(Palgrave Macmillan Studies in Banking and Financial Institutions) Franco Fiordelisi, Philip Molyneux (Auth.) - Shareholder Value in Banking-Palgrave Macmillan UK (2006) PDF

Shareholder Value in Banking
Palgrave Macmillan Studies in Banking and Financial Institutions

Series Editor: Professor Philip Molyneux
The Palgrave Macmillan Studies in Banking and Financial Institutions will
be international in orientation and include studies of banking within
particular countries or regions, and studies of particular themes such as
Corporate Banking, Risk Management, Mergers and Acquisitions, etc. The
books will be focused upon research and practice, and include up-to-date
and innovative studies on contemporary topics in banking that will have
global impact and influence.
Titles include:
Santiago Carbó, Edward P.M.Gardener and Philip Molyneux
FINANCIAL EXCLUSION
Franco Fiordelisi and Philip Molyneux
SHAREHOLDER VALUE IN BANKING
Munawar Iqbal and Philip Molyneux
THIRTY YEARS OF ISLAMIC BANKING
History, Performance and Prospects
Philip Molyneux and Munawar Iqbal
BANKING AND FINANCIAL SYSTEMS IN THE ARAB WORLD
Palgrave Macmillan Studies in Banking and Financial Institutions

Series Standing Order ISBN 978-1-4039-4872-4
You can receive future titles in this series as they are published by placing a standing order. Please
contact your bookseller or, in case of difficulty, write to us at the address below with your name
and address, the titles of the series and the ISBN quoted above.
Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire

RG21 6XS, England
Shareholder Value in Banking
Franco Fiordelisi
University of Rome III, Rome, Italy
and
Philip Molyneux
University of Wales, Bangor, UK
Franco Fiordelisi and Philip Molyneux © 2006
Softcover reprint of the hardcover 1st edition 2006 978-1-4039-9666-4
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No paragraph of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency, 90
Tottenham Court Road, London W1T 4LP.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified as the authors
of this work in accordance with the Copyright, Designs and
Patents Act 1988.
First published 2006 by
PALGRAVE MACMILLAN
Houndmills, Basingstoke, Hampshire RG21 6XS and
175 Fifth Avenue, New York, N.Y. 10010
Companies and representatives throughout the world
PALGRAVE MACMILLAN is the global academic imprint of the Palgrave
Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd.
Macmillan® is a registered trademark in the United States, United Kingdom
and other countries. Palgrave is a registered trademark in the European
Union and other countries.
ISBN 978-1-349-54557-5 ISBN 978-0-230-59592-7 (eBook)
DOI 10.1057/9780230595927
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources.
A catalogue record for this book is available from the British Library.
Library of Congress Cataloging-in-Publication Data
Fiordelisi, Franco, 1972–
Shareholder value in banking / Franco Fiordelisi, Philip Molyneux.
p. cm. – (Palgrave Macmillan studies in banking and financial
institutions)
Includes bibliographical references and index.
ISBN 978-1-349-54557-5
1. Banks and banking–Valuation. 2. Financial institutions–Valuation.
3. Risk management. I. Molyneux, Philip. II. Title. III. Series.
HG1707.7.F56 2006
332.120681–dc22 2005044625
10 9 8 7 6 5 4 3 2 1
15 14 13 12 11 10 09 08 07 06
To my wife Patrizia, my value-added
(Franco Fiordelisi)
To Del and my cheeky kids – Catrin, Lois,

Rhiannon, Alun, Gareth and Gethin
(Phil Molyneux)
This page intentionally left blank
Contents
List of Tables ix
List of Figures xii
About the Authors xiv
Preface xv
1 Why Study Shareholder Value Creation in Banking? 1
Introduction 1
Motivation and key questions 2
Structure of the text 3
2 Economic Objectives of Banks 9
Introduction 9
Economic foundations of the theory of the firm 10
Stakeholder models 13
Shareholder-value approach 18
The shareholder-value approach in European banking 29
Conclusion 36
3 Shareholder Value – a Literature Review 38
Introduction 38
The concept of shareholder value 38
Evolution of Shareholder Value Theory 40
Shareholder value metrics 49
A literature review of empirical studies dealing with 75
shareholder value
Conclusion 94
4 How Banks Create Shareholder Value 95
Introduction 95
Endogenous channels 97
Exogenous channels: financial consolidation 129
Conclusion 144
5 Bank Performance Measures and Shareholder Value 146
Introduction 146
Bank performance measures and shareholder value – 147
methodology
Sample description 160
Information content of bank performance measures 163
vii
viii Contents
Measuring shareholder value in Europe 171

Conclusion 191
6 Measuring Shareholder Value Drivers in Banking 194
Introduction 194
Measuring cost efficiency 195
Measuring profit efficiency 218
Measuring productivity changes 223
Measuring customer satisfaction 239
Conclusion 254
7 Determinants of Shareholder Value in European Banking and 256
Shareholder Value Efficiency
Introduction 256
Analysing the determinants of shareholder value 256
The concept of shareholder value efficiency 278
Measuring shareholder value efficiency 281
Conclusion 288
8 Conclusions 289
Limitations 292
Future research 293
Notes 295
References 304
Index 330
List of Tables
2.1 Number of banks in European banking systems between 32

1985 and 2002
2.2 Bank privatisation in Italy (1993–2001) 35
3.1 Discounting cash flow valuation: equity vs. firm valuation 55
3.2 Innovation of performance measures
3.3 FT’s European company performance survey: best European 71
sector performers
3.4 FT’s European company performance survey: bank sector total 78
shareholder return
3.5 Value relevance literature classified by methodology and 85
motivation
4.1 The percentage of customers at risk of leaving the bank 103
4.2 Studies of the efficiency of banks up to 1997 112
4.3 Restructuring of the banking sector 128
4.4 Merger and acquisition in the main industrial countries 130
4.5 A selection of major domestic bank mergers in Europe 132
4.6 A selection of cross-border acquisitions of merchant banks 133
4.7 A selection of cross-border acquisitions of commercial banks 133
4.8 Overview of the banking-related M&A research over the last 140
twenty years
4.9 Cumulated abnormal returns for event studies of the last 142
ten years
5.1 Descriptive statistics – sample of European listed banks 161
1996–2002
5.2 The number of listed and non-listed banks in the sample by 163
bank ownership
5.3 The relative information content performance measures of 164
European listed banks over the period 1996–2002
5.4 P-value of joint-F test among all couples of bank performance 166
measures analysed
5.5 The relative information content performance measures of 168
European listed banks over the period 1996–2002
5.6 The incremental information content of EVAbkg in European 169
banking between 1996 and 2002
5.7 Overall mean profits and EVAbkg created over the period 172
1995–2002
5.8 Annual mean levels of EVAbkg between 1995 and 2002 173
5.9 Top thirty EVAbkg creators in UK banking in 2002 177
ix
x List of Tables
5.10 Top thirty EVAbkg creators in French banking (in 2002) 181
5.11 Top thirty EVA creators in Italian banking (in 2002) 185
5.12 Top thirty EVA creators in German banking (in 2002) 189
6.1 Number of banks in samples used for estimating efficiency 202
with SFA and DEA
6.2 SFA cost X-efficiency mean levels in European banking 204
between 1995 and 2002
6.3 DEA cost efficiency mean levels in European banking 207
6.4 Alternative profit efficiency scores in European banking 220
6.5 Sample used for estimating Total Factor Productivity in 227
European banking between 1996 and 2002
6.6 Total Factor Productivity estimates in European banking 229
6.7 Mean levels of productivity changes between 1995 and 2000 in 240
European banking
6.8 Number of banks in the sample for assessing customer 245
satisfaction in European banking
6.9 Deposits and loans growth rates in European banking between 246
1997 and 2002
7.1 Relative information content of shareholder value drivers 263
analysing European publicly listed banks
7.2 Incremental information content of cost and profit efficiency 266
estimates (using SFA) analysing European listed banks between
1997 and 2002
7.3 Incremental information content of Total Factor Productivity 267
change components analysing European listed banks between
1997 and 2002
7.4 Incremental information content of cost efficiency (using DEA) 268
components analysing European listed banks between 1997
and 2002
7.5 Incremental information content of depositor and borrower 269
satisfaction estimates analysing European listed banks between
1997 and 2002
7.6 Relative information content of shareholder value drivers 271
analysing the overall European sample of listed and non-listed
banks
7.7 Incremental information content of cost efficiency (using DEA) 273
components analysing the overall European sample of listed
and non-listed banks between 1997 and 2002
7.8 Incremental information content of Total Factor Productivity 274
change components analysing overall European sample of listed
and non-listed banks between 1998 and 2002
List of Tables xi
7.9 Incremental information content of cost and profit efficiency 275

estimates (using SFA) analysing the overall European sample
of listed and non-listed banks between 1997 and 2002
7.10 Incremental information content of depositor and borrower 276
satisfaction estimates analysing the overall European sample
of listed and non-listed banks between 1999 and 2002
7.11 Relative information content of shareholder drivers using the 277
sample of listed and non-listed banks: a comparison among
countries
7.12 Number of banks in samples used for estimating efficiency 283
with SFA and DEA
7.13 Shareholder value efficiency scores in European banking 285
List of Figures
2.1 A possible implementation of the stakeholder approach 14

