ‘Bank Control and Corporate Performance in Germany: The
Evidence’, doctoral dissertation, University of Cambridge.
OECD 2007. Lack of Proportionality between Ownership and Control: Overview and
Issues for Discussion, OECD Steering Group on Corporate Governance.
2011. The Role of Institutional Investors in Promoting Good Corporate
Governance, Corporate Governance, OECD Publishing.
Renneboog, L. and P. Szilagyi 2011. ‘The role of shareholder proposals in corpo-
rate governance’, Journal of Corporate Finance 17: 167–88.
Roe, M. 2004. ‘Explaining Western securities markets’, in Grandori (ed.),
Corporate Governance and Firm Organization: Microfoundations and
Structural Forms. Oxford University Press: pp. 279–96.
Shearman & Sterling 2007. Report on the Proportionality Principle in the European
Union, external study commissioned by the European Commission.
Stout, L.A. 2007. ‘The mythical benefits of shareholder control’, Virginia Law
Review 93: 789–809.
Szilagyi, P.G. 2010. ‘Proxy contests’, in Baker and Anderson (eds.), Corporate
Governance: A Synthesis of Theory, Research and Practice. JohnWiley &
Sons, pp. 497–516.
White, H. 1980. ‘A heteroskedasticity-consistent covariance matrix estimator and
a direct test for heteroskedasticity’, Econometrica 48: 817–38.
Woidtke, T. 2002. ‘Agents watching agents? Evidence from pension fund owner-
ship and firm value’, Journal of Financial Economics 63: 99–131.
Zetzsche, D. 2008. ‘Shareholder passivity, cross-border voting and the
Shareholder Rights Directive’, Journal of Corporate Law Studies 8: 289–336.
8
Board elections and shareholder activism: the

Italian experiment
massimo belcredi, stefano bozzi
and carmine di noia
1. Introduction*
The board of directors performs an important role as a delegated super-
visor of executive management. A major issue in corporate governance is
how to ensure that the board performs this role effectively, in order to
guarantee that the incentives of management are aligned with those of
shareholders. The importance of board elections in this regard can
hardly be overestimated.
Board elections usually require shareholders’ vote. This is, however,
only the final step in a process which involves many different actors
operating under a set of rules. These are typically set out partly in the law,
partly in the company charter. A dysfunctional system may hamper the
effectiveness of the board as a delegated monitor, thereby affecting
agency costs and firm value.
Shareholder voting, particularly in corporate elections, has come into
the spotlight in the last few years. In the US, where the possibility of
shareholders to influence board elections was – apparently – at a histor-
ical minimum after the decline of cumulative voting (Gordon 1994), a
number of regulatory proposals aimed at empowering shareholders have
been put forward in the last decade. In the EU, the Commission issued
two Green Papers, in 2010 and 2011, respectively, claiming that a lack of
shareholder interest in holding management accountable may have
facilitated excessive risk-taking and contributed to the last financial
*
The authors wish to thank Ettore Croci, Guido Ferrarini and Eddy Wymeersch for their
helpful comments. The usual disclaimer applies. We appreciate research assistance from
Valentina Lanfranchi, Elisa Nossa, Silvia Saino and Cora Signorotto.
365
366 m. belcredi, s. bozzi and c. di noia
crisis. The adoption of voting rules reserving some board seats for
minority shareholders is one of the measures considered in this regard.
Shareholder activism has been studied extensively (Gillan and Starks
2000; 2007). Although active shareholders operate mostly through ‘quiet
negotiations’ (Becht et al. 2009; Mallin 2012), they may also target
shareholder meeting decisions: criticism by ‘voice’, shareholder voting
on management and, possibly, on shareholder proposals, could serve as a
device of external control (Cziraki et al. 2010; Renneboog and Szilagyi
2011). Shareholder activism may also target board elections.
Despite its growing importance for policy purposes, the empirical
literature on voting has not devoted particular attention to corporate
elections (Yermack 2010). Cai et al. (2009) provide the first comprehen-
sive study on board elections in US firms: shareholder votes are related to
firm performance, governance, director performance and voting mech-
anisms. However, apart from cases of gross misconduct, the probability
that incumbent directors are actually fired is negligible. A related stream
of literature investigated the effects of changes in the voting system
(Sjostrom and Kim 2007; Ertimur and Ferri 2011).
Ng et al. (2009), Cotter et al. (2010), Matvos and Ostrovsky (2010),
Choi et al. (2011) and Cremers and Romano (2011) analyse voting
strategies of US mutual funds, with particular attention to board elec-
tions. Iliev et al. (2011) extend the analysis to voting decisions of US
institutional investors in non-US firms. Mutual funds use various strat-
egies to economise on the transaction costs of making voting decisions.
They follow heterogeneous voting patterns and their voting decisions are
affected by the recommendations of proxy advisers. Finally, mutual
funds affiliated with financial institutions seem to be relatively unaf-
fected by the conflicts of interest of such institutions.1
Few studies have investigated shareholder voting in non-US firms.
Hamdani and Yafeh (2011) investigate the behaviour of institutional
investors in Israel and find that these adopt a predominantly passive
approach to board elections. De Jong et al. (2006) and Van der Elst
(2011) provide evidence on voting at Annual General Meetings (AGM)
(in Dutch firms and in large companies of five European countries,
respectively). However, none of these studies focuses on board elections.
1
Other studies, including Cai et al. (2009), Fischer et al. (2009) and Iliev et al. (2011) show
that a low shareholder vote approval in board elections may explain subsequent firm
decisions; in general, the behaviour of boards targeted by active shareholders seems better
aligned with shareholder interests.
board elections, shareholder activism: italy 367
The outcome of corporate elections may be influenced by the rules

governing nomination and voting, as well as by their interaction with the
‘constitution’ of the firm (i.e. under which conditions and to what extent
it is possible to change corporate rules, including voting rules) and with
the institutional context (e.g. firm ownership structure). Consequently,
analysing corporate elections in different contexts may offer insights into
the possible effects of various reform proposals.
Corporate elections in Italy are an interesting case, mentioned by
scholars (Zingales 2008; Ventoruzzo 2010) and the EU Commission as
a possible leading example. A board election system (‘slate voting’)
reserving some seats to representatives of minority shareholders was
introduced in Italian privatised enterprises in 1994. In 1998, slate voting
was mandated in all listed firms for elections of the Board of Statutory
Auditors members (BoSA: a vestigial supervisory body provided by
Italian corporate law). In 2005, slate voting was also mandated for
board elections. Minority shareholders sought board representation in
around 40 per cent of the cases. A second feature of the Italian example is
the succession of different regulatory systems over time, which allows
analysis of the effects of alternative rules. Furthermore, Italian regulation
is unique: few other countries have tried to introduce similar rules,
apparently with much less success. Italy is therefore an ideal candidate
to investigate corporate elections and to test the effectiveness of rules
favouring activism.
This chapter investigates empirically the probability that minority
shareholders submit a list of nominees in board elections of Italian listed
firms. It may therefore be interpreted as a ‘one-country–one-topic’ in-
depth analysis of shareholder activism. To the best of our knowledge,
this is the first non-US study in this field. Our results shed light on how
the Italian system has worked in practice and provide the basis for
evaluating the benefits of exporting it. Related papers include Barucci
and Falini (2005), analysing Italian corporate governance (including
decisions about voting rules) before slate voting was mandated in
board elections, Eckbo et al. (2011), illustrating the Italian legal frame-
work on general meetings and voting rules, and Malberti and Sironi
(2007), providing preliminary evidence about the appointment of
minority representatives (on a shorter time span and too early to fully
gauge the influence of the new legislation).
The rest of the chapter is organised as follows. Section 2 develops a
logical framework for understanding board elections and their links to
the legal and institutional context. Section 3 makes use of this framework
to analyse the main issues in the US and the European policy debate.
Section 4 describes the Italian slate voting system and its evolution over
time. Section 5 reports our empirical analysis. Section 6 presents policy
implications and concludes.
2. The anatomy of board elections

Board elections are a complex process which involves many actors,
including the incumbent board, shareholders, investment companies,
custodial institutions and service providers (head hunters, proxy advis-
ers, etc.). The rules governing the process are set out partly in the law,
partly in the company charter. From a comparative viewpoint, corporate
elections may be surprisingly diverse: within a unitary framework
(boards are appointed through a shareholder vote), details may differ
substantially.
According to the OECD Principles of corporate governance, a key
function of the board is ‘ensuring a formal and transparent board
nomination and election process’. Nomination refers to the selection of
candidates and disclosure of their names and personal characteristics to
shareholders. A formal submission ahead of the shareholders’ meeting
may be required, especially where nomination is subject to conditions set
out in the law and/or in the company charter.
In several jurisdictions, the standard arrangement is that candidates
are nominated by the incumbent board, possibly under the lead of a
nomination committee.2 This is rarely the result of a binding legal (or
charter) provision; rather, a ‘company slate’ is submitted under a set of
enabling provisions.3 The involvement of incumbent directors in the
election process (which often coincides with their own re-appointment)
may look curious, at first sight. The reasons for this have been explained
alternatively as a way to economise on transactions costs or as the
2
The establishment of such a committee is often recommended by codes of best practices;
it is also frequently recommended that such committee has a majority (or, alternatively, is
composed entirely) of independent directors.
3
An exception is Germany, where the incumbent supervisory board is bound by law to
make a proposal for the appointment of new shareholder representatives (employee
representatives are subject to a separate appointment process). Until 2004, in Dutch
companies operating under the so-called ‘structuur’ regime, new directors were selected
by incumbent directors without a shareholder vote, subject to review by the commercial
court to ensure adequate representation of shareholder and employee interests. Where
corporate law is enabling in nature, it may nonetheless require that the actual recourse to
a particular solution is authorised in the company charter.
natural consequence of a board primacy role.4 The OECD Principles

make a clear distinction between shareholders’ right ‘to elect and remove
members of the board’ and their ‘participation to the nomination and
election of board members’, which is merely recommended (‘should be
facilitated’).
Shareholders may propose their own nominees in addition to those
of the board. This will produce a contested election if the number of
candidates exceeds the number of board seats.5 The technicalities of
shareholder nomination may differ across jurisdictions and create
huge differences in transaction costs, which, in turn, could either facili-
tate or create obstacles to collective action. Two points deserve particu-
lar attention in this regard. On one hand, specific conditions (e.g. a
minimum shareholding, possibly for a determined period) may be required
to nominate a candidate. On the other hand, the mechanics of disclosure
may differ profoundly and significantly affect the final outcome. If disclo-
sure is ineffective or it is provided too late, shareholder nominees may have
no real chance to be elected.
A crucial point is that many shareholders do not have the possibility
or the incentive to attend the general meeting in person. To reduce
obstacles to collective action, shareholders are usually allowed to exercise
their voice in absentia through one of three mechanisms: (a) mail (or
remote) voting, (b) proxy solicitation by the board and/or major share-
holders, or (c) proxy voting through custodial institutions or other
agents. These mechanisms involve a number of technical issues (e.g.
may shareholders append their proposals to the company proxy? Who
bears the costs of a proxy collection? Do mutual funds, brokers and
custodial institutions have a positive obligation to vote the shares they
hold – or to abstain from voting – and to publicly declare their voting
policy?). How these issues are regulated may affect the voting outcome
and hence investors’ incentives to become active in the first place. Even
though shareholders have a right to nominate board candidates, they
4
According to proponents of this role, the board has better chances to make an informed
decision and select the best candidates; besides, unlike shareholders, the board has
fiduciary duties towards all shareholders, who still retain the possibility to choose
alternatives (Bainbridge 2012; Sharfman 2012). Of course, this position is at odds with
agency-based explanations, which argue that investor protection is insufficient and call
for shareholder empowerment (Bebchuk 2003; 2005; 2007).
5
This is not necessarily the case, if the election system provides for ‘quotas’ for different
types of candidates or if the board voluntarily submits a number of candidates which is
lower than that of the available seats.
may choose to abstain from such activity and vote for or against the
candidates proposed by the incumbent board.
Shareholder voting is the core of the process. There is nothing obvious
in the voting mechanism adopted for board elections. Voting regimes
may differ greatly and affect the allocation of power between share-
holders and the incumbent board and also among various shareholder
classes.6
An oft-cited classification distinguishes between plurality and major-
ity voting. Under plurality (also called ‘relative majority’), each share-
holder votes for one choice and the choice that receives most votes wins,
even if it receives less than a majority of votes. Plurality voting is the
default rule for board elections in a number of US states and is some-
times criticised on the basis that candidates who receive a low number of
votes (possibly by one shareholder holding one single share, if all other
shareholders withheld their votes) may nonetheless be elected. The
alternative is (pure) majority, where only candidates receiving a majority
of the votes cast at the General Meeting are elected. This is the common
standard in Europe and has also been adopted voluntarily by several
large US public companies.
Majority voting is technically a referendum on the nominees in the
company slate, in that votes withheld from a candidate may be counted
as votes against him. Unlike plurality, however, majority voting needs an
uncontested election, since it does not guarantee a sufficient number of
winners if the number of nominees exceeds that of the board seats. US
firms adopting majority voting switch to plurality if the election is
contested. Furthermore, a back-up solution must be provided in case
one or more candidates fail to gain a majority. In various US states a
‘holdover rule’ is provided, which allows for the ‘failed’ director to stay in
place until a different candidate reaches a majority (in a subsequent,
contested election). An alternative solution is a charter provision requir-
ing the failed director to tender his resignation to the board, who can
decide either to accept or reject it, sometimes according to the nomin-
ation committee’s proposal. The failed director (or a new board member
appointed after the previous has stepped down) will then remain in
charge until the next shareholder vote.
6
We focus our discussion on the mechanics of voting and take the initial allocation of
voting rights as given, i.e. we ignore the issues related to shares with differential voting
rights, voting caps, recourse to alternative systems (e.g. one-head–one-vote) etc., which
simply allocate initial voting rights across shareholders.
In a contested election, board seats may be allocated on the basis of a

single-winner or a multiple-winner method. In a single-winner system a
shareholder holding a majority of voting rights is able to appoint the
whole board; under a multiple-winner system, the outcome also reflects
the preferences expressed by minorities. This may be accomplished
through a number of mechanisms. One is to define quotas, which
allow minorities to appoint a pre-determined number of board mem-
bers. Another is proportional voting, whereby the allocation of seats
reflects the distribution of votes held by shareholders participating in
the ballot. An intermediate solution is cumulative voting, whereby each
shareholder receives, for each share, a number of votes equal to the
number of directors to be elected, which may be freely allocated
among various candidates.7 Multiple-winner systems are relatively
uncommon (Enriques et al. 2009b)8 and are prone to conflicts of interest
similar to the ones raised by single-winner systems. Therefore, a quorum
is often required for minority representation.
Independently of the method used to assign seats in a contested
election, shareholders may alternatively be free to pick individual can-
didates from a list (open list) or be forced to vote for a whole list (closed
list). In an open list system, shareholders are free to ‘unbundle’ a slate
and vote for individual nominees, while in a closed list system, the seats
are assigned to winning candidates in a fixed order chosen by the subject
who submitted the slate.
Removal rights are also important in understanding board elections.
They generally follow appointment rights: directors are removed by
7
Cumulative voting is not fully proportional because a group of voters divided among ‘too
many’ candidates may fail to elect any winners, or elect fewer than they could. The level of
proportionality depends on how well coordinated voters are. Well-known formulas
inform the minority (and the majority) how to allocate votes for maximum effect. The
greater the number of directors to be elected, the smaller the minority block necessary to
elect one director.
8
Cumulative voting is the regulatory standard in Japan, but is routinely avoided by charter
provisions. A number of jurisdictions allow cumulative voting which is, however, rarely
adopted (in the US this system has undergone a steady decline in the last few decades).
Proportional voting is the legal standard in Spain, but it rarely gives rise to a contested
election (possibly because shareholders holding a sufficient stake directly negotiate access
to the company slate with the incumbent board). In Iceland, the standard is majority
voting. However, shareholders controlling at least one-tenth of the share capital may
demand that proportional or cumulative voting is used. If there are conflicting demands,
the latter is to be employed. Similar regulatory proposals were discarded in other
Scandinavian countries (Björgvinsdóttir 2004). The Italian system based on quotas is
analysed in detail in this chapter.
dropping their names from the company slate or by failing to re-elect

them. Terms of office generally range from one-to-three years. Some
jurisdictions allow even longer terms.9 Of course, longer terms mean
stronger directors (and also weaker shareholders). Even more important
is whether, and under which conditions, shareholders may remove
directors before the end of their terms (Enriques et al. 2009a). A number
of legislations (e.g. the UK, France and Italy) give shareholders a non-
waivable right to remove directors without cause (ad nutum) and,
possibly, without compensation. In Germany, removal of shareholder-
appointed supervisory board members requires a three-quarters super-
majority (while the management board cannot be ousted without cause).
At the other end of the spectrum, a number of US jurisdictions pose
obstacles to directors’ removal.
Board election rules are at the core of corporate governance. They
allocate power within the firm and have far-reaching implications. What
may look like a minor change in the election/removal process, or in its
interaction with other aspects of legislation, may tilt the balance of power
in favour of a particular interest class. In a ‘board-centric’ system (Cools
2005), nomination, election and removal of directors are structured so as
to insulate the board from shareholder pressure, while in a ‘shareholder-
centric’ system incumbent board members may be easily ousted by
shareholders.10
9
For example, no pre-specified limit is defined by the law in the UK (where private
companies occasionally appoint directors for life). The UK Corporate Governance Code
recommends, on a comply-or-explain basis, that directors are ‘submitted for re-election
at regular intervals, subject to continued satisfactory performance’ (B.7); the (non-
binding) provision adds that all directors of FTSE 350 companies should be subject to
annual election by shareholders. All other directors should be subject to election by
shareholders at the first annual general meeting after their appointment, and to re-
election thereafter at intervals of no more than three years. Non-executive directors who
have served longer than nine years should be subject to annual re-election (B.7.1). If
terms of office are longer than one year, the board is said to be ‘classified’. A ‘staggered’
board is a classified board where the terms of office of individual directors are not
aligned (e.g. a third of the board is subject to re-election each year).
10
In the words of Cools (2005): ‘In a U.S. corporation, the center of power lies within the
board, or better, management. It can act autonomously in matters where a Continental
European board or management would depend on its shareholders. This fundamental
difference is supported by two other differences. First, it is easier for shareholders to set
the agenda of the shareholders’ meeting in Continental Europe than it is in the United
States. Second, the enabling approach of the Delaware legislature allows the board to
assume several powers of the shareholders’ meeting. In contrast, in Continental Europe,
the statutory allocation of powers is mandatory and even with the permission of the
The balance of power is deeply intertwined with the firm’s ownership

structure. If existing, the controlling shareholder will be able to appoint
the board (or a substantial part of it), whichever proposal is made by the
incumbent board. Consequently, the technicalities of the election pro-
cess are less important and the voting power is what matters. On the
opposite, if no controlling shareholder exists, or if a multiple-winner
system is in place, the technicalities of the election process may affect the
incentive to submit alternative candidates, the voting outcome and the
resulting board composition. The regulatory system may, however,
backlash and influence the firm ownership structure in the first place.
Where the board is insulated from shareholder pressure, the founder can
sell out virtually all equity without losing control: he only needs to stay
on the board when the company goes public. On the contrary, where
control is in the hands of shareholders, the founder must keep a consid-
erable portion of the voting rights to retain control of the company’s
decisions.
In sum, the process leading to the appointment (and removal) of
directors is a key issue in corporate governance. Inevitably, it came
into the spotlight after the financial crisis and reform proposals were
put forward in this regard. However, the US and European approaches
differ considerably.
3. The policy debate on board elections