2.2 The relationship between economic management–efficiency– 22
effectiveness in perfect markets
2.3 Shareholder value maximisation, customer satisfaction, customer 25
loyalty and competitive position: their relationship in a
competitive market
2.4 The service value chain 28
2.5 Number of banks in European banking systems between 1985 32
and 2002
2.6 Number of branches in European banking systems 32
2.7 Estimated bank transaction costs (Euro per transaction) 33
3.1 The Porter (1980) five forces model 47
3.2 Porter’s value chain 48
3.3 ROE decomposition in banking 52
3.4 Cash flow return on investment 58
3.5 Performance measures and risk adjusted performance measures 59
3.6 Market value added and market value of capital 60
3.7 EVA spectrum 67
3.8 Assessing corporate performance using DCF 72
4.1 The creation of shareholder value: drivers, actions, strategies 96
4.2 The shareholder value endogenous strategies 97
4.3 Customer satisfaction 100
4.4 Customer satisfaction, intangible assets and shareholder value 103
4.5 A customer satisfaction model for retail banking 105
4.6 Some causes of poor customer satisfaction in commercial 106
banking
4.7 Non-interest income of European banks during the 1990s 125
4.8 Bank’s supply: diversification vs. core business strategies 126
5.1 The incremental information content: EVA and its 157
components
5.2 Independent variables: flow and stock values 159
5.3 Return on equity and EVAbkg on capital invested in UK banking 176
5.4 Return on equity and EVAbkg on capital invested in French 179
5.5 Return on equity and EVA on capital invested in Italian banking 184
xii
List of Figures xiii
5.6 Return on equity and EVA on capital invested in German 188

6.1 Inputs and outputs used in estimating cost efficiency in 197
European banking
6.2 Inputs and outputs used for estimating alternative profit 219
efficiency
6.3 A graphical example of firm productivity and efficiency 224
6.4 Inputs and outputs for estimating Total Factor Productivity 226
change in European banking between 1995 and 2002
6.5 Total Factor Productivity change over the period 1995–2002 in 236
European banking
6.6 Cumulative Total Factor Productivity estimates in European 238
6.7 The chain of customer satisfaction indicators in banking 241
6.8 The relationship between customer satisfaction and bank sales 243
7.1 Outline of the relationship between shareholder value and its 257
determinants in banking
7.2 Determinants of shareholder value analysed 258
7.3 Inputs and outputs used for estimating alternative shareholder 282
value efficiency
About the Authors
Franco Fiordelisi (MA, PhD) is Lecturer in Banking and Finance at the

Faculty of Economics ‘Federico Caffè’ of the University of Rome III, Italy. His
main research interest relates to the economics of the banking and other
financial institutions focusing on efficiency, productivity and shareholder
value creation. His research has been published by leading academic journals.
He has acted as a consultant to many banks, leasing and factoring companies.
Philip Molyneux is Professor in Banking and Finance and Director of the

Institute of European Finance at the University of Wales, Bangor. He also has
a chair at Erasmus University, Rotterdam, Netherlands. His main area of
research is on the economics of the banking and financial services industry
and he has published widely in this area including recent publications in the
Journal of Banking and Finance and Journal of Money, Credit and Banking. Recent
texts include: Efficiency in European Banking (Wiley, 1998), Private Banking
(Euromoney) and European Banking: Efficiency, Growth & Technology (Wiley,
2001). Professor Molyneux has acted as a consultant to: New York Federal
Reserve Bank, World Bank, European Commission, UK Treasury, Citibank
Private Bank, Bermuda Commercial Bank, McKinsey’s and various other
international banks and consulting firms. In November 2001 he was the
Visiting Bertill Daniellson Research Fellow at the Stockholm School of
Economics and University of Gothenburg. Professor Molyneux has been
appointed as one of eight expert financial sector advisers to the European
Union’s Economic and Monetary Affairs Committee that implements finan-
cial services regulations for the European Community.
xiv
Preface
Managing to create sustainable shareholder value is currently recognised by

academics and practitioners as a major objective by both financial and non-
financial companies. One might ask why, since the 1990s, there is such a
strong interest towards shareholder value among practitioners, academics
and, even, regulators. The primary reason for the increasing interest of
banks and other firms towards the creation of shareholder value is that
business has become more competitive and this new environment requires
a new approach to keeping both stakeholders and shareholders satisfied.
Banks have to compete with one another to obtain valuable capital
resources to invest and to get such resources they have to demonstrate to
potential suppliers of capital (shareholders) that they will generate sufficient
returns – in other words banks need to demonstrate that they can create
value for potential shareholders. As such, shareholder value maximisation has
become the primary business goal and performance target for many banks and
other companies. Although managers and practitioners have traditionally criti-
cised the notion that the sole objective in decision-making should be to max-
imise firm value, over the last decade, they have recognised that shareholder
value maximisation is at least an important priority.
Generating stable shareholder value growth requires an intense focus on
delivering benefits to customers in the most efficient way, hiring and retaining
a motivated workforce, maintaining excellent supplier relationships, and being
a good corporate citizen in each of the company’s local communities. The
so-called service value chain shows the long-term relationship between customer
satisfaction, customer loyalty, productive efficiency, employee motivation and
satisfaction and shareholder value. In a competitive market, shareholder value
is based on customer satisfaction and productive efficiency, which also requires
employees’ and managers’ satisfaction.
This text aims to provide a detailed analytical assessment of shareholder
value creation in banking. It explains what shareholder value creation is and
its main drivers. Various shareholder value measures in banking are com-
pared with conventional and other performance indicators. We empirically
examine the drivers of shareholder value in banking and illustrate empirical
relationships between factors that are believed to add to the wealth of bank
owners. The first part of the text provides a framework for analysing share-
holder value theory by discussing how shareholder value can be defined, if
it can be considered a valid strategic objective for banks, how shareholder
value can be measured and how it can be created. The second part of the text
presents various empirical investigations in order to measure shareholder
value (using the Economic Value Added approach) and some of its drivers
xv
xvi Preface
(such as cost and profit efficiency, productivity changes and customer satis-
faction). The final part analyses the importance of these drivers in creating
shareholder value and also briefly develops a new measure of bank efficiency
(shareholder value efficiency).
We believe that this text provides the most extensive coverage on share-
holder value issues in banking and that it will be of use to financial sector
practitioners (who need to manage their operations so as to generate wealth
for their shareholders) as well as academics and students interested in mea-
suring bank and firm performance from a stockholder perspective.
Special thanks go to Alessandro Cazzette, Daniele Previati, Jon Williams
and Yener Altunbas for their comments, support and advice dealing with
many areas covered in the text.
Franco Fiordelisi
University of Rome III, Rome
Philip Molyneux
University of Wales, Bangor
1
Why Study Shareholder Value Creation
in European Banking?
Introduction
This book is concerned with the analysis of the relationship between share-
holder value and its drivers in European banking markets. One might address
the following four questions:
1. Why a book on ‘shareholder value’?

2. Is it necessary to have another contribution to the shareholder value
debate?
3. Why banks?
4. Why in Europe?
With regard to the first question, shareholder value is currently one of the most
widely studied areas in finance.1 Shareholder value maximisation has become
the primary business goal and performance target for most companies.
Although managers and practitioners have traditionally criticised the notion
that the objective in decision-making should be to maximise firm value, in the
last decade, they have recognised that shareholder-value maximisation is at
least an important priority for firms. It is possible to answer the first question
by concluding that ‘shareholder value’ is an area of research well-defined,
robust and interesting for academics, regulators and practitioners.
Concerning the second question, one might observe that there is a sub-
stantial literature on shareholder value issues and, consequently, one may
question if further study on these topics may enrich the existing literature. The
book is an advance on previous studies since it focuses on some specific aspects
that formerly have been investigated by only limited analysis. First of all, the
text proposes an independent assessment of various performance indicators
(standard profitability ratios and innovative shareholder value measures) focus-
ing on the banking sector: this is valuable since most analysis on shareholder
value and its creation has been undertaken on non-financial firms. Secondly,
while many studies support a strategy to create shareholder value, only a few
1
F. Fiordelisi et al., Shareholder Value in Banking
© Franco Fiordelisi and Philip Molyneux 2006
2 Shareholder Value in Banking
contributions attempt to empirically assess the importance of different drivers

of shareholder value. For example, many studies analyse banks’ efficiency and
propose management techniques (such as the adoption of divisional share-
holder value measures, new techniques for controlling risks and so on), imply-
ing that higher levels of efficiency lead to greater shareholder value. This
relationship is generally taken for granted, but it lacks empirical support and
accurate assessment. The book intends to investigate the relationship between
shareholder value created over a period and value-drivers (such as firm
efficiency and customer satisfaction). The third question further delimits the
area of study by focusing on financial institutions and, especially, banks.
During the 1990s, the banking industry was deregulated almost everywhere
and, consequently, competition seems to have increased. Bank efficiency and
shareholder value maximisation currently are primary goals in all developed
banking markets and attract the interest of managers, academics and regula-
tors. The fourth question defines the research area by identifying a specific
banking market: Europe. Existing literature dealing with shareholder value and
firm efficiency in banking mainly analyses the US system and there is only a
limited number of studies that focus on European markets.
Motivation and key questions

The literature dealing with ‘shareholder value’ is substantial and hetero-
geneous.2 Two main streams of research dealing with this topic can be
identified: some studies have focused on management techniques and their
ability to create shareholder value; other studies have analysed what mea-
sures best describe the creation of shareholder value (formally referred to as
the value-relevance of performance measures). These two lines of investiga-
tion have not been developed independently and are often the products
of consulting companies promoting proprietary management approaches
and valuation tools. By focusing on European banking, this text aims to
present a comprehensive analysis of shareholder value by analysing if value
creation is currently a feasible goal for European banks. We also aim to
identify the most reliable measures of shareholder value for banks and
analyse which drivers create shareholder value. In evaluating these issues
the book addresses some key questions:
• What is shareholder value (SHV) and how should it be accurately

measured?
Although the concept of SHV is one of the oldest features in business, the
problem of measuring created shareholder value is still open to debate and
controversial since there is no consensus as to the best method for gauging
the value that firms create for their owners. Starting from the definition of
value of Hamilton (1777) and Marshall (1890), this book illustrates how
Why Study Shareholder Value Creation in European Banking? 3
shareholder value can be measured in a relatively straightforward manner

for banks.
• What is the level of shareholder value created in European banking?
In order to answer this question, this book presents an empirical assessment of

how shareholder value is created by banking firms in Europe. This, we believe,
is a worthwhile exercise for the following reasons: firstly, publicly available esti-
mates of shareholder value in banking are common in the US, but not in
Europe; secondly, most of these studies are undertaken by consulting compa-
nies and there are few independent and comparable estimations of shareholder
value for banks; thirdly, our analysis presents a comparative analysis of share-
holder value created in four different European banking systems (France,
Germany, Italy and the UK) between 1995 and 2002.
• What are the drivers and strategies that create shareholder value? Is it possi-
ble to assess the importance of different drivers in creating shareholder value
in banking?
Finding a strong relationship between shareholder value and its drivers

(such as efficiency or customer satisfaction) is usually taken for granted by
practitioners, but has not, as far as we are aware, been empirically investi-
gated. This text empirically assesses these relationships by analysing which
factors are important in creating shareholder value in banking.
• Is it possible to develop new measures of efficiency? What is their role in

creating shareholder value?
Banks’ efficiency is usually measured focusing on production, costs and

profits by measuring the performance of banks (i.e. minimising inputs
given the level of outputs in productive efficiency; or minimising costs
given the level of outputs in cost efficiency; or maximising profits given the
level of outputs in profit efficiency) relative to the predicted performance of
the best practice firm in the industry if these best practice firms were facing
similar market conditions. The final aim of this text is to extend the estab-
lished literature on shareholder value by proposing a new measure of bank
efficiency taking into account created shareholder value.
Structure of the text

The book is organised into eight chapters and can be divided into three main
parts. Firstly, we provide a framework for analysing shareholder value theory
by discussing how shareholder value can be defined, if it can be considered a
valid target for banks, how shareholder value can be measured and how it can
be created (Chapters 1–4). Secondly, we present various empirical investiga-
tions in order to measure shareholder value (using the Economic Value Added
(EVA)* approach) and some of its drivers (such as cost and profit efficiency,
productivity changes and customer satisfaction) in the main European banking
systems between 1995 and 2002 (Chapters 5–6). Finally, we analyse the value-
relevance of these drivers in creating shareholder value and we develop a new
measure of bank efficiency (shareholder value efficiency) and examine its value
relevance in creating shareholder value (Chapters 7–9).
Chapter 2 analyses the economic objectives of a bank showing that manag-
ing to create sustained and sustainable shareholder value is the appropriate
goal for European banks. First of all, the economic foundations of the theory of
the firm are analysed by discussing the marginalist (which proposes the neo-
classical view) and behaviouralist (which proposes the managerial capitalist view)
approaches to shareholder value. The debate between shareholder and stake-
holder models follows the debate between marginalists and behaviouralists, but
there are several arguments which lead to different conclusions. While the
behaviouralist view has demonstrated the limits of the marginalist view (and, as a
result, the maximisation of instantaneous profit can be a firm’s primary goal
only if one makes unrealistic assumptions), stakeholder models seem to be not
entirely acceptable and do not prove that shareholder value maximisation
cannot be assumed as the primary target for a firm. Although it is undeniable
that stakeholders (such as customers, shareholders and managers) have their
own interests, these objectives are not incompatible and managing to create
shareholder value is not a zero-sum game where the shareholders prosper at
the expense of other stakeholders. On the contrary, we note that generating
stable shareholder value growth requires an intense focus on delivering benefits
to customers in the most efficient way, hiring and retaining a motivated work-
force, maintaining excellent supplier relationships, and being a good corporate
citizen in each of the company’s local communities (so-called ‘stakeholder
symbiosis’). The service value chain shows the long-term relationship between
customer satisfaction, customer loyalty, productive efficiency, employee moti-
vation and satisfaction and shareholder value. In a competitive market, share-
holder value is based on customer satisfaction and productive efficiency, which
requires employees’ and managers’ satisfaction. The second part of Chapter 2
analyses the shareholder value approach in banking and discusses whether
shareholder value creation is a valid long-term goal for all types of banks.
In Chapter 3, we outline the foundations of SHV theory focusing on the four
following issues: the concept of shareholder value and wealth; the evolution of
shareholder value theory; measurement performance methodologies; and a
review of the empirical literature. Regarding the first issue, we adopt a classic
definition of value: a company creates value for the shareholders over a given
* EVA is a registered trademark of Stern Stewart & Co. Since this book covers various
areas dealing with shareholder value in banking, the more general ‘EVA’ mnemonic
is used throughout to represent the broader context of this value added term.
time period when the return on invested capital is greater than its opportunity
cost (the rate that investors could earn by investing in other securities with the
same risk). In order to highlight that this concept of ‘value’ implies a com-
parison of the shareholder return with the opportunity cost, the word ‘added’
joins the term ‘value’. Regarding the second issue, we focus on two seminal
works: Miller and Modigliani (1961) and Porter (1980). Miller and Modigliani
(1961), commonly considered to be the origin of shareholder value theory,
show the irrelevance of dividend (distribution) policies in an ‘ideal’ environ-
ment, where the value of a business is linked to two elements: (1) the present
value of expected dividends and (2) the company’s residual value. This division
is the starting point for most of the subsequent research on financial evalu-
ation methods. Porter (1980) models economic sectors and the competitors
within them and highlights the importance of strategy. The value-chain
concept developed in Porter (1980) is the key to understanding competitive
advantages and, as such, to creating sustainable shareholder value. With regard
to the third issue, we review the main techniques proposed by consultants to
measure company performance and the creation of SHV, such as: Discounted
Cash Flow; Shareholder Value Added; Cash Flow Return on Investment; Risk
Adjusted Performance Measures; Market Value Added and Economic Value
Added. Finally, the chapter presents a detailed review of methodologies and
findings of empirical studies dealing with shareholder value.
In Chapter 4, we analyse how commercial banks can create SHV by outlining
a set of strategies that appear able to improve, ceteris paribus, at least one of the
SHV drivers. We analyse these strategies by distinguishing between endo-
genous and exogenous channels (i.e. strategies implemented within the bank
and strategies involving external parties). Regarding the endogenous channels,
these relate to improvements in customer satisfaction, greater cost or profit
efficiency, the redefinition of the bank business mix and the achievement of
optimal capital structures. The external channel for potentially creating share-
holder value relates to mergers and acquisitions (M&A), joint-ventures and
other forms of financial consolidation.
Chapter 5 presents various empirical investigations into shareholder value
creation in European banking. Firstly, we undertake a comparative analysis of
traditional (such as interest and intermediation margins, return on equity,
return on assets and net income) and non-traditional (such as residual income
and economic value added) bank performance indicators in the light of
creating SHV within the European banking industry. In order to assess which
performance metric is the most compatible with shareholder value creation,
our analyses examine both the relative information content (which is useful if
one needs to select a single performance measure that best explains share-
holder value creation) and the incremental information content (which aims
to assess if a performance indicator adds information to the explanatory power
provided by other measures). We examine the performance of publicly quoted
banks and our results suggest that an EVA measure that accounts for the
specifics of banking (EVAbkg) outperforms all other performance measures since

it has the greatest ability to explain variation of Market-Adjusted Returns
(MAR). In contrast, standard Economic Value Added measures (EVAstd), typic-
ally used to examine shareholder value issues for non-financial firms, do not
seem to explain better shareholder value creation in European banking com-
pared to a range of simple accounting and other performance measures, as it
has the lowest ability to explain variations in market-adjusted returns (MAR).
The investigation of the incremental information content of EVAbkg also shows
that the treatment of banking capital and various accounting adjustments is
central when one calculates shareholder value created in banking.
In the second part of Chapter 5, we discuss levels of shareholder value
created by both publicly quoted and also non-listed European banks. We find
that European banks generate, on average, substantial and widespread profits,
but these do not seem to have created shareholder value in many cases since
profits (measured from an economic viewpoint) are often lower than the
opportunity cost of equity invested capital. This situation is not common to all
countries analysed. UK banks appear to be the most profitable and to create
higher shareholder value than those in the other three countries. German
banks exhibit the lowest profits and created shareholder value. Both empirical
analyses undertaken in Chapter 5 can be considered as ‘preliminary analyses’
since this enables us to identify a performance measure ‘suitable’ for measuring
shareholder value for non-listed banks and assesses the mean levels of created
shareholder value in European banking.
Chapter 6 presents various approaches that empirically investigate the
drivers of shareholder value. Firstly, we estimated cost efficiency by using two
frontier methodologies, i.e. Stochastic Frontier Analysis (SFA) and Data
Envelopment Analysis (DEA), defining bank inputs and outputs according to
the value-added approach (i.e. bank inputs and outputs depending on the con-
tribution of bank items, on both sides of the balance sheet, to the creation of
bank value-added). According to our SFA estimates, European banks display
over the period 1995–2002 cost X-inefficiency scores ranging between 7.3 per
cent and 38.9 per cent. On average, the mean cost X-efficiency level over the
period 1995–2002 is 78.1 per cent. UK banks display the highest cost efficiency
levels compared to banks in France, Germany and Italy. The second empirical
analysis in Chapter 6 concerns the estimation of profit efficiency. We estimate
alternative profit efficiency measures developed by Berger and Mester (1997)
and define bank inputs and outputs according to the value-added approach.
We find that, on average, banks in the four European banking systems
analysed achieved only two-thirds of their potential profits. UK banks are
found to be the most profit efficient, while French, German and Italian banks
are found to have similar levels of profit efficiency (around 65 per cent). Our
profit efficiency estimates appear to be consistent with previous studies. The
third piece of empirical analysis in Chapter 6 concerns the estimation of Total
Factor Productivity (TFP) changes. We estimate the Malmquist productivity
index using DEA for estimating distance functions. We decompose TFP