3.1. The debate in the US . . .
The policy debate in board-centric US has centred, at least since the early
2000s, on two main issues: shareholder access to the corporate ballot and
the behaviour of brokers and custodial institutions.
The debate on proxy access has been particularly harsh (Gordon
2008). The election system in US companies is typically based on a
single-winner voting system. The nomination process is controlled by
the incumbent board and substantial impediments to directors’ removal
are in place. Board elections in US companies mostly go uncontested.
Company nominees (often the incumbent directors themselves) are
elected without substantial opposition. According to an often quoted
shareholders, the board cannot appropriate most of their powers.’ This definition
centres on election/removal rights and differs somewhat from that used in the first
chapter, which was based on the general division of powers between the board and the
shareholders (Davies et al. 2012).
statement by a former SEC chairman: ‘A director has a better chance of

being struck by lightning than losing an election’ (Levitt 2006).
Board control over the ballot should be put into a historical perspec-
tive. A single-winner voting system has not always been prevalent in the
US. Since the late nineteenth century, a number of states permitted or
even mandated cumulative voting for corporate elections (Gordon
1994).11 Consequently, active shareholders could get board representa-
tion. After a series of high-profile proxy contests, however, mandatory
cumulative voting started to decline in favour of the permissive form.
The elimination of cumulative voting on a firm-by-firm basis paralleled
the collapse of mandatory cumulative voting in the states. The final blow
came with the takeover wave of the 1980s, when a number of companies
chose to eliminate cumulative voting (possibly after re-incorporation in
a permissive state) as part of a complex anti-takeover strategy. In 1992
only 14 per cent of the Fortune 500 companies still retained cumulative
voting. The percentage declined further afterwards (Pozen 2003). This
phenomenon transferred further power from the shareholders to the
board.12 Unsurprisingly, the debate about shareholder access to corpor-
ate ballot became more intense when their powers in corporate elections
reached a historical minimum.
It is widely recognised that board control over the ballot depends
crucially on the regulation of proxy collection, which bans shareholders
from appending their nominees to the company slate. A shareholder
wishing to nominate different candidates may file his own proxy state-
ment. However, this implies he will bear campaign costs (unlike incum-
bents, whose costs are borne by the company). Since they will share the
benefits from improved corporate governance with other shareholders,
rational apathy is often the result (Easterbrook and Fischel 1991).
Therefore, proxy contests are quite rare (Bebchuk 2005; 2007; Kahan
11
By the late 1940s, twenty-two states had mandatory and fifteen had permissive cumu-
lative voting provisions. Cumulative voting was found in 40 per cent of a sample of 2,900
large corporations.
12
Gordon (1994) reports that the most common justification offered in proxy statements
was ‘a double-barreled attack on the principle and consequences of minority board
representation: directors “should represent all shareholders, rather than the interests of a
special constituency, and [. . .] the presence on the Board of one or more director
representing such a constituency could disrupt and impair the efficient management
of the Corporation”’. Anecdotal evidence shows that cumulative voting was sometimes
used by corporate raiders and greenmailers. The near-success of a labour union leader in
gaining a board seat in a major public utility is claimed to have been a major factor in the
repeal of mandatory cumulative voting in California.
and Rock 2011). Concurrent reasons for shareholder apathy are the limi-
tations imposed to institutional investors, notably through the regulation of
‘acting in concert’ (Roe 1994), which may imply strong limitations to
trading, onerous disclosure obligations, liability – as controlling persons –
for company obligations, claims’ subordination in case of bankruptcy and,
finally, a less favourable tax treatment. These features of the US legislation
may explain why institutional investors show limited interest in multiple-
winner systems (such as cumulative voting, which implies a conspicuous
role in the nomination process) and apparently favour majority voting,
where they may ‘just say no’ to company nominees.
Proponents of shareholder access to the ballot argue that competition
in the election process would increase the activism of institutional
investors and benefit all shareholders (Bebchuk and Hirst 2010; Becker
et al., 2010). Critics raise concerns on the risk that certain shareholders
could use their power to pursue objectives in contrast with firm value
maximisation (Bainbridge 2003; 2010; Larcker et al. 2010; Sharfman
2012). Other scholars cast doubts on investors having sufficient incen-
tives to become active even under different rules (Kahan and Rock 2011).
After a ten-year debate, the Dodd-Frank Act gave the SEC the authority
to enact a proxy access rule. In August 2010 the SEC adopted Rule 14a-11,
allowing shareholders holding at least 3 per cent of equity capital to use –
under certain circumstances – the company’s proxy statement to solicit
votes for their nominees (one or 25 per cent of the board, whichever is
greater). However, Rule 14a-11 never became effective, since two business
groups sued to block it and a federal appeals court agreed that the SEC had
failed properly to assess the rule’s economic impact. The Commission also
amended Rule 14a-8, allowing eligible shareholders to include proposals
regarding proxy access procedures in company proxy materials. The
amended Rule 14a-8 became effective after the court decision on Rule
14a-11. Consequently, the US debate on proxy access and, in general, on
voting procedures, is far from over. Proxy access and other corporate
governance proposals will be discussed on a company-by-company basis.
Broker voting has also been targeted by recent regulation. NYSE and
SEC rules require that brokers deliver proxy materials to beneficial
owners and request voting instructions. The previous NYSE Rule 452
permitted brokers to exercise discretionary voting authority on shares
held ‘in street name’ on ‘routine’ matters when they received no voting
instructions. Uncontested director elections have long been considered
routine matters and brokers usually voted for company nominees,
thereby contributing to reinforce the management position. Such rule
was amended in 2009:13 uncontested elections are no longer considered

routine matters. Consequently, uninstructed shares have de facto been
sterilised; this makes it easier for companies adopting majority voting to
fire directors with whom investors are unhappy, leaving the selection of a
substitute to the board.
3.2. . . . And in Europe

In shareholder-centric European countries, the nomination process does
not seem to be much of an issue. Even though nominees are usually
proposed by the incumbent board, shareholders may easily submit
alternative candidates, as long as they do not intend to take control of
the firm (in which case they could fall under the mandatory bid rule14).
No US-style bottleneck to shareholder participation seems to be present
in the nomination process. The EU Commission recommended the
creation of a nomination committee composed of a majority of inde-
pendent non-executive directors (Recommendation 2005/162/EC). The
Shareholders’ Rights Directive (2007/36/EC) granted shareholders
the right to table draft resolutions for items included or to be included
on the agenda of a general meeting. Where such right is subject to a
minimum shareholding condition, such minimum stake shall not exceed
5 per cent of the share capital.
Board election rules have received increased attention in the last few
years. According to the 2010 EU Green Paper on Financial Institutions
and Remuneration Policies, one of the reasons for the disappointing
13
The rule was further amended by the Dodd-Frank Act to prohibit broker discretionary
voting on executive compensation-related agenda items. The latest change to date (in
January 2012) prohibits uninstructed voting on corporate governance proposals. The
change reverses the previous policy of permitting discretionary voting on ‘shareholder
friendly’ governance proposals provided that management was recommending in favour
of the resolution (the NYSE deemed proposals to be ‘non-routine’ if management was
recommending against the resolution or made no recommendation). The NYSE men-
tioned as examples proposals to ‘de-stagger the board of directors, majority voting in the
election of directors, eliminating supermajority voting requirements, providing for the
use of consents, providing rights to call a special meeting, and certain types of anti-
takeover provision overrides’. The last amendment virtually eliminates discretionary
voting altogether with the exception of a few items.
14
Even though national rules in this field may differ (Santella et al. 2009), they do not
usually preclude collective action, insofar as nomination of shareholder candidates aims
at gaining a minority representation on the board. In the UK, the Takeover Panel
adopted clear guidelines for distinguishing between ‘acting in concert’ and corporate
governance activism.
performance of corporate boards in supervising managers of financial insti-

tutions during the financial crisis was the passive behaviour of institutional
investors, largely due to the information and contracting costs determined by
active engagement. Free-riding problems may affect non-financial companies
as well. The EU Commission issued a second Green Paper in 2011, putting
forward proposals aimed at improving the EU Corporate Governance frame-
work in general, with a particular focus on listed companies.
One of the proposals is to require institutional investors to publish their
voting policies and records, at the same time requiring proxy advisers to
disclose any conflicts of interest and the methods applied in the preparation
of their advice. These proposals are not aimed at directly influencing
investors’ incentives to become active, but rather, address – through
increased disclosure – the way conflicts of interests are managed.
A second proposal specifically regards firms with controlling share-
holders, which remain the predominant governance model in Europe.
According to the EU Commission, minority shareholder engagement
can be particularly challenging in such companies and the existing EU
rules may not be sufficient to protect minority shareholders’ interests
against potential abuse. As a possible solution, ‘certain Member States
(e.g. Italy) reserve the appointment of some board seats to minority
shareholders’. The Commission implies that a multiple-winner election
rule favouring the appointment of ‘minority directors’ could overcome
rational apathy and improve investor self-protection.
Such a claim leaves a number of questions unanswered. First, board
election rules in Europe are largely in the shareholders’ domain; assuming
that minority representation grants superior investor protection, it is not
clear why this has not yet emerged as a widespread market solution.15
Consequently, a call for regulation is not clearly justified. Furthermore,
even if a case for regulation can be established, it remains to be seen if
election rules should be mandated or should take an enabling form, aimed
at removing possible regulatory obstacles to minority representation.
Second, a rule granting board representation to minority shareholders
may influence benefits and costs of activism (Enriques and Volpin 2007).
However, the true extent to which board election rules may affect such
15
When a firm goes public, the controlling shareholder has the incentive to define the best
set of governance rules (including board elections) to maximise the returns from the
share issuance. The Italian case seems to fit well into this framework, since a multiple-
winner voting system was first introduced by the state for companies undergoing the
privatisation process.
costs remains unclear. Direct costs include expenses (the time of senior
executives involved in activities such as the selection of candidates,
coordination with other shareholders, plus out-of-pocket expenses for
proxy solicitation and other campaign costs) which are necessary to
successfully present a list of candidates. Indirect costs are less visible,
yet not less real, and may include limitations to trading implied by
market abuse regulation, suboptimal diversification (where activism
requires a large and/or long-term investment in the company), legal
liability for acting in concert and potential litigation costs (Pozen 2003).
Third, since minority shareholders are also self-interested, it is not
clear whether granting them access to the board will produce better
incentive alignment. In fact, an investor holding a small block of shares
will face higher conflicts of interest compared to a controlling share-
holder. Consequently, he may have an incentive to collude with manage-
ment (and/or with the controlling blockholder), to greenmail the
company or to use his position to other ends (e.g. to collect information
useful for other business purposes).
Any regulatory proposal should be analysed in detail in order to prove its
feasibility and efficiency, in terms of a proper cost–benefit comparison.
Within this framework, an analysis of previous national experiences may be
useful. The Italian case looks particularly interesting: Italy introduced a
multiple-winner system for corporate elections, which has proved quite
effective in stimulating activism. Minority shareholders got board represen-
tation in around 40 per cent of the cases (Assonime-Emittenti Titoli 2011).
Therefore, Italy offers a unique opportunity for investigating the influence
of regulation on shareholder activism in corporate elections.
4. Board elections in Italy

Until 2003, the Italian regulation mandated, substantially, a one-tier
board structure. Under this ‘traditional’ governance system, sharehold-
ers appoint both a board of directors (BoD, Consiglio di amministra-
zione) and a separate Board of Statutory Auditors (BoSA, Collegio
Sindacale).16 Since 2003, Italian companies may switch either to a true
16
The BoSA is composed of 3 (or, less frequently, 5) independent members (who have a
background in law or accounting), attending board meetings and monitoring compli-
ance with the law and with the company charter, as well as the adequacy of the
company’s organisational structure and of the internal control, administrative and
accounting system.
one-tier system (i.e. with an audit committee within the board and no
BoSA) or to a two-tier system, following the German model, albeit with a
number of important differences (e.g. no employee representation).
Alternative board models have had limited success, being chosen by
only a handful of companies.17 Almost all Italian boards are ‘classified’
(the standard is a three-year mandate) but only few are ‘staggered’ (the
whole board is almost always appointed through a unitary vote).
Board elections in Italy are based on a multiple-winner system.
Directors are drawn from lists of candidates (‘slates’), to be submitted
ahead of the shareholders’ meeting. At least one seat must be reserved to
minority nominees. Even though alternative solutions (e.g. proportional
voting) are occasionally adopted, quotas are by far the prevalent solu-
tion. The slate receiving the highest number of votes takes all but a
predetermined number of seats (set out in the by-laws), which are
reserved to candidates chosen from minority slates. Plurality voting is
the norm: in a contested election a director may be appointed even
though he receives a low number of votes. To avoid the issues related
to plurality voting, the bylaws usually provide for both a shareholding
and a voting quorum. A shareholding quorum, within an upper limit set
out by Consob – the Italian market supervisor – is required to submit a
slate. Furthermore, the election of minority candidates may require a
further voting quorum, which may not exceed 50 per cent of the share-
holding quorum.18 Slates are usually submitted by relevant shareholders
and a company slate prepared by the board is uncommon.19
Consequently, the nomination committee is substantially redundant.
Voting usually takes place on a closed list basis, i.e. shareholders
cannot express preferences for individual nominees. Directors are
17
At the end of 2010, only 7 (3) companies had adopted the two- (one-) tier model.
Numbers are substantially stable over time.
18
Occasionally one or more candidates have been proposed directly at the AGM by
shareholders not meeting the shareholding quorum requirement. This opportunity
has also been taken by mutual funds (e.g. in Saipem, a privatised company). This
might happen (with the consent of a majority of shareholders) in cases where no
minority slate was submitted within the prescribed term and the majority slate did not
include a sufficient number of nominees, to serve as a ‘backup’ in cases where minorities
remain passive.
19
Only a handful of companies adopted by-laws allowing incumbent boards to submit a
list. A company slate could, in fact, be problematic in a contested election, due to a ‘no-
link’ provision present in Italian regulation: minority slates may not be submitted and/or
voted by on shareholders ‘linked in any way, even indirectly, with the shareholders who
presented or voted the most voted list’ (see below for further details).
appointed according to a so-called ‘quotient’ method, whereby the votes

received by each slate are divided by a sequence of whole numbers, from
one up to the total number of directors to be elected. The resulting
quotients are assigned progressively to the candidates of each slate, in
the order, in which they are listed. The quotients attributed to all
nominees are then arranged in a single list, in decreasing order and the
persons with the highest quotients are elected.
Slate voting was first introduced in Italy by Law 474/1994, which regu-
lated the privatisation of publicly owned enterprises: article 4 states that the
by-laws of particular classes of companies undergoing privatisation should
adopt slate voting for board elections. Although the technical instrument
used (i.e. law) was uncommon, one could still argue that slate voting was
introduced on a voluntary basis to increase share marketability, since the
state was the ultimate shareholder of the companies to which the Law
applied. Under this law, slates could be submitted by shareholders holding
at least 1 per cent of share capital and at least one-fifth of board seats must go
to representatives of minority shareholders.20
According to the privatisation law, slate voting should be adopted by
firms meeting the following conditions: (a) they operated in one of the
‘strategic’ industries specifically enumerated in the law (defence, trans-
ports, telecommunications, energy, public utilities, financial); (b) the
state retained ‘golden share’ powers; (c) a cap to shareholdings (and/or
voting rights) held by private investors was provided. Slate voting was
therefore part of a complex project aimed either at creating a dispersed
ownership structure or at keeping control in public hands with a minor-
ity shareholding. The rationale underlying slate voting was to avoid an
entity holding a small fraction of equity capital dominating the board
and, at the same time, guaranteeing that privatised companies would
follow a shareholder value (i.e. not a stakeholder-oriented) approach.
Minority shareholders often sought representation in privatised companies.
Italian mutual funds (and their category association, Assogestioni) took a
leading role in the nomination and election process.
In 1998, the Consolidated Law on Finance (CLF) mandated slate
voting for BoSA elections in all listed firms. The success of slate voting
20
From a theoretical standpoint, a minority candidate may be identified only ex post, on
the basis of the actual voting outcome. However, ownership in Italy is typically con-
centrated, and there is usually little doubt about who is going to be a majority candidate.
The seats reserved to minorities were often voluntarily increased to one-third of the
board.
was, however, much smaller than in privatised enterprises. Consob

repeatedly remarked that minority representatives were appointed only
in a few cases. It was debated whether this was due to investors’ lack of
interest or to a high level of the quorum required to submit a list (for
which no legal cap was provided21). Current practice was also criticised,
since anecdotal evidence showed that minority candidates had some-
times been proposed by subjects ‘linked’ to the controlling blockholder.
Slate voting was mandated for board elections by Law 262/2005 (the
‘Protection of Savings’ Law), which considered board representation as a
tool for increasing minority shareholders’ self-protection against pos-
sible abuse.22 Following the new regulation, even companies where slate
voting was already in place had to profoundly revise their election rules.
According to the new system, voting rules are to be defined in the
company by-laws and are therefore subject to shareholder approval. At
least one seat must be reserved to minority candidates and the quorum
required to submit a list must not exceed 2.5 per cent of share capital or
the ‘different extent established by Consob’, on the basis of capitalisa-
tion, free-float and ownership structure. Finally, to avoid the adoption of
entrenchment strategies, Consob was delegated to define rules ensuring
that at least one director was appointed by minority shareholders not
21
In 2007, minority statutory auditors had been appointed in sixty-three companies (23
per cent of the aggregate). In 2004, Lamberto Cardia, Chairman of Consob, stated that
‘The request to lower the ownership thresholds required to submit a list, put forward in a
number of recent AGMs, seems a justified measure to avoid a substantial circumvention
of the rule.’ However, the average quorum to submit a slate was rather low (2.78 per cent,
as of February 2007; only in two small cap firms – Premuda and Exprivia – did the
quorum size exceed 5 per cent of share capital – 10 and 8 per cent, respectively). If
companies had adopted an alternative system (similar to the Spanish-style proportional
voting), a much higher quorum would have been necessary to gain representation: a 10
per cent stake would have been needed to appoint a director in the average 10-member
Italian board; a sky-high 33 per cent would have been necessary to appoint a statutory
auditor. The numbers reported are drawn from a ‘historical’ database, on file with the
authors, hand-collected from public documents (company by-laws and annual
Corporate Governance reports).
22
In December 2003 Parmalat, the food company, filed for bankruptcy. A number of
abuses emerged from criminal investigations, finally leading to the incrimination of
Calisto Tanzi, the chairman and founder of the company. Attempting to avoid similar
scandals was one of the reasons for legal reform. Interestingly, Parmalat had one
statutory auditor drawn from a minority slate submitted by Italian mutual funds.
However, the auditor in question was not re-appointed, since no minority slate was
submitted at the last election before default; this was interpreted by some commentators
as consistent with institutional investors ‘voting with their feet’, given the low degree of
disclosure about the true financial position of the company.
‘linked in any way, even indirectly, with the shareholders who presented
or voted’ the most voted list (‘no-link’ provision).
The new system created a major discontinuity. The implementation
rules proposed by Consob underwent a lengthy consultation process and
were not incorporated in company by-laws until June 2007 (to be applied
from 2008 elections). The 2.5 per cent quorum cap was replaced with six
different thresholds (ranging from 0.5 to 4.5 per cent of share capital),
inversely proportional to market capitalisation23 and criteria were
defined to sort out in advance which relationships would create a mate-
rial ‘link’ between shareholders submitting and/or voting on a list.24 If a
person linked to the ultimate shareholder voted for a minority list and
that person were pivotal for the election outcome, his votes would be
discounted for the election of the first minority representative.
Regulation of board elections was last revised in 2010, when the EU
Shareholders’ Rights Directive was transposed into the Italian
23
The 4.5 per cent category is accessible to small caps with a free-float above 25 per cent
where no individual shareholder (or group of subjects tied by a shareholders’ agreement)
holds a majority of voting rights. Privatised companies are still also subject to the rules
established by Law 474/1994. Finally, the ownership threshold is capped to 0.5 per cent
of share capital in cooperatives operating under a one-head–one-vote regime. The
number of thresholds was cut to four in May 2012.
24
A material link between the ultimate shareholder(s) and one or more minority share-
holders is deemed to exist at least in the following cases: (a) family relationships; (b)
membership of the same group; (c) control relationships between a company and those
who jointly control it; (d) relationships of affiliation pursuant to Article 2359, subsection
3 of the Italian Civil Code, including with persons belonging to the same group; (e) the
performance, by a shareholder, of management or executive functions, with the assump-
tion of strategic responsibilities, within a group that another shareholder belongs to; (f)
participation in a ‘control’ shareholders’ agreement (‘regarding the exercise of voting
rights’, according to art. 122, para. 5 CLF) concerning the issuer, its parent company, or
one of its subsidiaries. Although these rules refer to BoSA elections, they are also
interpreted as relevant parameters in board elections. In a handful of cases Consob
issued an official communiqué after a slate had been submitted but before the
shareholders’ meeting, arguing that a minority shareholder was linked to other share-
holders who had submitted or, simply, who were likely to vote on another list and that,
consequently, in case the election produced a particular outcome, his appointee could
fall under the no-link provision. In some occasions, the shareholders addressed replied
that no such link existed. However, the list was always withdrawn before the meeting.
One famous case was the Assicurazioni Generali BoSA election (a contested one, since
four slates had been submitted). Edizione Holding, a company owned by the Benetton
family, was targeted by Consob, arguing that the Benetton family was part of a ‘control’
agreement in Mediobanca, i.e. Assicurazioni Generali’s most important blockholder.
The remaining slates had been submitted – respectively – by the incumbent board of
directors, by mutual funds and by a hedge fund aggressively targeting the incumbent
management.
legislation, introducing the record date system. Investors can now exer-
cise their voting rights for all shares held at the record date and are no
longer required to deposit their shares until the AGM date to keep their
voting rights. The previous system had often been blamed by institu-
tional investors as the main culprit for their passive behaviour in board
elections, as reducing the liquidity of their trading portfolio.
5. Empirical analysis
5.1. Sample description
Corporate elections in Italy are unique. A multiple-winner board elec-
tion system has been in place since the mid 1990s. Active investors
submitted alternative slates fairly frequently. Activism is sufficiently
variable across companies to allow an analysis of its determinants.
Finally, different regulatory systems have been adopted over time (enab-
ling the analysis of the effects of alternative rules). Italy is therefore an
ideal candidate for investigating corporate elections and testing the
effectiveness of rules favouring activism.
We perform a regression analysis aimed at identifying the determin-
ants of minority shareholder activism. We use a proprietary dataset,
collected manually from AGM minutes, Corporate Governance reports
and additional official documents (e.g. company by-laws): data on voting
(e.g. election rules, identity of the shareholders who submitted a list,
characteristics of shareholders and candidates) are drawn from this
source. Financial and market data come from Datastream and
Worldscope. Ownership variables are calculated on the basis of official
data (published by Consob). The variables are defined in the Appendix.
Due to the limited availability of AGM minutes prior to 2008, we
focused our analysis on the 2008–10 period, i.e. following the imple-
mentation of the ‘Protection of Savings’ Law. We chose a three-year
sample period to cover elections in all listed firms and ignored the
appointment of individual directors, which typically did not make use
of slate voting. We identified 283 companies listed on the Italian Stock
Exchange where a board election took place in the sample period.25 After
25
Coverage of the Italian Stock Exchange is almost 100 per cent: the number of Italian
listed companies was 293 in 2008, 282 in 2009 and 273 in 2010. The number reported in
the text is lower than the 2008 total because some firms were delisted before an election
could take place according to the new rules (plus a handful of IPOs, where an election
took place only after 2010).
Table 8.1 Descriptive statistics: firm characteristics
25th 75th
Mean Median percentile percentile Std dev Observations
age 46.4 28.0 13.0 67.0 51.0 261
age from listing 16.0 9.0 3.0 19.0 22.3 261
assets 14,252 467 154 2,361 81,007 261
(€ million)
mktcap 1,936 188 60 776 7,827 260
(€ million)
q ratio 1.35 1.09 0.90 1.53 1.57 261
EBITDA/sales 15.0% 11.3% 4.4% 22.9% 144.7% 261
leverage 30.2% 30.1% 17.7% 41.1% 17.9% 260
ROA 6.2% 6.9% 1.7% 11.5% 12.0% 259
6-month stock −10.7% −13.9% −26.6% −2.2% 27.8% 261
returns
Note: Summary statistics for firm age, size, profitability, and capital structure for
the whole sample, period 2008–10. All variables are defined in the Appendix.
dropping 20 companies whose AGM minutes were missing or incom-

plete and two companies with other missing data, we end up with a
sample of 261 board elections. Summary statistics are reported in
Tables 8.1–8.3.
Table 8.1 reports data about age, size, financial structure and results.
Italian listed companies are usually small or medium enterprises. The
median firm has total assets of around EUR 500 million, a market
capitalisation of less than EUR 200 million, age of around 30 years and
has been listed for 9 years. Even in the midst of the worst crisis of the last
decades (the mean 6-month rate of return on common stock before the
election date was minus 11 per cent), leverage was not particularly high
(30 per cent) and financial results were substantially sound (median
EBITDA and net income in the previous year were positive, around 11
per cent and 2.6 per cent of net sales, respectively).
Tables 8.2(a) and 8.2(b) report data on ownership structure.
Consistent with previous literature, we classified firms according to the
identity of the ultimate shareholder, which may alternatively be: (a) a
family; (b) a financial institution (a bank or an insurance company); (c) a
private equity fund; (d) the state or another public entity; (e) other
Table 8.2(a) Descriptive statistics: ownership structure according to the identity of the ultimate shareholder
Type of No. of % of CFR (%) held by the ultimate VR (%) held by the Wedge Concentration Free Float
owner Firms Firms shareholder ultimate shareholder (%) index (96)
Family 174 66.7 51.09 54.61 3.52 394.680 32.44
Financial 5 1.9 42.38 43.64 1.27 366.458 33.04
institution
Private equity 10 3.8 41.71 41.71 0.00 227.338 41.09
State 22 8.4 40.04 44.59 4.55 253.067 35.07
Other 33 12.6 34.15 35.18 1.03 195.216 37.35
Widely held 17 6.5 7.44 7.44 0.00 10.211 79.58
Total 261 100.0 47.07 50.15 3.08 325.529 36.70
Note: Summary statistics for firm ownership structure, grouped according to the identity of the ultimate shareholder. Data for the
whole sample, period 2008–10. All variables are defined in the Appendix.
Table 8.2(b) Descriptive statistics: ownership structure according to the identity of the ultimate shareholder
% of Firms
No. No. of Firms with a 2nd Mean voting rights No. of Firms where a % of Firms where a
Type of of % of with a 2nd large large held by the 2nd largest shareholders’ shareholders’
owner Firms Firms shareholder shareholder shareholder (%) agreement is in place agreement is in place
Family 174 66.7 162 93 9.14 47 27
Financial 5 1.9 4 80 9.36 3 60
institution
Private 10 3.8 10 100 8.43 2 20
equity
State 22 8.4 21 95 11.20 6 27
Other 33 12.6 30 91 9.70 8 24
Widely held 17 6.5 11 65 5.30 5 29
Total 261 100.0 238 91 9.18 71 27
Note: Summary statistics for firm ownership structure, grouped according to the identity of the ultimate shareholder. Data for the
whole sample, period 2008–10. All variables are defined in the Appendix.
subjects.26 If no shareholder holds more than 10 per cent of voting rights,

the company is classified as widely held.
Around two-thirds of Italian listed firms are under the control of a
family, holding on average a 51 per cent block; 8 per cent are controlled
by the state or by other public entities; 12 per cent are controlled by other
subjects; while 6 per cent are widely held. The recourse to control-
enhancing mechanisms (dual class shares and pyramids) has been
quite limited in recent years. The average wedge between voting and
cash-flow rights is around 3.0 per cent and, curiously, is higher in
companies under public control. Listed companies have almost always
one or more other relevant shareholders. The average stake of the
second-ranking shareholder is substantial, around 9 per cent of share
capital. Shareholder agreements are present in around one-quarter of the
cases.
Table 8.3(a) and 8.3(b) reports data on the voting system in individual
firms. Board size (potentially affecting the voting outcome) varies
greatly, ranging from eight members – on average – in firms controlled
by private equity investors, to sixteen in widely held companies (which
need to accommodate several shareholders holding a small stake). The
average quota reserved to minority nominees27 also varies, from 1.1 seats
in firms controlled by private equity investors to 2.4 in state-owned
companies (a number probably driven by the privatisation law).
Consequently, board seats reserved to minorities range between 10 per
cent (in widely held firms) and one-quarter of the aggregate (in state-
owned companies).
As already noted, regulation caps the shareholding quorum required
to submit a list. The average quorum is low (2.3 per cent) and does
not seem to create an obstacle to investors seriously wishing to take an
active role (it is around one-quarter of the average stake held by the
26
A residual category, composed of firms where one or more shareholders hold a relevant
stake, but insufficient to exert control (10 per cent<X<30 per cent, where 30 per cent is
the relevant threshold for the mandatory bid rule under takeover legislation), plus
subsidiaries of such companies. A good example is RCS Media Group (the publisher
of the newspaper Il Corriere della Sera) where Mediobanca held a 15 per cent stake (i.e.
above the 10 per cent threshold used to define widely held firms), the Agnelli family held
a 10.5 per cent stake, while 9 Italian entrepreneurial families (Pesenti, Rotelli, Della
Valle, etc.) and 4 financial firms held stakes between 2 and 8 per cent (for a total around
70 per cent). Dada, one of its subsidiaries, is also included in this category.
27
Here a caveat is in order: in a handful of companies adopting proportional voting,
minority shareholders have been able to appoint a number of directors greater than the
number of seats (generally, the quota is one) reserved to them.
Table 8.3(a) Descriptive statistics: board elections according to the identity of the
ultimate shareholder
No. of Board % of Board seats Quorum

seats reserved to reserved to requested to
Type of No. of % of Board minority minority submit a
owner Firms Firms size candidates candidates slate (%)
Family 174 66.7 9.2 1.1 13.4 2.5
Financial 5 1.9 13.4 1.6 12.2 1.9
institution
Private 10 3.8 8.4 1.1 14.2 2.9
equity
State 22 8.4 10.4 2.4 23.6 1.4
Other 33 12.6 11.1 1.5 14.1 2.5
Widely held 17 6.5 15.7 1.5 9.8 1.4
Total 261 100.0 10.0 1.3 14.1 2.3
Note: Summary statistics for board elections, grouped according to the identity of the
ultimate shareholder. Data for the whole sample, period 2008–10. All variables are defined
in the Appendix.
Table 8.3(b) Descriptive statistics: board elections according to the identity of

the ultimate shareholder
No. of firms % of firms No. of %

where minority where minority contested Contested
Type of No. of % of slates were slates were minority minority
owner Firms Firms submitted submitted elections elections
Family 174 66.7 53 30.5 13 24.5
Financial 5 1.9 2 40.0 0 0.0
institution
Private 10 3.8 5 50.0 0 0.0
equity
State 22 8.4 19 86.4 3 15.8
Other 33 12.6 18 54.5 3 16.7
Widely held 17 6.5 9 52.9 1 11.1
Total 261 100.0 106 40.6 20 18.9
Note: Summary statistics for board elections, grouped according to the identity of the
ultimate shareholder. Data for the whole sample, period 2008–10. All variables are
defined in the Appendix
second-largest shareholder – 9.1 per cent). The quorum is lower (1.4 per
cent) in widely held and in state-owned companies and, typically, higher
in smaller, firms controlled by a family (2.5 per cent)28 or by private
equity investors (2.9 per cent).
Minority slates were submitted in 106 firms (around 40 per cent of the
aggregate). In a majority of cases only one list was presented, usually by
the controlling blockholder. The frequency of minority slates is lowest in
family firms (30 per cent) and highest in state-owned companies (86 per
cent). As slate voting is usually based on quotas, a contested election
takes place only when two or more minority slates run for the seats in
the quota. Contested elections are relatively rare (they take place in
less than one-fifth of the cases). Where minority slates are submitted,
their average number is 1.25. Since detailed information about the own-
ership structure is publicly available, investors choose to avoid the costs
of activism where an alternative list would have a low probability of
success.
5.2. Main hypotheses and methodology

We perform a cross-sectional regression analysis aimed at identifying
the factors affecting the probability that minority shareholders submit a
list of board nominees. We test the following (not mutually exclusive)
hypotheses:
(a) Activism may be driven by investors’ desire to monitor management
more closely through the appointment of minority directors.
According to this monitoring hypothesis, the submission of minor-
ity slates should be more likely where the risk of conflicts of interest
is higher.
(b) The voting system adopted at the company level (heavily influenced
by regulation) could also affect activism. According to a regulation
hypothesis, the submission of minority slates should be more likely
where regulatory conditions are more favourable.
(c) Activism may be influenced by the structure of the implied trans-
actions costs. According to a transaction costs hypothesis, the sub-
mission of minority slates should be more likely where the
cost–benefit comparison is more favourable.
28
Median total assets are EUR355 million in family firms and EUR165 million in compa-
nies controlled by private equity investors.
In the second stage of our analysis we turn to institutional investors’ (in

particular, mutual funds’) activism in corporate elections. At first sight,
mutual funds are not good candidates for this activity: their stakes in
individual companies are small, their investment horizon is short-term
and activism may negatively affect the liquidity of their portfolio.
Nonetheless, Italian asset managers have a long history in this field,
which may seem less strange once the institutional framework is taken
into account. The Italian stock market holds a medium weight in the
international arena (total market cap was around EUR450 billion at the
end of 2010). The single largest company (Eni) accounts for 15 per cent
of total market cap, while the first 5 (10) companies account for around
45 per cent (60 per cent) of the aggregate. Equity funds (both Italian and
global) wishing to ‘buy Italy’ have no real alternative to buying blue
chips.29 A considerable part of this investment is, actually, long-term and
may justify active behaviour.
How (and to what extent) could Italian mutual funds overcome the
obstacles to collective action underlined in other contexts (notably, in
the US)? As a first step, a Corporate Governance Committee was estab-
lished in 1994 by Assogestioni (the association of Italian investment
companies) in order to reduce transaction costs, and adopt a formal
and transparent decision-making process. The terms of reference of the
committee include the selection of candidates, to be drawn from a list
prepared by a primary executive search consultant.30 Slates are then
formally submitted by mutual funds collectively holding the needed
shareholding quorum.
This arrangement may expose mutual funds to three basic risks. First,
coordination could be interpreted as a shareholders’ agreement or as acting
in concert, thereby exposing asset managers to additional obligations.31
29
A number of reasons may justify this strategy, including higher liquidity, compliance
with regulatory benchmarks, availability of derivatives for hedging purposes and the
possibility to replicate the general market index with low transactions costs.
30
However, the committee (composed of executives of the main Italian asset managers)
seems to have the power to add (or delete) names to (or from) the list, which makes the
roles of the various actors unclear. The committee votes on individual candidates on a
one-head–one-vote basis; this should reduce the influence of individual asset managers
on the selection process and contribute to keeping possible conflicts of interest under
control. References are drawn from materials available (in Italian only) on the associ-
ation website and from additional information provided by Assogestioni officers.
31
This risk was considered so serious that Consob actually provided a ‘safe harbour’ for
shareholders cooperating to submit ‘minority’ slates; to this end the list must include a
number of candidates lower than half of the seats up for election, or – by design – be
Second, most asset managers are part of a banking or an insurance group