changes into technical efficiency changes, technical changes, scale efficiency
changes and pure technical efficiency. Overall, we find that TFP improved sub-
stantially in the UK (due to a greater pure technical efficiency change and a
heightened ability to exploit scale efficiency), in Italy for commercial banks
(due to technical change) and cooperative banks (due to pure technical
efficiency change), and in Germany for commercial banks (due to improved
pure technical efficiency and a reduction in input waste due to improved scale
efficiency). In contrast, we observe a reduction in TFP from 1995–2002 for
Italian savings banks and French cooperative banks generated by a negative
technological change. The final empirical analysis in Chapter 6 relates to our
analysis of bank customer satisfaction. The definition of an appropriate
measure of customer satisfaction (for our empirical investigation) is constrained
by data availability. As such, we define a proxy of customer satisfaction focus-
ing on deposits and loan growth rates relative to the industry average. We
assume that growth rates higher than the industry average are suggestive of
greater customer satisfaction (and vice versa). As deposit and loan growth rates
may vary over time for various reasons (and not only customer satisfaction),
our indicator of customer satisfaction requires various adjustments aimed at
reducing potential biases. This indicator provides a rank order measure of cus-
tomer satisfaction (from both a depositor and borrower standpoint).
Chapter 7 presents the empirical analysis of the value-relevance of three
main shareholder value drivers, i.e. customer satisfaction, cost efficiency (in all
its components, such as technical, allocative and scale efficiencies), and profit
efficiency and productivity changes (in all its components, such as technologi-
cal change, pure technical efficiency change, scale efficiency change). We assess
the information content of all these types of efficiency estimates since each of
these measures has a different economic meaning expressing a different ability
of a bank to boost its performance. Overall, we find strong evidence that
improvement in TFP and its main components leads to increases in bank
shareholder value. Cost efficiency is found to be the second most important
driver of shareholder value. Depositor and borrower satisfaction are also found
to be important in explaining shareholder value creation. Our findings are
similar for both listed and non-listed banks.
In Chapter 7, we also develop a new measure of efficiency labelled ‘share-
holder value efficiency’. A bank is defined as shareholder value efficient if it is
able to produce at the maximum possible shareholder value given a particular
level of output. Using the alternative profit efficiency estimation framework
advanced by Berger and Mester (1997), we develop a new efficiency measure
which we call ‘alternative shareholder value efficiency’. According to our esti-
mates, European banks display shareholder value inefficiency scores that
suggest, on average, banks are 36 per cent shareholder value inefficient. The sit-
uation, however, changes across years, countries and type of banks. Among the
four banking systems analysed, we found that Italian, German and UK banks
have similar shareholder value efficiency levels, while French banks are found
to generate only 58.39 per cent of their potential shareholder value. These
results are consistent with our previous findings (discussed in Chapter 5) about
the shareholder value generated by European banks.
Once we estimated shareholder value efficiency, the second part of Chapter 8
analyses the value-relevance of this new efficiency measure by focusing on
both relative and incremental information content. We apply a methodology
similar to that adopted in Chapter 7 to analyse the explanatory power of the
determinants of shareholder value. We use two samples: one comprises only
publicly listed banks and the second both listed and non-listed banks. Our
findings provide strong evidence that there is a positive relationship between
shareholder value efficiency and bank shareholder value. In other words, a
bank can generate value for its shareholders when this bank is able to reduce
the distance between best-practice banks in creating value for their share-
holders. We found that shareholder value efficiency has higher relative infor-
mation content than for cost X-efficiency and alternative profit efficiency. This
superiority is very small for listed banks, while it is substantial for non-listed
banks. As such, shareholder value efficiency seems to be particularly useful for
non-listed banks for investigating the ability of these banks to create value for
their shareholders/stakeholders.
Chapter 8 presents the conclusions.
2
Economic Objectives of Banks
Introduction
This chapter analyses the economic objectives of banks. As explained
in Chapter 1, managing to create sustained and sustainable share-
holder value is currently one of the main objectives in European bank-
ing. Although this view appears popular among managers and academics,
it cannot be taken for granted and it may not be accepted under
some circumstances. Shareholder-value maximisation implies an increase
of dividends and/or of share prices: the general view that managers
primarily concentrate on these targets can be subject to various
criticisms.
Banks are complex systems with several actors, having an own-interest,1 such
as:
• Shareholders who are interested in maximising the value of their

holdings by obtaining high dividends and/or increasing share prices.
However, this might not be an entirely accurate depiction. For ex-
ample, majority and minority shareholders may have different holding
periods and, consequently, minority shareholders are interested in
maximising their value over the shorter term, while majority share-
holders may prefer increased dividends and share prices over the longer
term (even sacrificing value maximisation in the short term).
• Board members may pursue their own goals (for example, the growth of
corporate size or the number of shareholders in order to reduce shareholder
power and increase their own power).
• Managers may pursue their own goals (for example, risk aversion and
maximising their pay and other expenses – expense preferencing), that may
differ from shareholders’ and CEOs’ goals.2
• Customers, who covet high-quality services (such as brokerage, payment
services etc.) at low prices. Among customers, it is necessary to distinguish
between:
9
(i) Funds-suppliers, such as depositors, who wish to receive high interest

and good payment services.
(ii) Funds-takers, such as borrowers, who aim to pay low interest on
their loans and obtain credit facilities easily.
• Suppliers, who prize steady sales. Among banks’ suppliers, the trade unions
are particularly important. The unions represent the bank’s workforce
(employees) and their objectives include high salaries and social security.
• Government, which levies taxes and requires that all corporate activities be
legal.
• Supervisory authorities, who are responsible for banking sector soundness.
• Other stakeholders, such as consumer associations. These associations may
play an important role, as demonstrated recently in Italy. In 2000, a
consumer association which had taken legal proceedings against the
Associazione Bancaria Italiana, obtained a ruling from the court that the
interest rates on outstanding mortgages (prior to 2000) be renegotiated
according to the new anti-usury law (which caps the maximum interest rate
charged).
Although this list of stakeholders may be further extended, it is sufficient to

show that the economic objective of any corporation cannot be defined
disregarding the interests of all stakeholders.
Economic foundations of the theory of the firm

In order to shed light on the shareholder-value approach, it is possible to iden-
tify within the economics literature two schools of thought providing different
sets of economic objectives for firms to pursue: the marginalists and the behavi-
ouralists.3 According to the marginalists (i.e. the neoclassical view), the ultimate
goal of a firm is to maximise profits. In contrast, the behaviouralists maintain
that a firm has a wider range of objectives.
Neoclassical vs. the managerial capitalist view

The neoclassical view assumes that firms: (1) have a limited size; (2) operate in
a perfectly competitive market and in a fully rational context; (3) have perfect
knowledge of market supply and demand; and (4) work in a regime charac-
terised by the perfect mobility and divisibility of productive inputs, where
frictions are absent and conditions are static.
Under these assumptions, a firm has to take two decisions: (1) selecting an
optimal production function; and (2) defining the optimal quantity for output.
Both problems are solved in order to maximise the instantaneous profit (i.e.
the orthodox objective function supported by the neoclassical view): as a con-
sequence, the optimal production function is the function which allows the
firm to produce with the lowest average cost per unit (given output quantity)
and the optimal output quantity is that corresponding to the minimum point
Economic Objectives of Banks 11
of the average cost per unit curve. The neoclassical approach has been dis-
cussed in several studies.4 Without entering into this debate, it is sufficient to
note that this view of the firm is determined by the constraints imposed by
the Walrasian system of general economic equilibrium.5 In other words, neo-
classical theorists do not concern themselves with microeconomic reality, but
define the firm’s behaviour according to general equilibrium conditions.
Following the seminal paper of Berle and Means (1932), Hall and Hitch
(1939) demonstrated that there are substantial differences between the neo-
classical view and reality. The first observation is that large firms are managed
by subjects distinct from the owners and this begs the question as to the extent
to which managers pursue owners’ objectives. Also, to what extent do man-
agers’ decisions produce a deviation from the neoclassical goal (i.e. profit
maximisation in the short run)?
The behaviouralist approach (i.e. the managerial capitalist view) rests on the
following assumptions:
(1) Firms are managed by two groups: managers and owners.

(2) Owners aim to maximise firms’ profits while managers aim to maximise
their own utility functions.
(3) Owners do not have the opportunity to directly manage firms (especially
large firms) and they have to delegate to managers. In addition, owners
cannot continuously control managers’ work and therefore they intervene
in managerial decision-making infrequently. As such, managers have dis-
cretion in managing the daily activities of the firm and this power is
curbed only for those decisions that clearly contlict with the owners’ inter-
ests.
(4) There cannot be a systematic clash between the managers’ utility function
and the owners’ goals: if there were, the managers’ behaviour might
damage owners or, at least, not fully pursue the owners’ objectives.
(5) Managers often behave in ways not consistent with profit maximisation.
Perfect competition is not assumed (in that case, profit maximisation
would be the only condition to ensure firm survival) and managers have
a margin of discretion. It is worthwhile noting that managers’ margin of
discretion is explained by removing the assumption of perfect competition
in the market.
Although the managerial capitalist views differ substantially on a wide

range of issues, all of them postulate the existence of a heterodox objective
function by recognising that a firm can have objectives other than the
maximisation of its instantaneous profit. Among the relevant studies, it is
interesting to summarise the following:
• Gordon (1945) identified several groups with an active role in the

firm’s management (i.e. shareholders, bondholders, bank lenders, regular
customers, board members, top and middle managers), and different levels
of management decisions (e.g. start-up decisions, management decisions,
restructuring decisions). Managers usually hold the leadership function and
aim to maximise their own power by increasing the number of employees
and the amount of resources managed.
• Monsen and Downs (1965) reached similar conclusions by showing that
CEOs and top managers tend to maximise their personal incomes over their
expected life. This goal is achieved by engaging in safe activities (i.e. opera-
tions with a low risk).
• Rothschild (1947) showed that managers’ primary interest is the corpora-
tion’s survival in the long run.
• Williamson (1963, 1964) makes another interesting point: managers usually
use their margin of discretion to undertake expense-preference activities.
They prefer to engage in specific activities by sustaining a level of expenses
which is higher than the level justifiable when the firm’s objective is profit
maximisation.
• Baumol (1959) stressed the importance of sales. Managers are interested in
maximising firm sales because this variable expresses the corporate size, and
managers’ incentives, status and prestige are correlated to sales.
• Marris (1964) presented a model based on a utility function composed of
the following elements: (1) the shareholders’ goal, which is valuation ratio6
maximisation; (2) the managers’ goal, which is the maximisation of the
firm’s growth rate. According to Marris (1964), managers attempt to max-
imise the firm’s growth rate by assuring a minimum acceptable valuation
ratio, in order to defend their power from internal threats (i.e. shareholders
may sell their shares or try to dismiss board members and top managers if
corporate performance is unsatisfactory) and external threats (i.e. takeover).
Debate about rationality