(Bianchi and Enriques 2001), which may hold a direct or indirect stake in
the firm.32 Consequently, they might sometimes fall under the ‘no-link’
provision. Finally, any contact with appointed directors exposes asset
managers to risks of violations (or limitations to trading) under the
market abuse legislation.33
Asset managers adopted a number of measures to keep such risks
under control. Nomination is formally separated from slate submission
(the former is managed by the Corporate Governance Committee, the
latter by individual funds). Mutual funds committed to running only for
minority quotas34 and to choosing high-profile professionals35 as nomi-
nees. Asset managers usually abstain from submitting board candidates
in their respective controlling bank (or insurance company).36 Finally,
asset managers recommend that appointees strive to keep their inde-
pendence and, at the same time, listen to the ‘opinion of the market’.37
preset for the election of representatives of minority interests. The ‘safe harbour’
provides, however, no general exemption: art. 44-quater Consob Regulation concerning
issuers states merely that such cooperation shall not ‘per se’ be classified as acting in
concert.
32
For example, in the Assicurazioni Generali BoSA election (mentioned in n. 24 above),
the asset managers which submitted the mutual funds’ list included an investment
company controlled by Unicredit (which was also part of the ‘control’ agreement in
Mediobanca, exactly like the holding of the Benetton family mentioned in n. 26 above;
the CEO of such investment company was actually an employee of the controlling bank)
and two investment companies controlled by Intesa SanPaolo (an important business
partner of Assicurazioni Generali). The stakes held by these asset managers, though
irrelevant for the final outcome, had been determinant to reach the shareholding
quorum needed to present a list.
33
This risk – present for all shareholders – is inherently more severe for asset managers,
who actively trade in the stock.
34
The number of nominees must be lower than half of the seats up for election, except
where the company by-laws dispose otherwise.
35
Nominees may not be executives of the associated asset managers (in order to minimise
insider trading risks) and must meet professional and independence standards in line
with best practice.
36
However, we found anecdotal evidence of cross-submission of candidates (e.g. invest-
ment companies controlled by Unicredit participated in the submission of board
nominees in Intesa SanPaolo, and vice versa).
37
Assogestioni recommends that any meeting between minority board members and ‘the
market’ (including institutional investors) take place in a formal and transparent
manner (i.e. not on a one-to-one basis). The aim of such meetings should be ‘to provide
minority board members with additional information’, also about the ‘opinions of the
market’: see ‘Principi sui rapport tra gli investitori istituzionali e gli amministratori
indipendenti e sindaci delle società quotate’, available (in Italian only) on the
Assogestioni website.
The robustness of such arrangements is, however, open to question. The

difference with shareholder agreements looks thin38 and the factors
which may justify a differential treatment for asset managers (i.e. that
they are not mere subsidiaries of their controlling banks) have not been
spelled out clearly.39 The viability of this model might not be guaranteed
without the ‘benign neglect’ of the market supervisor.
While mutual funds should conform to the general model (i.e. they
could use slate voting to monitor management subject to limitations
implied by transactions costs under the applicable law and, possibly,
their own conflicts of interest: Bianchi and Enriques 2001) an additional
hypothesis may be formulated about their activism: namely, that mutual
funds may follow a pension fund-style strategy, based on the recognition
of the limits of the Wall Street Rule, i.e. ‘voting with their feet’ in cases
where they are unhappy with management. According to a portfolio
composition hypothesis, mutual fund slates are more likely in blue-
chips and in other firms where mutual funds predominantly invest.
Previous literature has shown that institutional investors with strong
fiduciary responsibilities tend to invest in stocks that are viewed as
prudent investments (Del Guercio 1996; Parrino et al. 2003) and also
in companies with better governance quality (Chung and Zhang 2011).
Barucci and Falini (2005) and Bianchi et al. (2010) produce results for
Italian firms, compatible with this hypothesis. A number of factors may
reinforce this behavioural pattern. On one hand, institutional investors
may need to show that they take seriously their fiduciary duties as
delegated monitors on behalf of beneficial owners. This may lead them
38
According to art. 101-bis, subsection. 4-bis CLF, ‘[i]n any event, persons considered to
be acting in concert are: a) parties to an agreement, even if void, envisaged in article 122,
subsection 1 and subsection 5 paragraphs a), b), c) and d)’. Art.122, subsection 5 states
that the article shall also apply to ‘agreements, in whatsoever form concluded, that: a)
create obligations of consultation prior to the exercise of voting rights in companies with
listed shares or companies that control them’. The committee established by
Assogestioni effectively allows consultations regarding the nomination process (i.e.
the selection of candidates and the submission of slates) but not on the ‘exercise of
voting rights’ at the general meeting. The main difference seems that corporate govern-
ance committee decisions do not create an obligation for asset managers to vote for the
Assogestioni slate (i.e. an asset manager dissenting from the committee decision could
participate in the submission of the Assogestioni slate and then abstain or vote for a
different candidate at the AGM).
39
One possible factor may be the requirement (art. 40 para. 2 CLF) that asset managers
exercise the voting rights attached to the financial instruments belonging to the funds
under management ‘in the interests of the unit-holders’.
to concentrate on a few important cases where they can ‘show the flag’.40
On the other hand, just like US public pension funds, asset managers
(and their association) may be interested in maximising the political
returns from activism, arguably higher in large and/or politically rele-
vant companies (e.g. privatised firms).
To test our hypotheses we run a Logit regression model. More specif-
ically, we regress a variable capturing the probability that minority
shareholders (or alternatively, mutual funds) submit a slate (a 1/0
dummy variable based on the presence of such a list) on three different
sets of independent variables, also taking into account industry (defined
according to Fama and French (1993)) and year fixed effects. Our model
is the following:
Prob Multiple Listsi ¼ αi þ βij ½Characteristicsij þ βik ½Ownershipik
þ βil ½Votingsystemil þ εi ð1Þ
Where Characteristicsij, Ownershipik and Voting systemil are vectors of

variables capturing general characteristics of the i-th firm, its ownership
structure and the voting system in place, respectively, while εi is a
random error term. We used a number of specifications to check the
robustness of our estimates.
Endogeneity plagues much of the empirical literature in corporate
governance, particularly on the board of directors (Hermalin and
Weisbach 2003; Adams et al. 2009). However, our study appears less
affected by this problem, since we investigate the decisions of minority
investors, who had little direct influence over the firm governance
structure. Furthermore, we focus on the first election after a change in
legislation which created a remarkable discontinuity (only 28 per cent of
sample firms had slate voting in place before 2008 and previous election
rules were substantially different). However, we define firm character-
istics as lagged variables (i.e. they refer to the year – or the year-end –
before the election date), include appropriate control variables and
perform various robustness checks.
The first set of variables includes firm characteristics: age, size, finan-
cial structure, q ratio (a proxy for growth opportunities) and past
performance. In a second set of regressions, focusing on non-financial
40
The difference with the transaction costs hypothesis lies in the fact that asset managers
might choose to become active for reasons connected to their own agency relationship
with ultimate investors.
firms, we add a proxy for the risk of asset substitution (tangible assets/
total assets) and a proxy for the risk of over-investment (the cash/capex
ratio).41 Under the monitoring hypothesis, minority slates should be
more likely where the risk of conflicts of interest is higher, i.e. in younger
and smaller companies, in firms with higher growth opportunities, with
worse past performance and also where leverage is lower (since leverage
already limits the capacity of management to expropriate shareholders
or simply to waste money). In non-financial firms, the probability of
activism should be higher where tangible assets are smaller and cash/
capex is higher. Other hypotheses may have diverging implications.
Under the transactions costs hypothesis, minority slates should be
more likely in larger firms, since a significant part of such costs does
not vary with size. Under the portfolio composition hypothesis, mutual
fund activism is more likely in larger and better-performing firms and
where tangible assets are higher.42
The second set includes variables capturing different dimensions of
ownership structure: the identity of the ultimate shareholder, his cash-
flow and voting rights, the wedge between ownership and control and
other measures of ownership concentration proxying for the existence of
other relevant minority shareholders.43 We also include a dummy for the
presence of shareholder agreements. Under the monitoring hypothesis,
minority slates should be submitted more often where ownership con-
centration is lower (and voting and cash-flow rights are higher) and
where the wedge is higher. Under the regulation hypothesis, activism
should be more likely in State-owned firms (due to the long-run effects of
the privatisation law; this is also consistent with the portfolio composi-
tion hypothesis, since mutual funds should be more active in politically
rewarding cases) and less likely where a shareholders’ agreement is in
place (since the no-link provision limits the ability to submit a slate
41
Our choice was dictated by the fact that these variables are not available for financial
firms.
42
In other cases, two hypotheses may have converging implications: under both the
regulation and the portfolio composition hypotheses, activism should increase with
firm size (since quorum size is decreasing in firm size and investing in blue-chips is
preferable for prudent investors and also for institutions interested in political rewards
from activism). In the same vein, leverage and cash/capex have the same expected sign
under both the portfolio composition and the monitoring hypothesis: asset managers
may prefer investing in less financially-constrained firms.
43
Measured at the regulatory threshold for ownership transparency purposes (2 per cent),
which is often sufficient to submit a minority slate without coordinating with other
investors.
successfully).44 Furthermore, activism should be more likely where the

free-float is lower (a higher free-float allows small caps to increase the
quorum size). Under the transaction costs hypothesis, activism should
be more likely where minority shareholders have a larger investment in
the company, i.e. where the number of relevant shareholders is higher
(and where the free-float is lower). Finally, under the portfolio compo-
sition hypothesis, mutual fund activism is more likely where the number
of relevant shareholders is lower and where the free-float is higher (since
mutual funds hold a small stake, which is usually counted in the defi-
nition of free-float).
The third set includes voting variables: the number (and percentage)
of board seats reserved to minority nominees, a ‘new/old’ dummy taking
value one if slate voting was already in place before 2008, the share-
holding quorum required to submit a list and a dummy capturing the
type (quotas vs. proportional) of election rules. Under both the regula-
tion and the portfolio composition hypotheses, investors should be more
active where they have an opportunity to obtain more seats, where they
are already familiar with the voting system (where slate voting is ‘old’)
and where the quorum is lower. The quotas/proportional dummy has no
clear effect a priori. Voting variables (in particular, quorum size) are
correlated to some company characteristics, especially firm size
(ρquorum,size is −0.59). Consequently, we exclude voting variables from
a first set of regressions. In a second set of regressions, we add voting
variables and keep the concentration index as sole, comprehensive
ownership variable. We also avoid considering firm size and quorum
in the same specification.
5.3. Investor activism in board elections

Our results for investor activism in general are reported in Tables
8.4–8.6.
Table 8.4 shows the results when ownership structure is defined in
terms of concentration statistics.45 Firm characteristics do not seem to be
particularly relevant for investor activism, with the notable exception of
44
Consob includes ‘control’ shareholder agreements among the factors indicating a
material link between shareholders presenting or voting different slates.
45
Models 4 and 8 exclude ownership variables altogether in order to have a benchmark for
measuring their additional explanatory power. We have also run an alternative specifi-
cation, using past accounting performance with qualitatively equivalent results.
Table 8.4 Determinants of the decision to submit a ‘minority’ slate (ownership defined in terms of concentration)
Dependent variable = Minority Slate dummy

Whole sample Non-financial companies
(1) (2) (3) (4) (5) (6) (7) (8)
VR −3.3003** −3.4546*
[1.3771] [1.9330]
CFR −2.3315*** −2.7434***
[0.7075] [1.0156]
Wedge 1.3388 −1.1728 1.319 −2.091
[2.1996] [2.1144] [2.6849] [2.6684]
No. of shareholders>2% 0.2210*** 0.2775**
[0.0821] [0.1105]
Shareholders’ agreement −0.599 −0.7675*
[0.3678] [0.4632]
Free Float −2.5775* −3.9807*
[1.5052] [2.2280]
Concentration Index −0.0003*** −0.0004***
[0.0001] [0.0001]
Log age 0.5504 0.3951 0.4061 0.228 −0.09 0.0544 0.0668 −0.0825
[0.3455] [0.3221] [0.3248] [0.3092] [0.4496] [0.4243] [0.4252] [0.4067]
Log assets 0.5138** 0.3066 0.3600* 0.4270** 0.7892** 0.5967** 0.6791** 0.6970
[0.2207] [0.1949] [0.1928] [0.1864] [0.3094] [0.2793] [0.2711] [0.2638]
Leverage −0.2962 −0.3144 −0.3319 −0.3754 −2.8801** −2.8320** −2.8555** −2.8058**
[0.8539] [0.7972] [0.8011] [0.7780] [1.2846] [1.1725] [1.1839] [1.1477]
6-month stock returns −0.8633 −0.789 −0.7324 −0.8852 −1.1671 −1.2865* −1.2487 −1.2785*
[0.6244] [0.5810] [0.5882] [0.5704] [0.8398] [0.7636] [0.7885] [0.7471]
q ratio −0.0497 −0.0195 −0.0081 0.024 −0.1088 −0.0231 −0.015 0.0298
[0.1043] [0.0906] [0.0895] [0.0889] [0.1258] [0.1086] [0.1076] [0.1062]
Tangible assets (%) 1.8217*** 1.6103*** 1.6003*** 1.4648***
[0.5396] [0.5055] [0.5036] [0.4760]
Cash/Capex 0.0045 0.004 0.0041 0.0037
[0.0033] [0.0028] [0.0027] [0.0030]
Intercept −1.4283 −0.787 −1.245 −2.3876** −2.2494 −2.0884 −2.7138* −3.9359**
[1.7295] [1.2177] [1.1714] [1.1016] [2.4105] [1.7449] [1.6357] [1.5585]
Industry dummies Yes Yes Yes Yes Yes Yes Yes Yes
Calendar year dummies Yes Yes Yes Yes Yes Yes Yes Yes
Pseudo R-squared 0.1879 0.1225 0.1408 0.0892 0.2762 0.2081 0.2284 0.179
No. of Observations 259 260 260 260 201 201 201 201
Note: Results of a Logit regression analysis: whole sample, period 2008–10. Dependent variable: a dummy taking value 1 when at least
one minority slate was submitted for board elections, and 0 otherwise; standard errors are reported in square brackets. All other
variables are defined in the Appendix. Industry dummies and year dummies are included in all regressions. *, **, and *** denote
statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 8.5 Determinants of the decision to submit a ‘minority’ slate (ownership defined in terms of ultimate shareholder identity)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
Family −1.1588*** −1.2182*** −1.1181** −1.0681***

[0.4225] [0.3249] [0.5020] [0.3968]
State 1.0249 2.0298** 0.5901 1.5951*
[0.9065] [0.8302] [0.9636] [0.8505]
Financial institution −1.728 −1.2439
[1.3088] [1.2486]
Private Equity 0.0153 0.8481 −0.3886 0.4505
[0.8018] [0.7210] [0.9208] [0.8308]
Widely held −0.4914 0.2642 −1.7699 −0.9488
[0.6957] [0.5924] [1.5487] [1.4422]
Log age 0.3369 0.3441 0.3317 0.2246 0.2287 0.2287 0.0903 0.0398 0.0779 −0.09 −0.0919
[0.3300] [0.3169] [0.3191] [0.3115] [0.3087] [0.3094] [0.4301] [0.4186] [0.4230] [0.4070] [0.4077]
Log assets 0.3152 0.2812 0.3572* 0.4516** 0.4365** 0.4056** 0.5698** 0.5796** 0.5934** 0.7001*** 0.7136***
[0.2070] [0.1957] [0.1900] [0.1880] [0.1872] [0.1922] [0.2788] [0.2724] [0.2710] [0.2642] [0.2657]
Leverage −0.2821 −0.1486 −0.2092 −0.4643 −0.3904 −0.3572 −2.4294** −2.4174** −2.4905** −2.7915** −2.8617**
[0.8101] [0.8006] [0.7827] [0.7822] [0.7805] [0.7789] [1.1749] [1.1658] [1.1569] [1.1506] [1.1519]
6-month stock returns −1.0163 −0.7714 −0.8974 −0.8977 −0.861 −0.8664 −1.201 −1.2144 −1.2504* −1.2784* −1.2806*
[0.6215] [0.5795] [0.5747] [0.5707] [0.5728] [0.5719] [0.7465] [0.7479] [0.7457] [0.7475] [0.7477]
q ratio −0.0346 −0.0233 −0.0004 0.0254 −0.0061 0.0236 0.0156 0.006 0.0093 0.0221 0.0355
[0.0968] [0.0918] [0.0918] [0.0889] [0.0929] [0.0890] [0.1066] [0.1039] [0.1088] [0.1051] [0.1057]
Tangible assets (%) 1.2935*** 1.3492*** 1.2795*** 1.4682*** 1.4762***
[0.4841] [0.4761] [0.4765] [0.4766] [0.4793]
Cash/Capex 0.004 0.0041 0.0036 0.0037 0.0037
[0.0030] [0.0030] [0.0029] [0.0030] [0.0030]
Intercept −1.0882 −0.8606 −2.3818** −2.4868** −2.4481** −2.2819** −2.6833 −2.6488 −3.7511** −3.9654** −4.0331**
[1.2574] [1.2046] [1.1162] [1.1067] [1.1089] [1.1245] [1.6990] [1.6646] [1.5750] [1.5629] [1.5682]
Industry dummies Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Calendar year dummies Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Pseudo R-squared 0.1416 0.1311 0.1105 0.0924 0.0931 0.0897 0.2155 0.2066 0.1942 0.1801 0.1807
No. of Observations 258 260 260 260 260 260 201 201 201 201 201
Note: Results of a Logit regression analysis: whole sample, period 2008–10. Dependent variable: a dummy taking value 1 when at least one minority slate was
submitted for board elections, and 0 otherwise; standard errors are reported in square brackets. All other variables are defined in the Appendix. Industry
dummies and year dummies are included in all regressions. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 8.6 Determinants of the decision to submit a ‘minority’ slate (ownership concentration and voting rules)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
% board seats 4.2394** 4.6250* 4.5073 5.2935*

reserved
[2.1415] [2.0921] [3.0402] [2.9877]
No. of board seats 0.3176* 0.2841
reserved
[0.1926] [0.2910]
old/new 0.3423 0.4347 0.5184 0.5844
[0.3683] [0.3433] [0.4334] [0.4194]
Quorum −22.8128 2.0276
[17.8167] [23.6179]
Quotas/ 0.1797 0.1081 0.4665 0.442
proportional
[0.5100] [0.4816] [0.6320] [0.6122]
Concentration −0.0003*** −0.0003*** −0.0003*** −0.0003*** −0.0004*** −0.0003*** −0.0004*** −0.0004*** −0.0004*** −0.0004*** −0.0004*** −0.0004***
Index
[0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001] [0.0001]
Log age 0.5098 0.4691 0.3927 0.461 0.4858 0.4009 0.1924 0.1569 0.126 0.1617 0.229 0.0419
[0.3389] [0.3337] [0.3264] [0.3285] [0.3194] [0.3256] [0.4388] [0.4325] [0.4304] [0.4344] [0.4125] [0.4254]
Log assets 0.3569* 0.3912* 0.2859 0.3189 0.3610* 0.6748** 0.7238*** 0.5844** 0.6135** 0.6859**
[0.2052] [0.2007] [0.1978] [0.1964] [0.1927] [0.2835] [0.2780] [0.2864] [0.2765] [0.2715]
Leverage −0.6661 −0.5732 −0.4478 −0.4842 −0.2184 −0.3341 −3.2111*** −3.0324** −2.9099** −3.0268** −2.0746* −2.9170**
[0.8289] [0.8212] [0.8106] [0.8142] [0.7965] [0.8011] [1.2070] [1.1922] [1.1850] [1.1973] [1.1022] [1.1891]
6-month stock −0.7251 −0.678 −0.6646 −0.7948 −0.6162 −0.7287 −1.3657* −1.3047 −1.2445 −1.3250* −1.0587 −1.2444
returns
[0.5939] [0.5932] [0.5917] [0.5891] [0.5822] [0.5886] [0.7941] [0.7977] [0.7924] [0.7858] [0.7642] [0.7874]
q ratio −0.023 −0.0196 −0.0145 −0.0083 −0.0148 −0.0109 −0.046 −0.0305 −0.0183 −0.0146 −0.0407 −0.0328
[0.0911] [0.0896] [0.0899] [0.0900] [0.0911] [0.0904] [0.1131] [0.1065] [0.1079] [0.1090] [0.1051] [0.1125]
Tangible 1.5731*** 1.5945*** 1.5610*** 1.5405*** 1.6222*** 1.6299***
assets (%)
[0.5130] [0.5062] [0.5032] [0.5053] [0.5094] [0.5084]
Cash/Capex 0.0037 0.0037 0.0039 0.0041 0.0036 0.004
[0.0026] [0.0025] [0.0026] [0.0027] [0.0024] [0.0027]
Intercept −2.1166 −2.0845 −1.1765 −1.162 1.1772 −1.334 −3.8572** −3.7487** −2.635 −2.5544 0.568 −3.0691*
[1.3538] [1.2827] [1.1901] [1.1818] [0.8664] [1.2367] [1.8393] [1.7628] [1.6434] [1.6540] [1.1642] [1.7058]
Industry Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
dummies
Calendar year Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
dummies
Pseudo 0.1573 0.1548 0.1493 0.1453 0.1353 0.1409 0.2472 0.2408 0.232 0.2356 0.2039 0.2304
R-squared
No. of 259 259 260 260 260 260 201 201 201 201 201 201
Observations
Note: Results of a Logit regression analysis: whole sample, period 2008–2010. Dependent variable: a dummy taking value 1 when at least one minority slate was submitted for
board elections, and 0 otherwise; standard errors are reported in square brackets. All other variables are defined in the Appendix. Industry dummies and year dummies are
included in all regressions. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
firm size: activism is more likely in larger firms. On the contrary,

ownership structure has a strong impact. As expected, minority slates
are associated with lower ownership concentration in virtually any
specification. Minority slates are more likely where the ultimate share-
holder holds a lower stake, where the number of other relevant share-
holders is higher, where the free-float is lower, where no shareholders’
agreement is present and, finally, where the ownership concentration
index is lower. Wedge is the only ownership variable with no clear
impact on activism.
Taken together, our evidence is fully consistent with the transaction
costs hypothesis, while it is clearly at odds with the monitoring hypoth-
esis. Activism is more likely where minorities hold larger stakes.
Minority slates are submitted by ‘relevant’ shareholders which face
comparatively lower transaction costs. The risk of conflicts of interest
is substantially irrelevant.
We find only weak support for the regulation hypothesis: activism is
more likely in larger firms. This effect might be induced by the regulatory
caps to quorum size (decreasing in firm size), designed by Consob to
favour activism.46 At the same time, our evidence seems to contradict the
argument put forward by Consob to allow a quorum increase only where
the free-float is high (according to which a low free-float makes it
difficult to submit a minority slate). Our evidence might also be seen
as substantially consistent with the portfolio composition hypothesis,
insofar as the results are driven – at least partially – by the behaviour of
institutional investors: this point will be examined further below.
The results for non-financial firms are substantially in line with those
for the whole sample. Here we find some evidence consistent with the
monitoring hypothesis: leverage impacts negatively on activism and the
negative effect of past performance becomes marginally significant.
However, tangible assets have the ‘wrong’ sign (and the relation is
statistically significant): the submission of minority slates is more likely
where the risk of asset substitution is lower. This seems inconsistent with
46
Consob explained its regulatory intervention as striking a balance between two com-
peting goals, namely: (a) guarantee that minority shareholders have a real opportunity to
appoint their own representatives, as mandated by the law; (b) avoid greenmailing in
smaller firms, where buying the stake required to submit a list is cheap. This led the
market supervisor to set caps inversely proportional to market capitalisation and to free
small companies to set a higher quorum, conditional on having a sufficient free-float and
no majority blockholder (see the Consultation Document issued on 23 February 2007,
available on the Consob website).
a monitoring hypothesis (while it could be compatible with a portfolio

composition story).
Overall, our regressions show a reasonably good explanatory power,
especially where ownership variables are included in the model.
Table 8.5 reports the results when ownership is defined in terms of the
ultimate shareholder’s identity.
The identity of the ultimate shareholder is less relevant than owner-
ship concentration.47 Minority slates are less likely in family firms
(where ownership concentration is also higher) and more likely in
state-owned companies. This may be consistent with a continuing influ-
ence of the privatisation law and possibly with a portfolio composition
story. However, when all categories are considered in the same regres-
sion, the state dummy is not statistically significant: ownership concen-
tration is apparently the most influential variable. Table 8.6, reporting
results including voting variables, offers additional insights.
Consistent with the regulation hypothesis, activism is positively asso-
ciated with the percentage (and, marginally, with the number) of board
seats reserved to minority nominees. A higher quota increases both the
probability that at least one candidate will be appointed and the expected
influence of minority directors over the board. Interestingly, the per-
centage weight of the quota seems more relevant than the sheer number
of seats. The new/old dummy has the expected positive sign, but its effect
is weak. Activism seems to be equally likely in companies adopting a
quota/proportional voting system.
The shareholding quorum required to submit a slate has drawn a lot of
attention in the regulatory debate. However, its importance seems
grossly overstated: activism in board elections shows no association
with quorum size. This is even more striking since quorum size is
strongly and negatively correlated with firm size (which has a clear,
positive influence on activism). Surprising as it may seem, this result is
consistent with the data in Table 8.3: since quorum size is, on average,
one-quarter of the stake held by the second-largest shareholder, the
decision of this subject (the ideal candidate to seek board representation)
is hardly influenced by the quorum level. This, in turn, is consistent with
the transaction costs hypothesis.
47
Ultimate shareholder dummies and concentration proxies were not used in the same
regression to avoid collinearity. The Financial Institution dummy was dropped in the
right-hand panel (only two non-financial firms are controlled by a financial institution).
In general, substituting ownership with voting variables causes a

relevant drop in pseudo-R2 values. We interpret this result as a clear
symptom of the limits of the regulation hypothesis. Investor activism is
driven by firm characteristics and, in particular, by ownership structure
variables affecting transaction costs, while it is relatively unaffected by
regulation (in particular, by the technical details of the voting system).
5.4. Mutual fund activism in board elections

Analysing the activism of mutual funds may help to disentangle the
influence of firm characteristics and investor portfolio decisions.
Institutional investors choosing prudent man investments could follow
different strategies in corporate elections. This might, in turn, be linked
to the desire to reap political rewards. To test these hypotheses, we re-
run our regressions using as dependent variable a dummy taking the
value 1 where a minority slate was submitted by mutual funds coordin-
ated by Assogestioni;48 the methodology is otherwise unchanged. Our
results are reported in Tables 8.7–8.9.
The results, at first sight, may look similar to those for activism in
general. However, the magnitude of the size coefficient is striking (four
times the value it had in previous regressions). Furthermore, the explan-
atory power of even the poorest models in Table 8.7 is much higher than
that of our best-fitting regressions run on activism in general. Mutual
funds’ activism seems to be driven almost only by firm size.49
Ownership variables show a much weaker association with mutual
fund activism than with activism in general (see Tables 8.7 and 8.8).
Institutional investors tend to become active in State-owned, non-
financial companies where ownership concentration is lower. The
wedge is utterly irrelevant.
Mutual funds’ slates are more likely where leverage is lower. However,
the relation is weak (and is statistically significant only for the whole
sample).50 Furthermore, they are more likely where cash/capex (a proxy
48
Such coordination role was, usually, declared by the funds’ proxy-holders and is also
reported on the Assogestioni website.
49
Sample size is smaller since entire industries, where mutual funds have never submitted
a list, had to be dropped. The same is true for companies controlled by private equity
investors.
50
This result is probably driven by a weaker presence of mutual funds in AGMs of financial
firms, which might be associated with a desire of asset managers to avoid blatant cases of
conflict of interest.
Table 8.7 Determinants of the decision to submit a ‘mutual fund’ slate (ownership defined in terms of concentration)
Dependent variable = Mutual Fund Slate dummy

(1) (2) (3) (4) (5) (6) (7) (8)
VR 2.7197 2.9137
[2.9318] [3.5112]
CFR 0.6127 1.8847
[1.4712] [2.0694]
Wedge 3.3187 2.2923 6.0285 6.9531
[3.5141] [3.5308] [4.3731] [4.7825]
No. of shareholders>2% 0.2795* 0.1829
[0.1671] [0.2211]
Shareholders’ agreement −0.3385 0.2232
[0.7857] [1.0397]
Free Float 1.16 −0.0991
[3.2226] [4.0595]
Concentration Index −0.0001 −0.0001
[0.0002] [0.0003]
Log age 0.4774 0.7473 0.8708 0.8583 0.0806 0.3217 0.5925 0.5935
[0.7090] [0.6687] [0.6556] [0.6548] [0.8725] [0.7871] [0.7840] [0.7788]
Log size 2.2868*** 2.2884*** 2.1984*** 2.2575*** 2.5811*** 2.3808*** 2.2960*** 2.3555***
[0.5475] [0.5218] [0.5028] [0.4892] [0.7852] [0.6895] [0.6314] [0.6240]
Leverage −3.8550* −3.5123* −3.7912* −3.6941* −0.4277 −0.2717 −0.9664 −0.8098
[2.1004] [1.9776] [2.0203] [1.9613] [2.9392] [2.7111] [2.6078] [2.5277]
Table 8.7 (cont.)

(1) (2) (3) (4) (5) (6) (7) (8)
6-month stock returns 0.9435 1.1837 1.4053 1.3292
1.1178 1.3584 1.5179 1.5281
[1.3687] [1.2795] [1.2678] [1.2669]
[1.5673] [1.4514] [1.4725] [1.4565]
q ratio −0.1595 −0.1373 −0.1493 −0.1425
0.018 −0.1192 −0.1864 −0.1785
[0.2193] [0.1938] [0.1849] [0.1857]
[0.6312] [0.4121] [0.2936] [0.2986]
Tangible assets (%) 0.4224 0.3211 0.4004 0.4644
[0.9472] [0.8661] [0.8360] [0.8249]
Cash/Capex 0.0065** 0.0058** 0.0059** 0.0058**
[0.0032] [0.0029] [0.0029] [0.0029]
Intercept −19.6505*** −17.8342*** −16.7110*** −17.3828*** −23.5414*** −21.3068*** −18.9798*** −19.8813***
[4.4157] [3.9736] [3.7418] [3.4596] [6.8150] [5.7496] [5.1439] [4.7144]
Pseudo R-squared 0.4021 0.3807 0.3807 0.3776 0.4637 0.4517 0.4359 0.4324
Note: Results of a Logit regression analysis: whole sample, period 2008–2010. Dependent variable: a dummy taking value 1 when at least one
minority slate was submitted by mutual funds for board elections, and 0 otherwise; standard errors are reported in square brackets. All other
variables are defined in the Appendix. Industry dummies and year dummies are included in all regressions. *, **, and *** denote statistical
significance at the 10%, 5% and 1% levels, respectively.
Table 8.8 Determinants of the decision to submit a ‘mutual fund’ slate (ownership defined in terms of ultimate
shareholder identity)

(1) (2) (3) (4) (5) (6) (7) (8)
Family 0.3237 0.2348 −0.1805 −0.0874
[1.0680] [0.7004] [1.2151] [0.9167]
State 2.734 2.4942 3.64 3.9941*
[1.8971] [1.7504] [2.5934] [2.3538]
Financial institution −0.2514 0.1738
[2.1813] [1.9146]
Private Equity
Widely held −0.8452 −0.9354
[1.4626] [1.1092]
Log age 1.0549 0.8559 0.9612 0.8515 0.8287 0.6322 0.6098 0.9612
[0.6917] [0.6557] [0.6630] [0.6592] [0.6564] [0.9049] [0.7988] [0.8316]
Log assets 2.2953*** 2.2904*** 2.1943*** 2.2538*** 2.3501*** 2.6417*** 2.3483*** 2.2996***
[0.5174] [0.5005] [0.4865] [0.4908] [0.5110] [0.7305] [0.6283] [0.6246]
Leverage −3.2612 −3.7284* −3.4761* −3.6739* −3.5095* −1.6415 −0.7771 0.0594
[2.0065] [1.9677] [1.9815] [1.9765] [1.9664] [2.8316] [2.5408] [2.5977]
6-month stock returns 1.2889 1.3043 1.2359 1.3241 1.2719 1.9149 1.5244 1.4657
[1.3552] [1.2753] [1.2830] [1.2683] [1.2536] [1.5356] [1.4579] [1.5026]
q ratio −0.1784 −0.134 −0.19 −0.1427 −0.14 −0.2637 −0.1846 −0.3301
[0.2001] [0.1897] [0.1870] [0.1858] [0.1962] [0.3892] [0.3007] [0.2873]
Table 8.8 (cont.)

(1) (2) (3) (4) (5) (6) (7) (8)
Tangible assets (%) 0.3272 0.4583 0.1221
[0.9285] [0.8296] [0.9125]
Cash/Capex 0.0060** 0.0058** 0.0064**
[0.0028] [0.0029] [0.0029]
Intercept −18.8162*** −17.7487*** −17.6829*** −17.3523*** −17.9983*** −22.9678*** −19.8006*** −21.7382***
[3.8873] [3.6456] [3.5115] [3.4726] [3.6257] [5.8603] [4.7890] [5.4430]
Pseudo R-squared 0.4005 0.3784 0.3939 0.3777 0.3827 0.4981 0.4324 0.4677
Note: Results of a Logit regression analysis: whole sample, period 2008–2010. Dependent variable: a Dummy taking value 1 when at least one
significance at the 10%, 5%, and 1% levels, respectively.
Table 8.9 Determinants of the decision to submit a ‘mutual fund’ slate (ownership concentration and voting rules)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
% board seats 1.0859 3.0153 4.8285 5.2564

reserved
[3.5415] [3.2300] [5.1349] [4.5861]
No. of board seats 0.0901 0.114
reserved
[0.2877] [0.4077]
Old/new 1.0424 0.6692 1.3498 1.1564
[0.8340] [0.7144] [1.1089] [0.9434]
Quorum −79.1053** −68.685
[37.4897] [44.6354]
Quotas/proportional 2.8405* 2.2451
[1.5922] [1.4457]
Concentration index −0.0001 −0.0001 −0.0001 −0.0001 −0.0003* −0.0001 −0.0001 0 −0.0001 0 −0.0003
[0.0002] [0.0002] [0.0002] [0.0002] [0.0002] [0.0002] [0.0003] [0.0002] [0.0003] [0.0003] [0.0002]
Log age 0.7868 0.8669 0.886 0.9444 1.0265* 0.7457 0.6918 0.8279 0.6591 0.7899 0.632
[0.7120] [0.6713] [0.6576] [0.6640] [0.5979] [0.6768] [0.9481] [0.8162] [0.8193] [0.8248] [0.6797]
Log assets 2.3652*** 2.2036*** 2.1713*** 2.1594*** 2.3532*** 2.7145*** 2.3534*** 2.2696*** 2.2540***
[0.5469] [0.5065] [0.5114] [0.5041] [0.5344] [0.7756] [0.6496] [0.6420] [0.6289]
Leverage −5.0895** −4.0177* −3.8318* −4.0835* −1.9843 −4.3392** −4.5203 −1.3417 −0.9895 −1.1978 0.249
[2.2303] [2.0711] [2.0318] [2.0967] [1.6313] [2.0465] [3.2489] [2.5745] [2.5935] [2.6731] [1.9951]
6-month stock 1.4286 1.2536 1.4057 1.2195 1.6495 1.6362 1.7409 1.317 1.5417 1.1576 1.7029
returns
[1.3662] [1.2953] [1.2733] [1.2805] [1.1620] [1.3227] [1.6483] [1.4755] [1.4760] [1.5724] [1.4306]
Table 8.9 (cont.)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
q ratio −0.3726 −0.1663 −0.1515 −0.1608 −0.1053 −0.3018 −0.7448 −0.2255 −0.1898 −0.2031 −0.1528
[0.2350] [0.1824] [0.1843] [0.1857] [0.1594] [0.2227] [0.8025] [0.2733] [0.2908] [0.3071] [0.1964]
Tangible assets (%) −0.112 0.3273 0.3736 0.1319 0.0305
[0.9612] [0.8276] [0.8378] [0.8846] [0.7540]
Cash/Capex 0.0050* 0.0055** 0.0058** 0.0064** 0.0040*
[0.0030] [0.0027] [0.0029] [0.0031] [0.0021]
Intercept −20.3506*** −17.1527*** −16.7326*** −16.7629*** −1.7062 −19.2604*** −20.9870*** −20.8129*** −19.2258*** −19.4591*** −2.7416
[4.5544] [3.8411] [3.7623] [3.7799] [1.6790] [4.2624] [6.6905] [5.5183] [5.2919] [5.2637] [2.2549]
Industry dummies Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Calendar year Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
dummies
Pseudo R-squared 0.4207 0.3868 0.3813 0.3866 0.2183 0.4041 0.5041 0.4468 0.4366 0.4494 0.2758
No. of Observations 218 218 219 219 219 219 136 152 152 152 152
Note: Results of a Logit regression analysis: whole sample, period 2008–2010. Dependent variable: a dummy taking value 1 when at least one
significance at the 10%, 5%, and 1% levels, respectively.
for the risk of over-investment, but also a variable connected with a

prudent man portfolio) is higher. Asset composition (a proxy for the risk
of asset substitution) has no impact on mutual funds’ decisions.
Taken together, these results offer little support for the monitoring
hypothesis (the signs of leverage and cash/capex coefficients are correct,
while those of ownership structure and asset composition are not). On
the contrary, our evidence is totally consistent with the portfolio com-
position hypothesis. Mutual funds concentrate their efforts in a small
number of blue-chips which may be seen as prudent man investments
and offer potentially higher political rewards. Mutual funds’ choice of
prudent man investments and privatised companies as targets for board
election activism clearly shows that transaction costs are not the whole
story.
Further elements are offered by the market return coefficient, which is
positive in mutual fund regressions (while it was negative and marginally
significant in the previous analysis). Mutual funds unhappy with the
returns on a particular stock may still vote ‘with their feet’ (i.e. under-
weight the stock in their portfolio) and avoid activism.51 This result is at
odds with mutual funds exerting a monitoring role (while choosing their
targets on the basis of a proper cost–benefit analysis). On the opposite, it
is clearly consistent with institutional investors becoming active in a
small number of cases, carefully chosen from a prudent man portfolio.
An intriguing implication is that mutual funds tend to become active
where monitoring is less needed.
Table 8.9 reports results when voting variables are included in the
analysis. They are not particularly relevant for mutual fund activism,
with the notable exception of quorum size, which is negatively associated
with the submission of mutual funds’ slates.52
This may explain why the quorum issue received so much attention in
the regulatory debate: although quorum size is not important for rele-
vant shareholders, it may impact on mutual funds’ behaviour. Mutual
funds holding a tiny stake may face higher transaction costs (and choose
51
Alternatively, this result could be plagued by endogeneity: better managed companies
could be those where mutual funds exerted more pressure in the past. According to this
hypothesis, the result would simply be driven by previous activism. However, this seems
not to be the case, since the old/new dummy, capturing possible long-term effects of
activism, is not statistically significant.
52
However, the coefficient is statistically significant only in the regression on the whole
sample.
to stay passive) where the quorum is high. On the other hand, quorum
and firm size are correlated. It is easy to see that including quorum size in
the model (while excluding firm size to avoid collinearity) causes a
relevant drop in the explanatory power (compare model (3) in
Table 8.7 with model (5) in Table 8.9). A similar drop did not take
place in our analysis of activism in general (compare model (3) in
Table 8.4 with model (5) in Table 8.6). It seems safe to conclude that
mutual funds’ activism is less influenced by quorum than by firm size.
This result is, once again, consistent with the transaction costs and with
the portfolio composition hypothesis (but not necessarily with the regu-
lation hypothesis).
5.5. Robustness checks