The debate about the validity of the neoclassical view has focused on another
issue as well: rationality. This approach implicitly assumes that firms’ decision
processes take place under conditions of perfect rationality.7 An objectively
rational decision model can only be assumed if the decision-maker knows the
firm’s goals, resources available and to what extent the employment of each
resource contributes to the achievement of the goals.
The neoclassical company has a ‘maximising’ (or optimising) behaviour.
According to Machlup (1946) marginalism is the logical process aiming to find a
maximum; according to March and Simon (1958), a maximising behaviour
requires three fundamental conditions: (1) all the possible alternative actions
are known (or knowable); (2) all the consequences of these actions are known;
(3) the decision-maker has a utility function, which enables him to order
his options in terms of his preferences. The neoclassical firm clearly matches
these three assumptions. However, the rationality assumption has been cri-
ticised: when market competition is not perfect (perfect competition is a too

simplified view of reality), the decision-maker does not have all the know-
ledge required and, as a consequence, perfect rationality (implied by the
neoclassical view) cannot be adopted as an appropriate model to interpret
corporate behaviour.8
The behaviouralist view is based on the weakness of the perfect rationality
assumption. As noted by Cyert and March (1963), the perfect rationality of a
decision-maker is based on ‘too rigid’ a set of assumptions, which are nowhere
to be found in reality. Firstly, it is difficult to build a utility function when
there is more than one decision-maker and more than one objective (which,
for example, may conflict with each other). Secondly, it is impossible (or it is at
least too expensive) to know all the possible alternative actions and their
effects. Thirdly, the decision-maker essentially uses information which is not
objective, having been obtained through psychological and sociological
processes.
These observations raise doubts about the maximising behaviour as the
objective of the firm. As a rule, behaviouralists do not aim to identify the
most appropriate goal for a firm (i.e. the most fitting goal given market
conditions), but they focus on the decision process within the firm. In a
company, in fact, all the stakeholders ‘negotiate’ a mix of goals, which can
be seen as the process among them, where cash flows and political involve-
ment determine the outcome. As a result, the decision-making process is
undertaken by assigning sub-objectives to all parts or business units within
the firm. These business units can assess a limited number of actions and
select the action which best satisfies the decision-maker’s expectations. The
decision-making process consequently becomes an iterative process and
its output (i.e. the firm’s objective) is not a maximising behaviour, but a
satisfactory behaviour (i.e. the most easily achievable).
It is worthwhile mentioning a final point. According to the behaviouralist
view, a firm aims to avoid friction among the stakeholders and tends: (1) to
supply a level of factor compensation which is higher than necessary to ensure
its survival; and (2) keeps a higher stock of resources assignable to stakeholders
in order to face the variability of external factors. These excesses make even less
realistic the maximising behaviour assumed by the neoclassical view.
Stakeholder models
Following the marginalist view, the stakeholder approach assumes that a
firm is responsible to all stakeholders and that there are other corporate
objectives beyond shareholder-value maximisation. The stakeholder model
attempts to balance the interests of everyone with a stake in the company.
This approach aims to satisfy the needs of all stakeholders, individuals
and entities that may influence (and/or may also be affected by) the firm.
The management of a business requires the ‘balancing of the interests of
stakeholders’: as a consequence, the stakeholder approach is a subjective

process and is not connected to a specific methodology. In the literature,
however, it is possible to find different stakeholder models, which em-
phasise the interests of different specific groups (such as managers or
customers).
A basic stakeholder framework

According to the main principle of the stakeholder model, companies are
responsible to all stakeholders, and managing solely for shareholder value
creation will not fulfil that responsibility. The stakeholder model may involve
two different approaches: (1) involving stakeholders in the decision processes
(decisions are taken jointly in order to ensure that various interests are taken
into account); (2) setting specific targets for each group of stakeholders.
Figure 2.1: A possible implementation of the stakeholder approach
Number
Goals
Identification of
the stakeholders
Interdependencies
Threats and opportunities

Relevance of the
Experience
stakeholders
National/
international impact
Stakeholders' involvement
Information in the decision processes
Proactive measure vis-à-vis
the stakeholders
Board
Social involvement representation
Mapping of the
Aligning all management management processes
processes with the model
Re-engineering of the
management processes
Acceptance management
Reaction in case of
real problems Stakeholders' dialogue
Learning process
Source: adapted from Schuster (2000: 5).

These approaches appear to be complementary, rather than alternatives, and

both are instrumental to the stakeholder approach, which may be developed
through the following phases as shown in Figure 2.1:
• Definition of the group of stakeholders. In this phase, it is necessary to identify

a number of people, organisations and entities which hold a stake in
the company. Secondly, their interests have to be recognised. Thirdly, it is
opportune to acknowledge any interdependency among goals and needs.
It is worthwhile to note that the stakeholders’ objectives may be aligned
with shareholders’ goals: in any case, the interests of the stakeholders are
considered to be ‘at least of equal importance’ as the shareholders’.
• Assessment of stakeholders’ links to the company. The second step requires the
assessment of the stakeholders’ power on the firm. First of all, it is necessary
to measure this power in terms of opportunities (and threats) deriving from
the satisfaction (or dissatisfaction) of each specific group. In addition, the
stakeholders’ power assessment should also consider the experience of each
group and the national and international impact that the adoption of each
goal carries for the company.
• Proactive measures vis-à-vis the stakeholders. In this phase, stakeholders are
involved in the decision process. This may include, for example, a board
representation for some groups of stakeholders, according to the results of
the analysis carried out in the previous step. As a rule, stakeholder involve-
ment in the business is a subjective process and depends on the type of rela-
tionship, the communication and information channels and the company’s
social involvement.
• Aligning all management processes with the model. In this phase, management
processes are mapped and it is necessary to establish a stakeholder culture
and align management processes to the stakeholders’ goals.
• Reaction in case of real problems. This model is a learning process. If any prob-
lems arise during the implementation phase, it is necessary to plan some
corrective actions. This needs to be fully accepted by management (which
must be flexible and, if necessary, willing to modify its goals) and requires a
good dialogue with all stakeholders.
As a result, the stakeholder approach requires firms to conduct periodical

surveys to learn the needs of all stakeholders, setting objectives and measuring
results. Companies also need to publish externally audited reports on how the
interests of different stakeholders are being fulfilled and regularly use focus
groups to assess how well stakeholders’ needs are being met.
Social responsibility model

Since the 1960s, environmentalists, social activists and consumer advocates
have affirmed that companies have a ‘social responsibility’ and have to operate
in accordance with public interest, rather than ‘only’ aiming at shareholder-
value maximisation. The importance of banks’ social responsibility has

recently been highlighted in the case of the failure of Parmalat when bond-
holders argued that banks colluded with Parmalat against the interest of
investors.
Although the concept of the ‘social responsibility’ of the firm changes
according to various environments, all advocates of social responsibility affirm
that companies have a purpose beyond shareholder-value maximisation.
Ambiguity and the lack of enforceability with regard to meeting social
objectives are the two major difficulties connected with this view. As a result,
the social responsibility model currently receives only limited support from
policy-makers and corporate governance activists, although one should note
that social responsibility is a part of many firms’ overall objectives.
The ‘untouchable’ management model

The expression ‘untouchable management’, developed at the end of the
nineteenth century,9 refers to commonly observed behaviour10 which is likely
to occur under the following conditions: (1) managers, like other people, act in
their own self-interest: as a consequence, in some situations, managers’ goals
differ from those of the firm’s shareholders; and (2) in large companies, there is
a separation between ownership and management. Managers naturally have a
dominant position over the board, where they are often heavily represented,
either personally or through colleagues from other companies.
In order to shed further light on this problem, it is useful to consider the
analysis carried out by Marris (1964), who presented a model based on a
utility function composed of the following elements: (1) the shareholders’
goal, which is the maximisation of the valuation ratio (i.e. share price to
total asset per share); and (2) the managers’ goal, which is the maximisation
of the firm’s growth rate. The firm’s growth is a key objective for managers
because it enables them to increase their power: the managers’ interest in
achieving this objective can be so strong that growth initiatives with returns
below the cost of capital may be undertaken. In addition, this helps to avoid
becoming a target of takeovers.
Efforts to maximise a firm’s growth are constrained by managers’ con-
cerns about preserving their jobs. In fact, a growth rate that is too low and
generates poor profits may induce shareholders either to sell their shares or
to dismiss board members and top managers. It is therefore necessary to
identify the relationship between the firm’s growth rate and profits.
(According to Marris (1964) this function increases for an initial interval
and then decreases.)
In addition, it is necessary to assess the relationship between the firm’s
growth and the sources of finance employed by managers to fund this growth.
If the strategy aims to use only internal financing, the firm’s growth rate may
not be superior to the profit growth rate: in this case, maximisation of the
firm’s growth rate requires profit maximisation. However, when firm growth is
financed by external sources (a more realistic assumption), there is a trade-off

between managers’ and shareholders’ goals. In this case, the funds’ supply
depends on the profit rate and the valuation ratio.
The relationship among these variables can be explained as follows:
(1) Managers (in order to maximise the firm’s growth rate) increase the
percentage of corporate profit destined for future self-financing.
(2) The accounting value of the assets increases and the valuation ratio
decreases since dividends decrease. There is therefore a trade-off between
the firm’s growth rate and the valuation ratio.
(3) If the valuation ratio is too low, there is a gap between the firm’s market
value and its accounting value and the possibility of a takeover bid
increases.
(4) Managers (who are worried by the possibility of takeover) attempt to
maximise the firm’s growth rate, by assuring a minimum acceptable
valuation ratio, in order to defend their power from internal threats
(e.g. shareholders may sell their shares or try to dismiss board members
and top managers if the corporate performance is unsatisfactory) and
external threats (e.g. hostile takeovers).
The ‘untouchable management’ model has recently received increasing public

criticism, mainly because of the perceived greed of some board members. In
addition, compensation plans in some companies have been considered
excessive and not adequately connected with firms’ performance. From a
shareholders’ perspective, ‘untouchable management’ has a negative impact
on share prices. Lack of transparency and of clearly stated objectives can
create extra risks for investors, and some of the initiatives to prevent takeovers
seem to have a negative influence on share prices. Companies that have
pursued growth below the cost of capital may be criticised for not distributing
capital to shareholders.
Customer value approach