We performed two robustness checks. First, we considered elections to
the Board of Statutory Auditors. BoSA elections have a number of
features which may, in principle, lead to different results. BoSA and
BoD perform different functions, possibly creating different incentives
to become active. BoSA and BoD elections present some technical differ-
ences, possibly affecting the results.53 Investors might be more familiar
with slate voting in BoSA elections (mandated since 1998), even though
foreign investors might be less familiar with a corporate body having
few equivalents in other jurisdictions. After dropping companies
with incomplete and/or missing data, we end up with a sample of 260
elections. The results (not reported) are in line with our previous
analysis.
Second, we extended our analysis to 2011 elections, to extract infor-
mation on activism after the implementation of the Shareholders’ Rights
Directive (which introduced the record date system, possibly reducing
transactions costs, especially for mutual funds). A caveat is necessary:
since Italian boards are typically classified, we could analyse only a
subsample of 73 BoD (74 BoSA) elections. Consequently, the test has
only limited explanatory power. Once more, the results (not reported)
are substantially in line with our previous analysis.
53
BoSA size and minority quotas are substantially constant (three seats, one of which is
reserved to the most voted minority nominee; since 2008 he is automatically appointed
as chairman). If only one list has been submitted, further lists may be submitted up to the
fifth working day after the original expiry date and the thresholds established in the by-
laws shall be halved.
6. Conclusions and policy implications

Board elections are a key issue in corporate governance and have come
into the spotlight after the financial crisis. In the US, where shareholders’
influence over board elections is – apparently – at a historical minimum,
a number of regulatory proposals have been put forward to increase the
role of shareholders. In Europe, the adoption of multiple-winner voting
rules granting board representation to minority shareholders is consid-
ered by the EU Commission to support the alignment of managerial
incentives, particularly in companies with a controlling shareholder. An
analysis of previous experiences may provide useful insights into the
possible effects of such measures.
Italian regulation offers a unique opportunity in this regard: listed
companies are required to reserve at least one board seat to minority
nominees, and even investors holding a small block of shares have a real
opportunity to gain board representation. We analysed the minority
shareholders’ decision to submit a slate of candidates in Italian board
elections. Our main conclusions may be summarised as follows.
First, activism in corporate elections is overwhelmingly influenced by
firm ownership structure. In particular, it is associated positively with
the number of relevant shareholders (apart from the controlling block-
holder) and negatively with the stake held by the ultimate shareholder,
with the presence of shareholder agreements and with the company free-
float. The wedge between cash-flow and voting rights is irrelevant.
Minority slates are presented more (less) frequently in state-owned
(family) firms. However, the identity of the ultimate shareholder seems
less important than ownership concentration.
Second, activism depends on some firm characteristics. In particular,
it is associated positively with firm size. In non-financial firms it is also
positively associated with tangible assets and negatively with leverage.
Third, activism is relatively insensitive to voting rules. Only the
proportion of board seats reserved to minority nominees seems to have
an influence on investor decisions. The novelty of the voting system, its
type (quotas vs. proportional) and quorum size appear to have little
impact. The results are, in turn, influenced by ownership structure.
Activism is only slightly affected by quorum size, since this is, on
average, one-quarter of the stake held by the second-largest shareholder
(the ideal candidate to seek board representation).
Fourth, mutual funds tend to concentrate their efforts in a small
number of blue chips and in companies which may be seen as prudent
man investments (with a better past performance and lower financial

constraints, as proxied by leverage and cash/capex). Their decisions are
not influenced by ownership structure and voting variables, with the
notable exception of quorum size.
Taken together, our evidence provides little support for the hypothesis
that investors (in particular, mutual funds) become active to monitor
management where the risk of conflicts of interest is higher. Moreover,
investors are not sensitive to the technicalities of the voting system. Even
the results for quorum size (a key issue in the political debate) seem to be
driven mostly by firm size. On the other hand, our results are strongly
consistent with the hypotheses that investors consider transaction costs
in their decisions concerning activism, that mutual funds’ decisions are
affected by their portfolio composition and that they may be sensitive to
the political returns from activism. This may explain why they concen-
trate their efforts on large and/or politically relevant firms, despite the
lower need for monitoring in such companies.
Our analysis has a number of implications for the policy debate on
corporate elections, ranging from the conditions affecting activism to the
identity of shareholders who are more likely to become active and the
effectiveness of voting regulation as a tool for improving minority
shareholders’ protection. First, a multiple-winner voting rule spurs acti-
vism. Minority shareholders used the opportunity to submit a slate of
nominees in around 40 per cent of our sample firms. However, in a
majority of cases they chose to remain passive. Apathy was not due to a
high quorum size: almost every company had one or more relevant
shareholders who could easily get board representation. Even though
such shareholders may sometimes privately negotiate the inclusion of
their nominees in the majority slate, quorum size seems low and should
not create insurmountable obstacles to activism. A more reasonable
explanation for the 60 per cent of cases where minority quotas are
apparently neglected is simply that board representation might not be
a cost-effective way to monitor management. It could also imply other
regulatory deficiencies. An investor-friendly voting regulation may be
insufficient to overcome rational apathy. Decisions in this field are best
left to individual investors, who are in the position to fully appreciate
costs and benefits of alternative strategies.
Second, minority shareholders’ incentives depend significantly on
firms’ characteristics (in particular, on firm size, ownership concentra-
tion and, to a lesser extent, the identity of the ultimate shareholder).
Policy proposals should consider that the effects of regulation may be
different, depending on such characteristics. This point seems particularly

relevant at the EU level, since regulatory measures may need to consider
industrial and ownership structures which are quite diverse across Member
States. One size may not fit all in corporate elections either.
Third, once a multiple-winner system is in place, activism is relatively
insensitive to the specific voting rules adopted at the company level.
Consequently, tinkering with voting technicalities (e.g. fine-tuning
quorum size in relation to firm characteristics and/or ownership struc-
ture, or to market capitalisation, in either bullish or bearish markets) is
unlikely to affect investors’ incentives in a significant manner.
Fourth, the behaviour of institutional investors has a number of
peculiar features: their incentives to become active seem to be driven
by portfolio composition and, possibly, by the political returns expected
from such activity. Furthermore, institutional investors wishing to
become active in corporate elections face regulatory problems which
are not easily overcome. A favourable stance taken by the supervisory
authorities seems to be a necessary (although not a sufficient) pre-
condition for mutual funds’ activism.
Overall, our results show that the incentives for minority investors to
become active are rooted in firm characteristics (primarily ownership
concentration) and are relatively insensitive to regulatory measures. An
interesting corollary is that the identity of investors who choose to
become active also varies with firm characteristics: Italian mutual
funds submit their candidates in a small number of blue-chips, while
minority lists in family firms are submitted by individual shareholders
holding a relevant stake (on average, 9 per cent of share capital).
A second corollary is that granting board representation to minorities
will not, per se, increase monitoring or investor self-protection.
Transaction costs are a serious issue. Investors holding very small stakes
will suffer from conflicts of interest, which are at least as serious as those
faced by controlling blockholders. Institutional investors may face seri-
ous regulatory obstacles. Other active shareholders may have an incen-
tive to collude with the controlling blockholder and serve as an
entrenchment device. Further tests are needed to analyse these issues.
Finally, our results cast doubt on the opportunity to recommend (or
even to mandate) a multiple-winner voting system at the EU level.
Transaction costs implied by activism are substantial, while benefits
from additional monitoring are, at best, uncertain. The incentives to
become active depend on characteristics of the institutional context (e.g.
firm ownership structure, other regulatory issues, the discretionary
position of supervisory authorities, etc.) which may vary across Member

States. No clear case for an intervention at the EU level can be made.
Decisions in this area are best left to individual firms or, at most, to
national legislators.
APPENDIX
DEFINITION OF VARIABLES
age Number of years since foundation

age from listing Number of years since the company was first listed on the
Italian Stock Exchange
log age Natural logarithm of age
total assets Total assets (book value)
tangible assets (%) Tangible assets/total assets
cash Cash and cash equivalents (book value)
capex Capital expenditures
log assets Natural logarithm of total assets
equity Book value of equity
mktcap Market value of equity
q ratio (Total assets – equity + mktcap)/ Total assets
EBITDA/sales (Earnings before interest, taxes, depreciation, and
amortisation)/Total sales
leverage Financial debt/Total assets
ROA Earnings from total assets/total assets
6-month stock Rate of return on common stock over the 6 months
returns preceeding the AGM
ultimate shareholder The subject holding ultimate control (at the conventional
10% threshold) of a listed firm
family Dummy variable taking the value of 1 if ultimate
shareholder is a family, and 0 otherwise
financial institution Dummy variable taking the value of 1 if ultimate
shareholder is a bank or an insurance company, and 0
otherwise
private equity Dummy variable taking the value of 1 if ultimate
shareholder is a private equity fund, and 0 otherwise
state Dummy variable taking the value of 1 if ultimate
shareholder is the state or another public entity, and 0
otherwise
widely held Dummy variable taking the value of 1 if no shareholder
owns more than 10% of total voting rights, and 0
otherwise
other Dummy variable taking the value of 1 if the firm does not
belong to any of the categories described above, and
zero otherwise
VR Voting rights (%) held by the ultimate shareholder
CFR Cash-flow rights (%) held by the ultimate shareholder
wedge Difference between VR and CFR
relevant shareholder Any shareholder holding more than 2% of share capital in
a listed firm
No. of Number of relevant shareholders (other than the ultimate
shareholders>2% shareholder)
shareholders’ Dummy variable taking the value of 1 if a shareholders’
agreement agreement is in place, and 0 otherwise
Free Float Share capital (%) held neither by the ultimate shareholder,
nor by other relevant shareholders
Concentration Index Herfindhal index, i.e. sum of the squared percentage of VR
held by all relevant shareholders (including the ultimate
shareholder)
board size No. of board seats
No. of board seats No. of board seats reserved to minority nominees (on the
reserved basis of existing regulation and the company by-laws)
% board seats No. of board seats reserved/board size
reserved
old/new Dummy variable taking the value of 1 if slate voting was
already in place in 2007, and 0 otherwise
quorum Share capital (%) required to submit a slate of board
nominees
quotas/proportional Dummy variable taking the value of 1 if the company has
quotas in place, and 0 otherwise. In a quota system the
winning list takes all board seats except those ‘reserved’
to minority candidates (if present)
References
Adams, R. B., B. E. Hermalin and M. S. Weisbach 2009. ‘The Role of Boards of
Directors in Corporate Governance: A Conceptual Framework & Survey’,
Assonime-Emittenti Titoli 2011. ‘An Analysis of the Compliance with the Italian
Corporate Governance Code (Year 2010)’, available at www.Assonime.it.
Bainbridge, S. M. 2003. ‘A Comment on the SEC Shareholder Access Proposal’,
UCLA School of Law, Law & Econ. Research Paper 03–22.
2010. ‘The Corporate Governance Provisions of Dodd-Frank’, Working Paper
available at www.ssrn.com.
2012. Corporate Governance after the Financial Crisis, Oxford University
Press.
Barucci, E. and J. Falini 2005. ‘Determinants of corporate governance in the Italian
financial market’, Economic Notes 34: 371–405.
Bebchuk, L. A. 2003. ‘The case for shareholder access to the ballot’, The Business
Lawyer 59: 43–66.
2005. ‘The case for increasing shareholder power’, Harvard Law Review 118:
833–917.
2007. ‘The myth of the shareholder franchise’, Virginia Law Review 93:
675–732.
and S. Hirst 2010. ‘Private ordering and the proxy access debate’, The Business
Lawyer 65: 329–60.
Becht, M., J. Franks, C. Mayer and S. Rossi 2009. ‘Returns to shareholder activism:
Evidence from a clinical study of the Hermes UK focus fund’, Review of
Becker, B., D. B. Bergstresser and G. Subramanian 2010. ‘Does Shareholder Proxy
Access Improve Firm Value? Evidence from the Business Roundtable
Challenge’, Harvard John M. Olin Center for Law, Economics and
Business Discussion Paper 685.
Bianchi, M. and L. Enriques 2001. ‘Corporate governance in Italy after the 1998
reform: What role for institutional investors?’, Quaderni di finanza Consob 43.
Bianchi, M., A. Ciavarella, V. Novembre and R. Signoretti 2010. ‘Comply or
Explain? Investor Protection through Corporate Governance Codes’,
Björgvinsdóttir, Á. 2004. ‘Icelandic company law’, in Wahlgren (ed.), pp. 45–66.
Cai, J., J. L. Garner and R. E. Walkling 2009. ‘Electing directors’, Journal of Finance
64: 2389–421.
Choi, S. J., J. E. Fisch and M. Kahan 2011. ‘Voting through Agents: How Mutual
Funds Vote on Director Elections’, University of Pennsylvania, Institute for
Law & Economics Research Paper No. 11–28.
Chung, K. H. and H. Zhang 2011. ‘Corporate governance and institutional
ownership’, Journal of Financial and Quantitative Analysis 46: 247–73.
Cools, S. 2005. ‘The real difference in corporate law between the United States and
continental Europe: Distribution of powers’, Delaware Journal of Corporate
Law 30: 697–766.
Cotter, J., A. Palmiter and R. Thomas 2010. ‘ISS recommendations and mutual
fund voting on proxy proposals’, Villanova Law Review 55: 1–56.
Cremers, M. and R. Romano 2011. ‘Institutional investors and proxy voting: The
impact of the 2003 mutual fund voting disclosure regulation’, American
Law and Economics Review 13: 220–68.
Cziraki, P., L. Renneboog and P. Szilagyi 2010. ‘Shareholder activism through
proxy proposals: The European perspective’, European Financial
Davies, P., K. J. Hopt, R. Nowak and G. van Solinge 2012. ‘Boards in law and
practice: a cross-country analysis in Europe’, Forum Europaeum Corporate
Boards (FECB), mimeo.
De Jong, A., G. Mertens and P. Roosenboom 2006. ‘Shareholders’ voting at general
meetings: Evidence from the Netherlands’, Journal of Management and
Governance 10: 353–80.
Del Guercio, D. 1996. ‘The distorting effect of the prudent-man laws on institu-
tional equity investment’, Journal of Financial Economics 40: 31–62.
Easterbrook, F. H. and D. L. Fischel 1991. The Economic Structure of Corporate
Law, Harvard University Press, Cambridge, MA and London.
Eckbo, B. E., G. Paone and R. Urheim 2011. ‘Efficiency of Share-voting Systems:
Report on Italy’, ECGI Law Working Paper 174/2011.
Enriques, L., H. B. Hansmann and R. Kraakman 2009a. ‘The basic governance
structure: The interests of shareholders as a class’, in Kraakman et al.,
pp. 55–88.
2009b. ‘The basic governance structure: Minority shareholders and non-
shareholder constituencies’, in Kraakman et al., pp. 89–114.
Enriques, L. and P. Volpin 2007. ‘Corporate governance reforms in continental
Europe’, Journal of Economic Perspectives 21: 117–40.
Ertimur, Y. and F. Ferri 2011. ‘Does the Director Election System Matter? Evidence
from Majority Voting’, Working Paper available at www.ssrn.com.
Fama, E. F. and K. R. French 1993. ‘Common risk factors in the returns on stocks
and bonds’, Journal of Financial Economics 33: 3–56.
Fischer, P. E., J. D. Gramlich, B. P. Miller and H. D. White 2009. ‘Investor percep-
tions of board performance: Evidence from uncontested director elections’,
Journal of Accounting and Economics 48, 172–89.
Gillan, S. L. and L. T. Starks 2000. ‘Corporate governance proposals and share-
holder activism: The role of institutional investors’, Journal of Financial
2007. ‘The evolution of shareholder activism in the United States’, Journal of
Applied Corporate Finance 19: 55–73.
Gordon, J. N. 1994. ‘Institutions as relational investors: A new look at cumulative
voting’, Columbia Law Review 94: 124–92.
2008. ‘Proxy contests in an era of increasing shareholder power: Forget issuer
proxy access and focus on e-proxy’, Vanderbilt Law Review 61: 475–96.
Hamdani, A. and Y. Yafeh 2011. ‘Institutional Investors as Minority Shareholders’,
Hermalin, B. E. and M. S. Weisbach 2003. ‘Boards of directors as an endogenously
determined institution: A survey of the economic literature’, Federal
Reserve Bank of New York Economic Policy Review 9: 7–26.
Iliev, P., K. V. Lins, D. P. Miller and L. Roth 2011. ‘Shareholder Voting and Corporate
Governance around the World’, Working Paper, available at www.ssrn.com.
Kahan, M. and E. B. Rock 2011. ‘The insignificance of proxy access’, Virginia Law
Review 97: 1347–434.
Kraakman, R., J. Armour, P. Davies, L. Enriques, H. B. Hansmann, G. Hertig,
K. J. Hopt, H. Kanda, and E. B. Rock 2009. The Anatomy of Corporate Law:
A Comparative and Functional Approach, 2nd edn, Oxford University Press.
Larcker, D. F., G. Ormazabal, and D. J. Taylor 2010. ‘The Market Reaction to
Corporate Governance Regulation’, Working Paper, available at www.ssrn.com.
Levitt Jr., A., 2006. ‘Stocks populi’, The Wall Street Journal (October 27).
Malberti, C. and E. Sironi 2007. ‘The Mandatory Representation of Minority
Shareholders on the Board of Directors of Italian Listed Corporations: An
Empirical Analysis’, Bocconi University Institute of Comparative Law
‘Angelo Sraffa’ Legal Studies Research Paper Series 18.
Mallin, C. 2012. ‘Institutional investors: The vote as a tool of governance’, Journal
of Management and Governance 16: 177–96.
Matvos, G. and M. Ostrovsky 2010. ‘Heterogeneity and peer effects in mutual fund
proxy voting’, Journal of Financial Economics 98: 90–112.
Ng, L., Q. Wang and N. Zaiats 2009. ‘Firm performance and mutual fund voting’,
Journal of Banking and Finance 33: 2207–17.
Parrino, R., R. W. Sias and L. T. Starks 2003. ‘Voting with their feet: Institutional
ownership changes around forced CEO turnover’ Journal of Financial
Pozen, R. C. 2003. ‘Institutional Perspectives on Shareholder Nominations of
Corporate Directors’ Harvard Law School John M. Olin Center for Law,
Economics and Business Discussion Paper 429.
Renneboog, L. and P. Szilagyi 2011. ‘The role of shareholder proposals in corpo-
rate governance’, Journal of Corporate Finance 17: 167–88.
Roe, M. J. 1994. Strong Managers Weak Owners, Princeton University Press.
Santella, P., E. Baffi, C. Drago and D. Lattuca 2009. ‘A Comparative Analysis of the
Legal Obstacles to Institutional Investor Activism in the EU and in the US’,
Sharfman, B. S. 2012. ‘Why proxy access is harmful to corporate governance’, The
Journal of Corporation Law 37: 387–413.
Sjostrom Jr., W. K. and Y. S. Kim 2007. ‘Majority voting for the election of
directors’, Connecticut Law Review 40: 459–510.
Van der Elst, C. 2011. ‘Revisiting Shareholder Activism at AGMs: Voting
Determinants of Large and Small Shareholders’, ECGI Finance Working
Paper 311/2011.
Ventoruzzo, M. 2010. ‘Empowering shareholders in directors’ elections: A revolu-
tion in the making’, ECGI Law Working Paper 147/2010.
Wahlgren, P. (ed.) 2004. Company Law, Stockholm Institute for Scandinavian Law.
Yermack, D. 2010. ‘Shareholder voting and corporate governance’, Annual Review
of Financial Economics 2: 103–25.
Zingales, L. 2008. ‘Italy leads the way in protecting minority investors’, Financial
Times (April 14).
INDEX
Abowd, John M. 254 Assonime (Associazione fra le Società