In recent years, the concept of customer value (or customer satisfaction)
has grown in importance among academics and practitioners. According to
this approach, the firm’s objective is customer satisfaction, which is
achieved when the customers’ perceptions (about the products and/or
services received) are aligned with customer expectations.
The supremacy of customers among all stakeholders is captured by a
simple motto: ‘the customer is king’. According to this view, a company
produces ‘value’ only when a customer judges the products and the services
received to meet his or her expectations. When this condition is met, cus-
tomers become loyal and are even willing to pay higher prices. When
customers feel dissatisfied, they can be easily tempted by other competitors:
even a sound company can lose its customers and become unsuccessful and
unprofitable if it fails to meet customer expectations of its products and

services. For example, Cleland and Bono (1997) argue that one underlying
difficulty for companies stems from their inability to precisely identify and
meet their customers’ shifting needs. As a result, customer satisfaction is the
primary target for every firm and customers are considered as the most
important group of stakeholders. As such, banks need to engage a wide
range of resources to achieve customer satisfaction.
The CEO and top managers need to operate their business so as to meet
customer perceptions of their products and services. Customers usually
have their own (subjective) perception of the products/services received,
which may not entirely reflect the true characteristics of these products/
services. This gap can be due to several factors, such as the bank’s image,
the active role of the consumer in the production of the service (i.e. pro-
sumer11) and the poor quality of the services provided by the intermedi-
aries employed by the bank to deliver its services. For example, customer
dissatisfaction with e-banking services may be due to the company chosen
by the bank to design and manage the website, rather than the bank itself.
As such, the CEO should aim to improve the bank’s brand, select excellent
IT service providers and accurately design the bank’s productive and dis-
tributive processes so as to meet service quality and customer satisfaction
requirements. Furthermore, the CEO and top managers need to have a
strong commitment to managing managers and employees as described
below.
Bank managers have to understand customers’ expectations. Customers’
expectations are determined by several factors: past experience, cross-customer
communications, customers’ needs, quality awards or International Standard
Organisation (ISO) certification. An imperfect understanding of customers’
expectations is the first source of customer dissatisfaction. At the same time,
bank managers have to define the standards of the bank’s products/services
according to their comprehension of the customers’ needs. Once managers
have comprehended customers’ needs, they consequently define the stand-
ards for the bank’s products and services: however, product and service design
may not reflect perfectly the managers’ comprehension. This is the second
source of customer dissatisfaction.
Bank employees have to supply the bank’s services conforming to the
chosen standards. Although this is usually taken for granted, it often
happens that the products (and, especially, services) supplied differ from
the products and services as designed. Several factors may generate this
gap, such as inadequate management of human resources (e.g. unclear
definition of tasks or insufficient training). This gap is the third source of
customer dissatisfaction.
When one of the above-mentioned activities is not accurately carried
out, customers will be unsatisfied since customer satisfaction achievement
involves all the human, technical and productive processes of the bank.
Shareholder-value approach
Although the shareholder-value approach has been described in several ways, it
can be defined simply as ‘a management approach aiming to create long-term
value for the shareholder’.12 According to the advocates of this approach,
shareholders are the most important stakeholders in a company and their satis-
faction should consequently be the primary strategic objective of a company.
In perfect markets, all shareholders agree on an optimal investment policy –
that is, they choose the project with the highest net present value. Shareholder
value is therefore assessed in terms of dividends and share price increases.
However, when the market is not perfect (e.g. in the presence of asymmetric
information and transaction costs), shareholder unanimity is compromised. As
noted in Loderer and Zgranggen (1999: 91), ‘when some shareholders have
claims against the firm in addition to their shares … the same policy decision
can increase the wealth of one group of shareholders while reducing the wealth
of another. One group of shareholders might even be tempted to appropriate
wealth from other groups. Shareholder disagreement represents a situation that
management can use to serve its own ends – for example, to fend off an
unwanted takeover.’ Despite these problems, shareholder-value maximisation
is largely recognised as the most accurate representation of shareholder interest
when market imperfection reduces (e.g. transaction costs decrease and in-
formation increases).13
Limits of stakeholder models

Although the debate between shareholder and stakeholder models follows the
pattern of the debate among marginalists and behaviouralists, there are several
differences which lead to different conclusions. While the behaviouralist view
has demonstrated the limits of the marginalist view (and, as a result, the
maximisation of instantaneous profit can be a firm’s primary goal only under
unrealistic assumptions), the case made in favour of the stakeholder models is
not entirely acceptable and does not reflect the view that shareholder-value
maximisation can be the primary strategic objective for a firm.
Limits of the social responsibility model

Although the arguments posed by the promoters of social responsibility
models (which focus, for example, on the educational role of companies or
on their commitment to environmental protection) are challenging, it
should be noted that corporate managers do not have the political legit-
imacy or the expertise to define the social goal of the firm. In a democratic
political system, elected legislators (rather than corporate managers) are
entrusted with the collective choice mechanisms. As noted by Rappaport
(1998: 5), ‘ironically, costs that social responsibility advocates would
impose on corporations often are costs that voters through the political
process would be unwilling to bear’.
Fortunately, several market incentives make shareholder-value maximisa-

tion compatible with the adoption of socially responsible corporate strat-
egies. It could reasonably be argued that the leading social responsibility
of a firm in a market-based economy is to create shareholder value by
operating legally and with integrity.
Limits of the management model

Concerning management models, it is certainly incontestable that managers
have their own self-interest (which can differ from shareholders’ objectives)
and there can be a separation between ownership and management (and man-
agers have a strong position), as stakeholder advocates maintain. Habermayer
(1997: 497), for instance, analyses the extent to which managers are com-
mitted to shareholders’ goals and notes that the percentage of managers
aligned to shareholder interests is substantially higher in the UK and US
(76 per cent and 70 per cent, respectively) than in France, Germany and Japan
(22 per cent, 18 per cent and 3 per cent, respectively). However, it is neverthe-
less plausible that managers’ objectives converge with that of shareholders’ for
(at least) four reasons:14
(1) Stock ownership by management leads executives’ interest towards share-

holder objectives and, as a result, it motivates executives to act consistently
with shareholder-value maximisation. On theoretical grounds, one may
expect the management’s shareholder orientation to be directly correlated
to the portion of personal wealth invested in the company stock or tied to
stock options. However, these arguments have two main limitations: (a)
the convergence between managers’ and shareholders’ objectives may be
effective for executives (who often hold a relatively large participation in
the firm), but this is probably not apparent for divisional and business unit
managers (who do not usually receive shares as incentives); (b) although
stock ownership by management increases their shareholder orientation,
managers and shareholders still have different risk tolerance levels. While
shareholders have the opportunity to diversify their portfolio, managers’
compensation derives from the performance of their business unit, divi-
sion or company (according to their position in the firm). In other words,
while shareholders can balance the risks arising from investment in a risky
project against other investments in other companies, managers can only
balance a project failure against the other activities of the division or the
company. As a consequence, it is rational to expect their risk tolerance to
be lower than that of shareholders.
(2) Probably the most efficient measure for aligning managers’ interest
with the shareholder’ objective is to realise an ‘adequate’ incentive
system which links management compensation to shareholder-value
maximisation. The first possibility is to link managers’ compensation to
the market returns realised by shareholders. Although this is the most
direct system, it faces several limits such as the fact that share prices are
influenced by several factors beyond the control of managers. For
example, managers may believe that shareholder returns are influenced
by excessively optimistic or pessimistic market expectations and in this
case it is hard to link the performance of divisional and business unit
managers to share prices. This is a particularly serious drawback as the
objective of shareholder-value maximisation can and should guide
the decision-making process at all levels of the organisation. Several
companies (and banks) have found stock-based compensation systems
inadequate.15 For example, Knight (1997: 221) notes that ‘Stock-
option-based performance is largely based on changes driven by the
stock market and has little to do with operating decisions. Managers
are left with no clear link between what they do and the results that
determine their pay.’ In order to face these problems, several share-
holder-value measures, such as Economic Value Added (EVA), have
been developed.16 These shareholder-value measures can be applied to
the whole company (and measure the ‘overall’ shareholder value
created by the business), but also to every business unit within the
company (and measure the value created for shareholders by each busi-
ness unit). By measuring shareholder-value maximisation with these
measures and, subsequently, linking the managers’ compensation to
these measures, all managers throughout the organisation are given a
clear and comprehensive objective and their interests tend to be better
aligned with those of shareholders.17
(3) The threat of takeover is another instrument for increasing the share-
holder orientation of managers by constraining those who might
choose to pursue personal goals at the expense of shareholders. In fact,
as market efficiency increases, any abuse or impropriety is reflected in a
lower stock price. If the market share price declines, the firm becomes
an attractive takeover target. If takeover occurs, new owners are likely
to replace senior management. As a consequence, the threat of takeover
in efficient markets tends to reduce the divergence of interests between
managers and shareholders. One might note that Marris (1964), who
originally considered how a takeover threat affects managers’ behav-
iour, concluded that managers attempt to maximise the firm’s growth
rate, rather than converge to the shareholders’ objective. This conclu-
sion is debatable nowadays as the maximisation of the firm’s growth
rate may be problematic in increasingly competitive and open markets.
In these circumstances, we would argue that executives are likely
to minimise the likelihood of takeover by maximising shareholder
value and, in this way, shares will not be undervalued, shareholders
will be satisfied and less open to any bid and, even if a takeover
occurs, management replacement is not necessarily the most probable
option.
(4) Also, the presence of a competitive labour market for executives tends
to reduce the divergence of interests between managers and share-
holders. In fact, when the labour market for executives is competitive,
managers compete for places on the basis of their ability to achieve
results. As the competitiveness of this market increases, the motivation of
executives to act according to shareholders’ objectives increases.
Limits of the customer value approach