Adams, Renée 12, 221, 233–4 italiane per Azioni) 90
AFEP-Medef (French national code) AstraZeneca 259–60
82–3 audit committees 230
AFG (Association française de gestion ‘juridification’ 119
financière) 83–5 auditors
monitoring alerts 84–5 role of 127–8
agency, defined 21 statutory appointment 381
agency costs 13–20 Austria
and corporate law 15–17 corporate governance codes 75–6,
diffuse vs. concentrated ownership 120; drafting 121
20, 253 termination payments 281
discretionary powers 23–4 Aviva 259
and executive pay 30–2
market solutions 14–15, 24 Bae, Kee-Hong 144
and relationship between controllers Bainbridge, Stephen M. 318
and other stakeholders 21 banks 10–12
role of shareholder activism 40–1 asset substitution 11
types 13 binding nature of codes 119
agency theory 228–30 bonus payments 302
dominance 229 Chief Risk Officer, role of 12
objections to 228–9, 234–5 compensation levels/practices 296,
AGM (Annual General Meeting) see 299, 302–3, 304–6; modification
general meetings 306
AMF (Autorité des marchés financiers, directors’ remuneration 34–7, 38, 51,
French regulator) 81–2, 84 251, 260–4, 272–3, 282–4, 308–9
Andres, Christian 148, 152 failure 255
Arcot, Sridhar 112 international regulation 37
Armour, John 39, 320 lack of regulation/supervision 58
Asian crisis (1997), family firms’ legal action against 88–9
response to 145 mismatch between liabilities and
ASMI case 99–100 assets 10
Assicurazioni Generali 382, 391 and moral hazard 10–11
Association of Generally Accepted nationalised 87
Principles in the Securities Market in the Netherlands 97
(Sweden) 107 prudential regulation 19–20, 49
Assogestioni 391 reasons for failure 11–12
423
424 index
banks (cont.) minimal involvement levels 241–2
response to crises 145, 151, 183–4 types of involvement 240–5
risk management 19–20 see also Matrix of Board Interaction
risk-taking tendencies 11, 49, 51, board size 192, 193, 195–9, 206–9,
254–5 215–17
state aid 283–4 defined 194
Barclays 259 EU–US comparison 197–9
Barontini, Roberto 254 and firm characteristics 206–9,
Bartz, Jenny 118 216, 220
Barucci, Emilio 367 impact of global crisis 209
bearer shares 343–4 impact on voting outcomes 387
Bebchuk, Lucian A. 229, 254, 318 national variations 199, 200, 206–9,
Becht, Marco 9, 320 217–20
Begeman case 100 reductions in 195–6, 220
Belgian corporate governance codes relationship with firm size
76–9, 122 206–9
ambiguities in national law 77–9 board structures 50–1
case law 79 alternative specifications 215–17
explanations for diversity 230 (see also gender
non-implementation 117 diversity)
measures to improve compliance EU recommendations 51
78–9 and leverage 217
monitoring 77–8, 134 national variations 50, 204, 205
national legislation 76–7 policy initiatives 221–2
Belgium post-crisis changes 204
directors’ remuneration 256–7, 281 regulation 50–1, 56–7, 247–8
see also Belgian corporate variations 27–9, 203–15
governance codes see also board size; independence
Berle-Means corporate model 2, BoardEx (database) 194–5
154, 253 boards 2, 23–9, 191–222
Bertrand, Marianne 143–4, 303–4 accountability 248
Bianchi, Marcello 91 collective responsibility 5–6
Bizjak, John 253 collegial nature 24–5
Black, Fischer 291 composition 5, 25–7, 247–8
blockholder ownership see criteria in governance codes
concentrated ownership 26–7
Board GPS 226, 236–7 differences of opinion with investors
Abilities 237 95–6
Essences 237 division of powers with shareholders
Group lens 236 42–4
Person lens 236 economic analyses 228–30
System lens 236 as focus of EU reform 6, 7–8
Traps 237 impact of crises 225
Board on Task 226, 235–7, 247 importance of role 191
defined 235–6 inability to prevent malpractice 225
facilitation of review process 245–6 individual interactions within 226,
fields of interaction 238–40, 241–2, 231–2, 235, 236, 238–9
244–5 labour representation 204
index 425
legal disputes with shareholders Cadbury Report/Code 9, 120
98–101 Cai, Jie 366
legal requirements 227–8 Cairn Energy 259–60
limits of quantitative approach 29 Caprio, Lorenzo 151, 152, 154–5
methodology of study 203–4 Cardia, Lamberto 381
misunderstandings of role 226, CEOs
234–5, 238–9 compensation 38, 291–303, 308–9;
modernisation 5–6 factors affecting 300, 303–6;
national variations 27–8, 193, 203 relationship with firm
organisational theories 237–8 performance 306
performance, evaluation of 226–7, in family firms 153, 165, 170,
246–7 171–3
policy recommendations 247–8 hiring/firing 24, 132
in practice 25–7; departures from Cheffins, Brian R. 39
theory 25 Chen, Carl R. 255
professional requirements 25 Chizema, Amon 252, 253
reform proposals 192 Clacher, Iain 125
regulation in line with stakeholder CMVM (Portuguese securities
interests 10 regulator) 101–4
remit 24 CNMV (Spanish securities regulator)
reporting procedures 72 104–7
reviews of own performance 245, 248 annual reports 105–6
role in implementation of codes Coffee, John C., Jr. 254
114–16, 127 compensation 15, 254, 291–306
sample data 194–5 alignment with risk measurement
setting of pay levels 31 35–6
shareholder powers vis-à-vis 42–4 composition of packages 266,
studies 191–2, 193, 227, 230–4, 296–302
246, 393 data 266–9
in theory 24–5 excessive 51–2
understandings of own role 245–6 factors affecting 300–1, 303–6,
variations by firm characteristics 307–9
193, 203 in financial vs. non-financial firms
variations by firm performance 193 296, 298–9, 302–3, 304–6
variations over time 28, 192–3 individual 288
weak 43 international standards 35–6
see also board size; board structures; national variations 291–6
directors; elections; gender performance-based packages 302
diversity; non-executive relationship with firm performance
(supervisory) directors; voting 303–5
Bognanno, Michael L. 254 relationship with firm size 303–5
Bolkestein, Frits 4 shareholder dissent 335
BoSA (Board of Statutory Auditors), statistical summary 291–303
elections to 378, 412 total, evolution of 296
Bozzi, Stefano 254 competition
Bruno, Valentina 112 impact on agency costs 14–15
Buchanan, Bonnie 319 compliance, as shareholder policy
business judgment rule 16, 17 131, 133
426 index
‘comply or explain’ principle 17–19, 20, company departures from 69–70, 89
55–6, 58, 116–17, 136, 137 comparative studies 135
ambiguities 18, 116–17 double-layered system 114–16
criticisms 87 drafting 119–20, 121
in EU legislation 70 Dutch 92–8
and executive pay 32–3 extent of adoption 71, 118, 120–2,
flexibility 55 318–19
in national codes 76–7, 79, 83, German 85–8
89–90, 93, 97, 101, 103, 104–5, governmental monitoring see
106, 111 corporate governance
problems of implementation 55–6 commissions
see also explanation(s) improvements to effectiveness
concentrated ownership 20–2 133–6
and adherence to codes 115, 124–5, incentives to adopt 94
130–1, 132 interpretation 69, 94
agency costs 20–1 Italian 89–91
and board elections 377 mandatory adoption 69–71, 101,
and compensation packages 305 107–8, 116
remuneration practices 31–2, 289, Member State 70, 74–113;
307–8 similarities between 120
and suboptimal diversification 21–2 Portuguese 101–4
see blockholder ownership (proposed) common principles 137
Consob (Italian regulator) 379, 381–2, public/private character 114
390–1, 402 questionnaires 76
Conyon, Martin J. 254 recommendations for future
Cools, Sofie 372–3 137–8
corporate governance relationship with hard law 68–70, 119
and banks 10–12, 49 remuneration provisions 279–80
commissions 73 scholarship 125
defined 9 scope 73–4
flaws 3, 58 variations in companies’ approach to
‘law matters’ theory 21–2 121–2
links with firm value 145–6 voluntary aspects 116–17
monitoring bodies 18–19 see also ‘comply or explain’
policy 49–58 principle; implementation of
problems of harmonisation 19, 55–6 corporate governance codes
regulation 9 corporate governance commissions
soft law instruments 82 (national bodies) 121–2,
varieties 8–12, 49–50 133–6, 138
see also corporate governance codes; engagement with stakeholders 135
implementation of corporate interaction with companies
governance codes; reform (of 135–6, 138
corporate law) interaction with each other 136–7, 138
corporate governance codes 17–18, 248 private law status 134–5
adherence to 115 right of public comment/naming
aims 85–6 136, 138
companies’ choice of non-home corporate governance statements
state 70 69–70
index 427
correspondence with company term of office 371–2
practice 71–2 see also boards; independence;
national bodies dealing with see non-executive (supervisory)
corporate governance directors; remuneration
commissions disclosure 17–19
national requirements 75–6, 90 board structure policy 56–7
corporate law 15–17 compliance levels 280–2
approach to boards 227–8 EU reforms 4–5, 6, 58
effectiveness 118 governance function 17
impact on corporate governance individual 282
13–14 legislative function 17–18
limitations 16–17 mandatory vs. soft-law 34, 52–3
protection of investors 16 requirements in national laws/codes
relationship with codes 68–9, 119 77, 112, 123–4, 280–2
credit crisis (2008–12) see global verification 135
financial crisis see also remuneration; transparency
crises Djankov, Simeon 317, 340, 344
impact on employment levels Dodd-Frank Act (US 2010) 375, 376
179–83 dotcom crisis (2001–3) see governance
impact on family firms 158–81 crisis
impact on firm performance 166–83 Drobetz, Wolfgang 125
see also Asian crisis; dotcom crisis; Dutch Corporate Governance Code
global financial crisis 92–4, 231, 232
Croci, Ettore 254 adherence to 115
Cziraki, Peter 319–20 drafting 121
explanations for
De Jong, Abe 320, 366 non-implementation 117
de Larosière, Jacques/de Larosière monitoring 123–4
Report 7
Denmark, corporate governance codes Eastern Europe, former communist
80, 123 countries 203, 212
adherence to 115 adoption of codetermination
diffuse ownership 20–2, 269 principles 204
and adherence to codes 115–16, 124, ECGI (European Corporate
126–7, 130–2 Governance Institute) 120
agency costs 20–2 Eckbo, B. Espen 367
remuneration practices 30, 289, economics, and board performance/
307–8 regulation 228–30
and shareholder activism 1, 43, 321 Economiesuisse 108–9
directing, as role of non-executives elections 43, 54, 365–416
241, 242 centrality to corporate governance
directors 372, 413
margin of discretion 227–8 contested 369
opportunities for malpractice 228, empirical analysis 383–412;
235, 242 robustness checks 412
personal characteristics 236 multiple-winner systems 48–9, 371,
personal liability 227 379, 383, 413, 414, 415–16
removal 43, 371–2 open vs. closed lists 371, 379–80
428 index
elections (cont.) shareholder rights legislation
and ownership structures 373, 388 41–2, 46
policy debate 373–8 expertise, directorial 232–3
processes 368–73 impact on firm performance 232–3
quota system 369 testing 233
removal rights 371–2 explanation(s)
single-winner systems 371, 373–4 ambiguity of requirements 116–17
uncontested 370, 373–4, 375–6 failure to provide 72, 94
see also nomination ‘proper,’ defined 117, 121
employees, board representation 43,
204, 321 Fahlenbrach, Rüdiger 255
enforcement, instruments of 102–3 Falini, Jury 367
engaging, as role of non-executives family firms 22–3, 143–86
240–1, 242 access to outside capital 145
Ertimur, Yonca 319, 339 age 156–7, 165
Eumedion 94–6, 123 behavioural differences from other
European Commission firm types 143–50, 165–86
policy on shareholder engagement/ cash holdings 157, 174
rights 58–9, 315–16 CF/K variable 174
reports on general meetings 323, 340 ‘dark side’ 150–1
Action Plan on EU Company Law defined 151–3
(2012) 56–8 downsizing 145, 155, 175–7, 179,
Communication on Modernising 183, 185
Company Law (2003) 4–6 employees 165, 179–83, 185; share
Green Paper on Corporate ownership 185
Governance in financial exacerbation of crisis 184–5
institutions (2010) 6–7, 20, 45, family members in top positions
256, 365–6, 376–7 153, 170, 171–3
Green Paper on the EU Corporate fears of expropriation 144, 147–8,
Governance framework (2011) 185–6
6–8, 9, 20, 27, 34, 43–5, 48, 56, governance codes 87–8
150–1, 183–4, 185, 186, 256, 315, implicit contracts 149, 185
365–6, 377 investment-cash flow sensitivity (I/K
Recommendation on directors’ pay ratio) 148, 155, 159, 170
at non-financial companies investment levels during crises
(2009) 33–4, 256 174–9, 183
Recommendation on the investment policies 170–9, 184–5
remuneration of directors (2004) investment regressions 157
32–3, 255–6, 264 in Italy 156–7
Recommendation on the role of ‘market-to-book’ variable
non-executive or supervisory 157, 170
directors (2005) 26–7, 32–3, 231, maximisation of utility 144, 147
255–6, 376–7 methodology of study 157–9
European Union national variations 150, 154
corporate law 13–14 outperformance of non-family firms
corporate reforms 3–8 145, 170, 186
legal basis of governance codes 134 outside CEOs 165, 170, 171–3
remuneration reforms 255–64 ownership criteria 151–2
index 429
‘patient capital’ 143–4, 146 role of shareholders 83–5
performance regressions 158 shareholder activism 320, 328,
policy issues 150–1 331–4
positive/stabilising role 150–1 soft-law instruments 82
private benefits of control 147, fraud, incidences of 3
150–1, 156 FRC (Financial Reporting Council,
relationship of firm performance UK) 111–13, 114, 121
and ownership characteristics Fried, Jesse 229
166–9 FSB (Financial Stability Board),
reliance on bank debt 151, 183–5 Principles and Standards on
response to short-term price compensation practices 35–7,
movements 146, 184 260–1
responses to crises 143–6, 158–9, as acceptable compromise 36
163–81, 183–6 implementation 36–7
role of founding family 152 main areas covered 35–6
sales 157, 165 FSMA (Belgian regulator) 77–9
sample 154–5 FTSE (Financial Times Stock
shareholder engagement 184 Exchange) group 264–5
shareholders’ propping behaviour
148–9 G20 group 35, 260–1
studies 143–9, 151–5, 170 gender diversity (on boards) 25–6, 27,
tendency to conservatism 143–4, 28–9, 50, 56, 213–15, 217, 220–2,
147–8 230
total assets 165, 179 EU–US comparison 198–202
univariate analysis 159–65 and firm characteristics 213,
variables 155–7; performance- 214, 219
related 156 governmental policy initiatives
wages 149, 155–6, 165, 179–83, 185 200–2, 213, 220
Ferreira, Daniel 203, 233–4 increase in 192–3, 197, 215
financial institutions see banks ‘juridification’ 119, 134
Finland measurement 194–5
board characteristics/regulation national variations 202, 213
199–201, 220 proposals for improvement 192
shareholder proposals 336 relationship with firm performance
Forbes, Daniel P. 234 217–20, 233–4
Fortis 79 general meetings 42, 94, 316–58
France 80–5 addressing of governance
AGM voting trends 84 concerns 317
board structures 204 attendance rates 320–1, 328, 357
corporate governance codes 82–3, cross-border participation 323
114, 116, 122, 123 databases 329, 339–40
directors’ remuneration 258, 259, in French law 82, 84–5
276–9, 307 legal disputes relating to 98–101
disclosure requirements 126, 281 multivariate analysis 339–57, 360
family firms 149 national variations 320–1, 331
market supervision 81–2 negative votes 84
monitoring system 134 notice periods 343
provisions in national law 81–2 powers 42–3
430 index
general meetings (cont.) impact on family firms 150–1, 183,
proposals submitted to 328–39; data 185
sources 329–31 (see also impact on market capitalisation 276
management proposals; impact on remuneration 34–5, 38,
shareholder proposals) 53, 264–89, 296, 308–9
regulatory conditions 316, 324–5 impact on shareholder rights/
relationship with firm characteristics activism 315, 318–19, 357, 358
340–4 impact on stock returns 276
role in implementation of codes 128, start/end 154
131–2 Gordon, Jeffrey N. 374
role of activism 326–8 governance crisis (2001–3) 144–5,
setting aside of decisions 88–9 153–4
setting of remuneration policy 96 impact on boards 225, 229
shareholder dissent 316–17 impact on family firms 166–9, 183,
studies 366 185
Georgeson (consultancy) 323, 340 start/end 154–9
German Corporate Governance Code Greece, directors’ remuneration 265
70, 85–8, 120, 123 Grinstein, Yaniv 252
central provisions 86
criticisms 86–7 Hamdani, Assaf 366
drafting 121 harmonisation 136–7
effectiveness 118 Harris, Milton 318
remuneration provisions 279 Hartzell, Jay C. 304
Germany 85–9 Hau, Harald 232
board elections 368 Hayes, Rachel M. 252–3
board sizes 199 hedge funds 126–7, 319
board structures 204 Heifetz, Ronald A. 246
business judgement rule 17 hygiene, as board function 237, 239
case law 88–9 ORA applied to 244
codetermination system 1
directors’ liability 127 Iceland
directors’ remuneration 258, 276–9, board characteristics/regulation
291–6, 307 200–1
disclosure requirements 281 board elections 371
investor associations 87 implementation of corporate
legal scholarship 125 governance codes 67–138
removal of directors 372 ambiguities in national systems 77
role of AGM 128 case law 71, 72, 79, 88–9, 98–101
shareholder activism 328 drivers 72–3
see also German Corporate ex ante 68
Governance Code extra-company factors 73
Gillan, Stuart L. 39 failures of 67–8, 94, 103
Girard, Carine 320 measurement of effectiveness 71–2,
global financial crisis (2008–12) 6–8, 118
144–5, 153–4 measures to improve compliance
impact on banks 11–12 78–9
impact on board structures/ methods 70–1
regulation 209, 212, 225, 229–30 monitoring 77–8, 81–2, 91
index 431
national variations 71, 74–113, 133 directors’ remuneration 256–7, 259,
preliminary findings 113–37 262–4, 276–9, 307
role of shareholders 124–7 disclosure requirements 281
self-regulation 68, 136; problems of firm characteristics 384
119 market regulation 90
see also explanation(s) privatisation law 380
independence, directorial 192–3, Protection of Savings Law
209–12, 217, 231–2 381–2, 393
EU–US comparison 197–9, 220 reporting procedure 90
improvements in 196, 220 shareholder activism 328
increased stress on (post-2008) 230 shareholder voting system 47–9,
national variations 201, 209–12, 279 54, 331
(problems of) definition 194, 231–2 stock market 383, 390
relationship with firm characteristics
209–12, 218 Japan, board elections 371
relationship with firm size 212 Jensen, Michael 228, 229, 253
insider trading 137
institutional investors 94–6 Keynes, J.M. 247
behaviour specific to 379 Kim, Jeong-Bon 255
differences of opinion with boards Kirchmaier, Tom 221
95–6
obligation to vote/report on La Porta, Rafael 269
votes 129 labour market 15
participation at general Lang, Mark H. 252
meetings 328 LaSalle Bank 99
pressure-sensitive, shareholdings Lehman Brothers 153
343 Li Xiao 125
proposals submitted by 337 Linsky, Marty 246
publication of voting policies/ Lipton, Martin 318
records 377 Loewenstein, Mark J. 30
recommendations concerning 130, Lundholm, Russell J. 252
137 Luxembourg
role in implementation of codes corporate governance codes
129–30 92, 123
insurance contracts, implicit 149 directors’ remuneration 262
intervention, as shareholder policy 131,
132 MacNeil, Iain 125
investment funds 131, 337 Malaysia, corporate governance
Ireland, directors’ remuneration 265 code 67
Italy Malberti, Corrado 367
board appointments 186 management proposals 330
board elections 367, 378–412 objectives 335, 345–7
Consolidated Law on Finance (1998) proposal characteristics 344–8
380–1 voting outcomes 331–5, 344–8;
corporate governance codes 89–91; relationship with firm
amendment 90–1 performance 348; role of national
Corporate Governance Committee regulation 348
390 Masouros, Pavlos 228, 318, 327
432 index
Matrix of Board Interaction 241–5, 247 pension fund-style strategy 392–3
dynamic nature 242 risks 390–1; management 391–2
facilitation of review process shareholder activism 404–12
245–6
ORA applied to 243–5 Netherlands 92–101
McCahery, Joseph A. 328 Banking Code 233
McKinsey Quarterly 233 board elections 368
Means, Gardiner see Berle-Means board sizes 199
corporate model case law 98–101, 132, 238
Meckling, William H. 228, 253 corporate governance
Milliken, Frances J. 234 recommendations 95–8
minority shareholders corporate law 79, 227, 230
access to boardroom 50, 327 directors’ remuneration 258
board appointments reserved to 186, disclosure requirements 126
371, 377–8, 380, 395, 403, 415 Enterprise Chamber
communications between 357 (Ondernemingskamer) 98–101,
costs of participation 321–3 136
in family firms 144, 147–8, 150–1, institutional investors 94–6
185–6 Monitoring Commission 93
fears of controllers/management 22 public investors 96–7
in Italian system 380–2, 383 shareholder activism 328
links with controlling shareholder see also Dutch Corporate
382 Governance Code
protection of interests 3, 20, 43–4, Nigeria, corporate governance code
186, 344, 352, 377–8 319
‘rational apathy’ 316–17, 321, 357, nomination
374–5, 414 committees 368–9
submission of board candidates 48, processes of 368–70, 376–8
367, 389–412 restrictions 391
variables 393 separated from slate submissions
monitoring 391–2
absence 135 non-executive (supervisory)
commissions, nature/role 133–5 directors 5
external 133–6, 138 composition 26
internal 127–33, 137 in Dutch law 227
tools 135 enhancement of monitoring role
monitoring hypothesis 389, 394 (post-2008) 225–6, 230, 237–8,
lack of support for 402, 411, 414, 248
415–16 impact on firm performance 230–4,
support for 402–4 247–8
moral hazard 10–11 interaction with executives 29, 236,
Mullainathan, Sendhil 303–4 238–46; research into 246–7
Murphy, Kevin J. 229, 255 legal liability 127
Muslu, Volkan 252 number/balance 26
mutual funds 48, 413–14 personal relationships 231
hypothesis specific to 392 professional profile 26–7
in Italy 380, 390–3, 404–12 removal 44
leverage 404–11 remuneration 288
index 433
role in management 26 people, role of board in relation to
self-evaluation 26 239–40
shift in expectations of 242 performance, as board function 239
strategic role 238, 244–5 Philippines, corporate governance
taking control of investigations 242 code 319
types of involvement 240–5 portfolio composition hypothesis 392–
understandings of own role 245–6 3, 394–5
see also independence linked to mutual funds 392–3
non-financial firms 50 support for 402, 411–12, 414, 415
compensation levels 296, 298, 302–3, Portugal
304–5 directors’ remuneration 256–7, 265
consequences of excessive risk- shareholder proposals 336
taking 51–2 see also Portuguese corporate
directors’ remuneration 32–4, governance codes
274–5, 308–9 Portugal Telecom 103
shareholder activism/submissions Portuguese corporate governance
402–3 codes 101–4, 114, 122
Norway monitoring system 102–3, 118, 123,
board characteristics/regulation 133–4
200–1, 220 specific features of regime 103–4
directors’ remuneration 258 verification process 102
Nowak, Margaret 125 Prigge, Stefan 125
NVB (Dutch Bankers’ Association) 97 probing, as role of non-executives 240,
241–2
OECD (Organisation for Economic ORA applied to 243–4
Co-operation and Development) proportionality principle 262–4
11, 254, 368–9 protection (of investors) 16
Organisational Role Analysis (ORA) mandatory rules 16
236, 237–8, 242–5, 246 public investors 96–7
defined 242–3
ownership structures 20–3 Q Ratio 155, 166, 170
and election process 373, 388 impact of crisis periods 159,
and general meetings 321, 341, 343, 163–8
354–5 quorum (of shareholders) 387–9
impact on remuneration 253, 269, ownership thresholds 382
290–, 303, 304 proposals despite absence of 379
in Italian firms 385 size, impact on voting/submissions
national variations 21, 23 395, 402, 403, 411–12, 413–14
optimal, difficulty of establishing 22
relationship with shareholder ratifying, as role of non-executives 240
activism 394–402, 403, 404, 413 ORA applied to 243–4
see also concentrated ownership; Raviv, Artur 318
diffuse ownership; family firms RCS Media Group 387
reform (of corporate law) 3–8
Padgett, Carol 125 coordination of national efforts 6
Parmalat 186, 381 ‘pillars’ 4–6
Partnoy, Frank 233 regulation hypothesis 389, 394–5
pension funds 319, 392–3 lack of support for 402
434 index
remuneration (of directors) 251–310 performance-based 36, 229;
as agency cost 30–1 contracts 30
and blockholder ownership 31–2 policy statements 287–8
committees 33, 253, 257, 266, 279, reforms 288, 309
287, 307; composition regulatory recommendations 229
requirements 287; variations 289 relationship with firm size 269, 289
competing views 30–2 relationship with firms’ financial
compliance with EU standards characteristics 263, 277–91, 289,
261–4, 279, 284–9, 309–10; lack 304–285, 308–9
of progress 285; national remedying of agency costs 30–2
variations 287 reporting process 264
dataset/methodology of study 251, statistical summary 269–76
264–9, 284–5 studies 251–5
disclosure 5, 32–4, 36, 52–3, 252–3, see also compensation; termination
267–8, 276–9, 277–, 280–2, 283, payments
288, 307; flaws in system 33; Rio Tinto 112
improvements in 285, 287–9; ROA (accounting performance
mandatory 34; national variations variable) 155, 166, 170
33–4 impact of crises 159, 167–8
empirical analysis 37–8 Roe, Mark J. 41
EU policy 51–3 Rosen, Sherwin 303
evolution across jurisdictions
270–8, 287–90 Sarbanes-Oxley Act (US 2002) 3, 231
in financial institutions 34–7, 51–2, Schaefer, Scott 252–3
260–4, 272–3, 282–4 Schoar, Antoinette 143–4
fundamental considerations 53 Scholes, Myron 291
governance 267–8, 276–80, Schwalbach, Joachim 254
277–91 shareholder activism 3, 39–49
harmonisation, plans/need for and adherence to codes 115–16,
309–10 128–9
impact of 2008 crisis 34–5, 38, in board elections 54
264–89, 308–9 changes 39
implementation of code provisions circumstances favourable to 1–2, 44,
68–9, 71, 96, 124 48–9, 375
implementation of EU costs of 40–1, 57, 327, 377–8
recommendations 37, 52, 255–64, cross-border 323
288–9, 307–10 in family firms 184
incentive schemes 1, 16, 30–8, 302 at general meetings 326–8
information lacking on 266, 280 impact of regulation 41, 54, 316, 317,
influence of shareholders 258–60, 348, 352, 356
282 impact of voting rules 414
investors’ rejection of plans 124–5 and implementation of codes 73
‘juridification’ 119 in Italy 47–9, 404–12
mandatory regulation 256–8 national variations 42–4, 46–7,
monitoring 57–8, 103 334–5, 349
national variations 33–4, 37, 38, 251, nature of impact 40
261, 276–82, 284, 309–10 (need for) harmonisation 46, 47
in non-financial firms 32–4, 274–5 obstacles to 321, 322–5
index 435
reform proposals 44–7, 53–5, 57, as focus of EU reform 6, 7–8, 44–6
357, 358 influence on directors’ remuneration
relationship with firm characteristics 258–60, 282
393, 395–404, 413–16 lack of involvement 128
relationship with firm size 402, 413 legal disputes with boards 98–101
relationship with voting rules 400–1, monitoring tools 127–33
413 in national laws 80, 83–5
studies 315–16, 318–20, 357, nomination of board candidates 369,
366–7 374–5
targets 53–4, 348–52 powers 42–4
types 39–41 relationship with companies 130–3,
see also shareholder proposals; 137
takeovers, resistance to right to remove directors 372
shareholder proposals rights, role in corporate governance
frequency of submission 330, 331, 228, 342
338, 349–52 role in implementation of codes 72–
government-submitted 336 3, 114–16, 124–7
(lack of) obligation to implement ‘short-termism’ 44–5
339 States as 87–8, 336
national variations 330, 335–9, strengthening of rights 5, 41–2, 53,
340–4 57–8, 87, 315, 323–6
objectives 336, 353 ‘three-layered’ form of involvement
proposal characteristics 352–6 131–3
relationship with firm characteristics see also blockholder ownership;
340–4, 350–5 diffuse ownership; general
relationship with governance meetings; minority shareholders;
environment 342, 350–5, 356–7 shareholder activism; voting
sponsors 336, 339, 353 Shaukat, Amama 125
success 337; relationship with firm Sironi, Emiliano 367
performance 349, 356; Slovenia, board characteristics/
relationship with firm size 349, regulation 200–1
356 small firms
support base 328, 349 elections/shareholder activism 384
voting outcomes 331–9, 352–7 public bond market 183–4
Shareholder Rights Directive (2007) Spain
315–17, 318–19, 323–6, 343, 348, board characteristics/regulation
376 201–2
criticisms 326 board elections 371
key provisions 323–6 directors’ remuneration 259, 265,
limits 327, 357–8 276
transposition into national law 382–3 shareholder activism 328
shareholders see also Spanish corporate
and agency theory 228 governance codes
agreements 392 Spanish corporate governance codes
anti-self-dealing measures 344 104–7
approach based on 1, 8–10 double nature 105
dissent from management proposals implementation 105–7, 114
259–60, 316–17, 328–35, 336 improvements in reporting 105–6
436 index
Spanish corporate governance codes Thesmar, David 157
(cont.) Thum, Marcel 232
independence criteria 106 total assets, changes in level 179
monitoring system 118, 133–4 transaction costs hypothesis 389, 393,
Unified Code, drafting/adoption 394, 395
104–5, 122 support for 402, 403, 412, 414,
Sraer, David 157 415–16
stakeholder approach 1, 8–9 transparency
followed by EU legislators 9–10 failures of 45
in national laws 80 moves to enhance 47, 54–5, 106–7
stakeholders, boards’ interaction optimal degree 52–3
with 240 Trinity Mirror 259–60
Starks, Laura T. 39, 304
stewardship UK Corporate Governance Code
as shareholders’ position 131–3 111–12, 114, 117, 121, 231
theory 238 (absence of) case law 113
stock grants 291, 302 implementation 111–12,
Storck case 126 115–16
Stout, Lynn A. 318 United Arab Emirates, corporate
strategy, as board function 238, 239 governance code 319
ORA applied to 244–5 United Kingdom 17
‘streetlight effect’ 246 board elections 372
Stulz, Rene M. 255 board regulatory policy 220, 221
Sun, Jerry 253 board sizes 199, 206–9, 217–20
Sweden board structures 204
Annual Accounts Act 107 directors’ remuneration 257–8, 259,
annual reports 108 262, 279, 280, 291–6, 307; reform
corporate governance codes 107–8, proposals 258
123 disclosure requirements 126, 281
directors’ remuneration 258 family firms 154
self-regulation 107 Listing Authority 111
Switzerland, corporate governance shareholder activism 320, 321,
codes 108–10, 123 327–8
compliance 110, 122, 281 shareholder proposals 337–9
directors’ remuneration 307 Stewardship Code 45, 54–5, 111,
drafting 109 112–13, 126, 127
Listing Rules 109–10 voting procedures 344
role of stock exchange 108–10 see also UK Corporate Governance
Code
Tabaksblat Code (Netherlands) 92–3, United States
99, 100 board elections 365–6, 370, 371,
takeovers 376 372–6, 413
governance role 14–15 board structure/composition 28,
impact on voting practice 374 192–3, 194, 197–9, 204
resistance to 335, 337–9 corporate law 3, 16–17
Tanzi, Calisto 381 directors’ remuneration 38, 229, 258,
termination payments 281–2 260, 308
compliance levels 281 disclosure requirements 126
index 437
pension funds 319 remote 369–70
removal of directors 372 rules, relationship with shareholder
shareholder activism 46–7, 53, 318, activism 377–8, 400–1, 409–10,
321, 326–7, 331, 338, 349 413, 414–15
Troubled Asset Relief Program 12 shareholder rights 41
‘slate’ system (Italy) 379–82, 389
van der Elst, Christoph 125, 366 variables 395, 411–12, 416–17
van Zijl, Niels 232 see also elections; general meetings;
VEB (Vereniging van Effektenbezitters, management proposals;
Association of Securities shareholder proposals
Investors) 71, 96–7, 123
Versattel case 99 ‘Wedge’ variable 156
VIP (Vereinigung Institutionelle Werder, Axel von 86–7, 118
Privatanleger, Association of William Hill Ltd 259–60
Institutional Shareholders) 87 Williamson, Oliver 24
voting 370–1 Winter, Jaap 4, 131
agents 129–30 women, labour force participation 221
broker 375–6 see also gender diversity
cross-border 5, 41 workers, participation in governance
cumulative 371, 374 10
dissemination of results 329 World Bank 159, 317, 340
national variations 370
plurality vs. majority 370, 379 Yafeh, Yishay 366
processes 344, 370
proportional 371 Zetsche, Dieter 321
proxy 369–70, 374–5 Zhou, Y.M. 145
regulation 365–6 Zimmermann, Jochen 125

Boards and Shareholders in European Listed Companies PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Boards and Shareholders in European Listed Companies PDF

Hochgeladen von

Copyright:

Verfügbare Formate

BOARDS AND SHAREHOLDERS IN

EUROPEAN LISTED COMPANIES

With contributions by distinguished scholars from legal and ﬁnancial

massimo belcredi is Professor of Corporate Finance at the Università

The papers collected in this volume, written by a group of leading

Emittenti Titoli, a company promoted by Assonime, and whose capital

1 Corporate boards, incentive pay and shareholder activism in

7 Shareholder engagement at European general

christian andres is Professor of Empirical Corporate Finance, Otto

luc renneeboog is Professor of Corporate Finance, Tilburg University

Corporate boards, incentive pay and shareholder

In general, we assume that boards are an essential mechanism for

1.2.1. After Enron

analysed brieﬂy in the following paragraph, and more extensively in

were translated into Articles 50b and 50c of Directive 78/660/EEC

1.2.2. The recent ﬁnancial crisis

between ﬁnancial institutions and other ﬁrms. Therefore, two Green

As highlighted by the de Larosière report, it is clear that boards of

This statement helps understand the remaining contents of the Green

1.3. Varieties of corporate governance

1.3.1. Shareholder v. stakeholder governance

and Devolopment (OECD) Principles of Corporate Governance. These

1.3.2. Bank governance

instead beneﬁt from potential gains (for example, by having an implicit

during the crisis. On the contrary, banks with more pro-shareholder

2. Controlling agency costs

EU powers in this area, since EU legislation can easily cover disclosure by

2.1. Market solutions

2.2. Corporate law

provide a complete solution to agency problems (Barnea et al. 1981).

exempting them from ex post legal scrutiny (Hopt 2011).17 Indeed,

2.3. ‘Comply or explain’

that a motivation is given for non-compliance with one or more of its

other side, public authorities are reluctant to lend their assistance to

2.4. Bank prudential regulation

3.1. Dispersed v. concentrated ownership

keeping managerial agency costs under control. Consequently, good

3.2. The case of family ﬁrms

of incomplete contracts. This leaves room for governance mechanisms

monitors and grants superior decision-making under a number of cir-

organisational reforms also translated into effective improvements of

recent and relevant background and experience in ﬁnance and accounting

4.3. Empirical analysis

country-level governance rules play an important role in the determin-

Furthermore, while protection of equal opportunities can hardly be

4.4. Limits of a quantitative approach

5.1. Two views

2004). In practice, pay is not set by the shareholders; rather, it is set on

may also be a non-executive director), the owner monitors the manager’s

5.2. Non-ﬁnancial ﬁrms

conformity with the Recommendations at Europe’s largest companies

to the governance process, while the remuneration committee is gener-

5.3. Financial institutions (banks in particular)

institutions rescued by the states, whilst compensation structures were

difﬁcult to measure, such as liquidity risk, reputation risk, and capital

of bankers and hostility to their remuneration practices are strong, the

5.4. Empirical analysis

Barontini, Bozzi, Ferrarini and Ungureanu also analyse the dynamics

6.1. Types and role of activism

Activism interferes with corporate decision-making and, in a sense,

investors, and the active monitoring of portfolio company strategy and

6.2. Regulatory impact and reform proposals

6.2.1. Shareholder rights