The shareholder-value approach incorporates most of the views advanced by
the advocates of the customer value approach. As noted by Rappaport (1998:
8), ‘even the most persistent advocate of shareholder value understands that
without customer value there can be no shareholder value … On the other
hand, providing customer satisfaction does not automatically translate into
shareholder value.’18 In other words, customer satisfaction is a necessary but
not sufficient condition for shareholder-value maximisation. This relationship
reflects a basic management principle in a perfect market as every business unit
(independently of the stakeholders’ and shareholders’ goals) should be
managed in an economic manner. The concept of ‘economic management’19
implies that every firm has to operate ‘efficiently’ and ‘effectively’, as shown in
Figure 2.2. Productive efficiency requires the firm to be able to maximise its
income, given a certain cost level, or minimise its costs, given a certain income
level; effectiveness is the firm’s ability to achieve goals. In a competitive market,
customer satisfaction expresses the ‘effectiveness’ of the management.
Business management
principle
SHAREHOLDER
VALUE
ECONOMIC
MANAGEMENT
EFFICIENCY EFFECTIVENESS
PRODUCTIVE CUSTOMER
EFFICIENCY SATISFACTION
Translation in a
competitive market
Figure 2.2: The relationship between economic management–efficiency–effectiveness

in perfect markets
One may have some doubts about the above view and the following two
questions can be advanced: (1) Is it always true that customer satisfaction is a
necessary but not sufficient condition for shareholder-value maximisation? Or,
in other words, under what conditions is this relationship observed? (2) While
it is straightforward that efficiency and effectiveness are compatible objectives
and that they jointly lead to an ‘economic management’, are productive
efficiency and customer satisfaction conflicting goals ?
Concerning the first question, one might note that, as market conditions
tend to oligopoly and monopoly, customers have fewer opportunities to
acquire products and services from different competitors (because either the
numbers of companies in the market decrease or there are high transaction
costs and little information) and unsatisfied customers cannot ‘migrate’ to
other companies. In these market conditions, the firm’s success may be less
influenced by customer satisfaction and the business focus will tend to be on
productive efficiency. In contrast, as the market tends to perfect competition,
customers have a greater opportunity to acquire products and services from
different competitors without sustaining any costs (i.e. migration costs) and
unsatisfied customers have therefore the chance to choose other companies.
On the other hand, a firm’s ability to achieve customer satisfaction is insuf-
ficient (although it is necessary) to guarantee its success and customer satisfac-
tion can be an extremely costly goal if it is achieved without accounting for
productive efficiency. As such, customer satisfaction is in itself insufficient to
guarantee an ‘economic’ management and customer satisfaction seems to be a
necessary, but not sufficient, condition for shareholder-value maximisation
when a company operates in a perfect market (i.e. a perfectly competitive
market, where information is perfectly available, there are no transaction costs,
etc.). It is worthwhile noting that perfect markets can be viewed mainly as a
theoretical case. In reality, most markets (such as banking) can be seen as
‘quasi-perfect markets’, being competitive, with low transaction costs and with
information largely available. Under these conditions the relationship between
customer satisfaction, productive efficiency and shareholder value is likely to
be observed in the medium to long term. Although there may be migration
costs, unsatisfied customers may not have the opportunity to leave ‘imme-
diately’ all those companies unable to satisfy their expectations (these cus-
tomers are usually described as ‘hostage’), but it is reasonable to assume that
they will leave these companies in the medium to long term (the length of this
period depends on the industry, on the number of products and services
acquired and on switching costs).
Concerning the second question, at first glance, one might certainly view
productive efficiency and customer satisfaction as conflicting goals. Customer
satisfaction requires a stronger focus on customers and this implies increasing
research and design expenses, higher production costs and a more sophist-
icated customer service. However, one can argue that there is no substantial
trade-off because improving quality and customer satisfaction reduces costs
associated with defective goods and services, such as recovery costs, warranty
costs, reworking/replacing defective goods and services, etc. In addition,
customer satisfaction is usually achieved by implementing specific strategies
(such as Total Quality Management (TQM)) and by re-engineering business
processes. In this way, companies are able to satisfy their customers and, at the
same time, to minimise inefficiencies. In other words, customer satisfaction is
achieved by rethinking the whole production and distribution process and
eliminating all the activities which do not add customer value (e.g. waste).
Finally, customer satisfaction does not necessarily imply higher cost, but it
should bring higher profits by increasing customer loyalty (repeat purchasing)
and improved competitive position.
The above discussion has shown how productive efficiency and customer
satisfaction are not necessarily conflicting goals, and why they need to be
jointly achieved in a competitive market. It is now clear why shareholder-value
maximisation requires customer satisfaction, even though this still cannot be
the ultimate goal for companies.
The shareholder value approach: a non zero-sum game

Although the stakeholder models discussed cannot be fully accepted in a
perfectly competitive market environment, the previous remarks have shown
that stakeholders (such as customers, shareholders and managers) clearly have
specific interests. This section aims to show that these objectives are not
incompatible (as it may appear at first sight), and managing to create share-
holder value is not a zero-sum game where shareholders prosper at the expense
of other stakeholders.
In order to shed light on the dispute among stakeholders, we will start by
looking at a historical event. The conflict among stakeholders appears to be
similar to the plebeians’ secession against the patricians in 494 BC during
the Roman Empire. Roman society was organised as follows: the working
class (i.e. plebeians) were to serve and obey the patricians (rich and noble
class), who took them under their patronage. In 494 BC the plebeians
rebelled against the patricians by gathering on the Aventino (one of Rome’s
seven hills) and refusing to work. Menenio Agrippa solved the dispute with
his famous apologue. He implicitly affirmed the equality between the two
classes by comparing Roman society to the human body, where the patri-
cians were the stomach and the plebeians the hands. Agrippa affirmed that,
although it is irrefutable that the hands have the most tiring job (i.e. they
continuously work to feed the stomach) and that the stomach has the
simplest task (i.e. to digest food), the stomach is indispensable because it
supplies the nourishment for every part of the body (hands included).
The dispute between plebeians and patricians is similar to the dispute
between customers and shareholders. A ‘perfect market’ can in fact to be com-
pared to a ‘perfect body’ – both are composed of several parts, whose life
depends on the work of the other parts. Following this view, customers play
the same role of the stomach in the human body (the patricians in ancient
Rome): they take advantage of the work (i.e. products and services received)
done by the rest of the body (i.e. companies) and supply the nourishment (i.e.
sale proceeds) for the survival of the rest of the body (i.e. companies).
Companies can be considered as the rest of the body (the plebeians in ancient
Rome): they work to make products and survive by receiving nourishment
from customers. It is worthwhile to note that: (1) the body (i.e. companies) can
only survive by working and, therefore, feeding the stomach (i.e. customers);
(2) not all food (i.e. products and services) supplied to the stomach (i.e. cus-
tomers) is suitable to produce nourishment. It is therefore necessary to under-
stand which food is more energetic (i.e. profitable); (3) the human body (i.e.
companies) expends energy during its activities: the nourishment for the body
increases when the body is able to perform more efficiently.
As such, customer satisfaction and shareholder-value maximisation are two
compatible and dependent goals in a competitive market. Customer satis-
faction is fundamental for shareholder-value maximisation since it increases
customer loyalty and improves the competitive position as illustrated in
Figure 2.3.
To outline the relationship between customer satisfaction and shareholder
value in more detail:
(a) Once satisfied, customers tend to be more loyal and profitable. The rela-
tionship between customer loyalty and profits is commonly accepted in
the literature; for example, Reichheld and Sasser (1990) found (collecting
factual experiences from a number of organisations) that customer loyalty
is the factor most often associated with higher profits. This relation-
ship (customer satisfaction → customer loyalty → shareholder value) is
explained by the fact that as expectations are established and met, cus-
tomers gain knowledge of the service delivery process, thus reducing future
marketing and transaction costs and existing customers become easier to
serve and a lower marketing effort is required. In addition, customer satis-
faction helps to secure future revenues making customers less sensitive to
• reducing future marketing and transaction costs for

Customer existing customers
loyalty • secure future revenues
• reduce price elasticity
• increases margin on sales to existing customers
Customer Shareholder
satisfaction value
• improving the bank's image
• sustaining lower marketing costs
Competitive • increasing its volume of activities
position • finances processes to monitor customer satisfaction
and to continuously improve the quality of the
banking services
Figure 2.3: Shareholder value maximisation, customer satisfaction, customer loyalty

and competitive position: their relationship in a competitive market
commercial offers from other competitors. In this sense, customer satisfac-

tion reduces the fluctuation of future income. Customer satisfaction also
reduces price elasticity. Several studies20 have found that in long-term
commercial relationships (such as in banking), customers are willing to pay
higher commissions in order to receive higher quality services. Under these
circumstances, if customers are satisfied, a bank may well increase the price
level before customers consider the possibility of moving to another bank
so customer satisfaction also increases margins on sales to existing cus-
tomers. As customer loyalty increases, companies can more concretely
exploit opportunities for cross-selling and enhancing revenues. This is pos-
sible because existing customers are known (in terms of expectations,
needs and desires) and the products and services supplied are also known
(in terms of customer needs satisfied by the products/services supplied).
Companies have therefore a higher probability of selling products and ser-
vices while sustaining lower marketing costs.
(b) Customer satisfaction improves the competitive position and, con-
sequently, profits increase because satisfied customers communicate their
satisfaction to other potential customers (i.e. cross-customer communica-
tion) thereby improving the bank’s image. As noted by Anderson (1995),
there are three causes for the existence of the cross-customer communica-
tion: altruism (the wish to help other people), instrumentality (the wish to
appear well-informed and expert) and confirmation (the reduction of
the post-acquisition cognitive gap, due to uncertainty, relating to having
made the best purchase). Cross-customer communications (and the bank’s
image) are particularly important in banking since customers cannot ex-
perience banking products and services before their acquisition. A strong
competitive position enables a bank to attract new customers by sustaining
lower marketing costs: the cost of attracting new customers is usually esti-
mated to be at least five times the cost of retaining existing customers. As
such, a bank can focus on improving the quality of its services to increase
customer satisfaction and be confident of attracting new customers by
cross-customer communications. In this way, marketing cost savings
should balance out (in the medium term), the initial expenses incurred
being covered by enhanced revenue through an increase in the quality of
banking services and greater customer retention. In the long term, the
overall effect should be an improvement in operating profits. Following
this process, a bank improves equity returns and, at the same time,
finances those business processes intended to monitor customer satisfac-
tion and to improve the quality of its banking services.
The dispute between stakeholders is not completely resolved yet. The above
discussion has in fact assessed the relationship between shareholders and cus-
tomers, but it has omitted consideration of the dispute between managers,
employees and shareholders. In other words, following Agrippa’s apologue,
there is still an open question: With which part of the body can shareholders
and managers be compared?
In our opinion, shareholders in a company play the same role as the brain in
a human body. They give the impulse for managing every part of the body (i.e.
companies) and receive the necessary nourishment from the stomach (i.e. cus-
tomers). As the brain, shareholders have two main tasks. First, they identify
how best to satisfy customer expectations, selecting senior managers and over-
seeing their activities and creating an efficient structure. This is exactly what
the brain does for the human body: the brain manages the body, attempting to
avoid waste of energy (i.e. efficiency) and chooses the most appropriate food
for the stomach (i.e. effectiveness). Secondly they need to satisfy the work-
force’s expectations, for example, by paying fair salaries and supplying good
incentives. This is exactly what the brain does in the human body as it distrib-
utes the nourishment coming from the stomach (i.e. customers) to the body
(employees and managers). In contrast, managers play a role similar to that of
the nervous system. The nervous system receives impulses from the brain (i.e.
shareholders) and these impulses are transmitted to the other parts of the body
(lower-level managers and employees). Nourishment is produced by the
stomach (i.e. customers) and allocated by the brain (shareholders).
This apologue shows that, in a perfect market (as in a perfect body), all
stakeholders are dependent on each other for their success. Managing to
create sustained and sustainable shareholder value is in fact not a zero-sum
game where the shareholders prosper at the expense of other stakeholders.
On the contrary, creating stable shareholder value requires an intense focus
on delivering benefits to customers in the most efficient way, hiring and
retaining a motivated workforce, maintaining excellent supplier relation-
ships, and being a good corporate citizen in each of the local communities
where the company has a presence.
This approach is usually referred to as ‘stakeholder symbiosis’ where
shareholders must take responsibility for all stakeholders because com-
panies that take into account the interests of all their stakeholders will
benefit from a steadily growing value. Under these circumstances, when the
shareholders win – that is, when wealth is created – all stakeholders win. It
is only when wealth is destroyed that stakeholder conflicts emerge. As
shown by a study of the Institut International d’Etudes Bancaire,21 this
view is currently common among bankers: ‘Deutsche Bank’s overriding aim
is to create and grow shareholder value in line with increasing our cus-
tomers’ and employees’ satisfaction, thus creating public benefit. We do
not believe that there is an inherent conflict between the interests of share-
holders, customers, employees and the public in the long run’ (Dalborg
1999: 14).
The stakeholder symbiosis leads one to identify a service value chain22
(as shown in Figure 2.4), which illustrates the long-term relationship
between customer satisfaction, customer loyalty, productive efficiency,
INTERNAL EXTERNAL
Loyalty
Satisfaction Motivation
Customers Shareholder
Human Output Satisfaction Loyalty Value
Capability
Ca resources Quality
Productive Service
efficiency quality
Figure 2.4: The service value chain
employee motivation and satisfaction and shareholder value. In a com-

petitive market, shareholder value rests on customer satisfaction and pro-
ductive efficiency, which also requires employees’ and managers’
satisfaction.
One may claim that the ‘stakeholder symbiosis’ is always observable in a
perfect market, but it may be weakened in the short term in a quasi-perfect
market. For instance, when information is largely (but not perfectly) avail-
able, and there are few transaction costs and competition is high (but not
perfect), in the short term, shareholders may increase their value at the
expense of customers, managers and employees. For example, shareholders
may attempt to maximise profits by reducing staff and negotiating wage
reductions and charging excessive prices or cutting the expenses for
research and development.
In the long run, however, these strategies will not assure sustained share-
holder value in a quasi-perfect market, because, in most industries (among
them banking), a company’s success depends on the skills of its workforce,
not only those of a few individuals. Staff reductions and wage negotiations
usually make managers and employees less motivated and unsatisfied by
reducing productive efficiency and the quality of products and services sup-
plied. Shareholder-value maximisation therefore requires fair salaries and
opportunities for personal development through training programmes, job
rotation, career building, etc. In the medium and long term, shareholder-
value creation implies more job opportunities (than those associated with
other corporate goals) and fairer contracts. Staff cuts and wage negotiation
should be implemented, therefore, only if they are intended to lead to sus-
tained and sustainable shareholder value rather than to reach a short-term
goal. In some cases, staff reductions may be necessary in order to improve
productive efficiency and, in this case, shareholders should take the res-
ponsibility for the placement of redundant staff. MeritaNordbanken,23 for

example, launched a programme to reduce the number of employees
(300 people out of a total of 7300). The programme was designed to help
these employees by offering them either a new job in the bank, training for
a job outside the bank and/or support for self-employment. Although this
programme implied a 30 per cent higher cost than the minimum down-
sizing cost, MeritaNordbanken’s employees (and prospective employees)
noticed that the bank gave proper attention to its human resources
development in pursuing sustained shareholder value.
Staff reductions or the charge of excessive prices will not assure sustained
shareholder value in a quasi-perfect market since competition forces tend
in the long run to eliminate truly excessive pricing or other unreasonable
terms of sale. The crucial question is when will customers judge ‘excessive’
the price charged? The answer depends on customer satisfaction. As men-
tioned earlier, in a long-term commercial relationship (such as banking),
customers are willing to pay higher commissions in order to receive higher
quality services: if customers are satisfied, a company may increase the
price level before customers consider the possibility of moving to another
competitor. The price therefore becomes excessive when it is higher than
what customers ‘expect’, where the customers’ expectations are determined
by their experience, their needs and cross-customer communications. A
price increment may allow shareholders to achieve sustained and sustain-
able value if the services supplied are of high quality; and to achieve short-
term value if the services supplied do not meet certain quality standards,
although in this latter case, the higher price will eventually reduce cus-
tomer satisfaction and, in a competitive market, customers will eventually
switch to other competitors.
The shareholder value approach in European banking

This section deals with the shareholder value approach in banking. The
analysis is organised as follows:
• Recent trends in European banking are noted in order to support the

idea that, under the current operating environment, the creation of sus-
tained shareholder value is a main strategic target for European banks.
• Shareholder value issues are analysed by distinguishing among different
types of bank and banking systems. Hence we assess the shareholder value
approach (developed in the US) to gauge its applicability to European banks
and consider whether shareholder value creation is also an appropriate goal
for mutual (e.g. cooperative) and state-owned banks.
Trends in European banking and the shareholder value approach

This section presents some recent trends in European banking in order
to support the idea that, under the current operating environment, the
creation of sustained shareholder value is a main strategic target for

European banks.
Over the last decade, most governments have begun a deregulation process
and have eliminated a wide range of protective rules. These protective rules,
adopted as a response to the banking and financial crises of the 1930s,
shared the basic idea that competition had to be constrained to preserve
the stability of the banking and financial systems. As noted by Padoa-
Schioppa (2001: 14), ‘It was widely believed that in an oligopolistic envir-
onment banks could enjoy extra-profits, which would foster the stability of
individual banks and the banking system as a whole … until very recently,
there was no embarrassment in pursuing and supporting explicit limitation
to competition.’ Gradually, a positive view of competition has developed
and it has now become a primary concern for supervisors in order to
increase banks’ efficiency. Up to the 1980s, regulators aimed to guarantee
market stability and implemented a structural regulation approach, a policy
that prevented competition among banks, reducing productive efficiency
and sacrificing the quality of banking services. Since the start of the 1990s
in Europe (and after the EU’s Second Banking Directive in 1992), deregula-
tion seems to have made European banking industries more competitive,
increasing corporate efficiency and the quality of banking services.
The structural features of many banking systems (such as in the US and
Europe) have changed dramatically during the 1990s, as shown by the trends
in number of branches and banks. Table 2.1 and Figure 2.5 show that the
number of banks has decreased in almost all European countries: the number
of banks was stable over an initial period (1985–90) and, since then, has started
to fall. Focusing on the period 1999–2001, the number of banks entering the
market is substantially lower than leavers; this may be interpreted as a signal
that competition in European banking industries has become effective as less
profitable banks are eliminated from the market or acquired by more profitable
banks. This view seems to be confirmed by Padoa-Schioppa (2001: 16) who
notes ‘competition means selection. Hence, such an attitude also implies that
authorities must be ready to let the weakest banks leave the market.’
In the main European banking systems Figure 2.6 shows that the number of
branches substantially increased in most European countries, such as in Italy,
Spain, the United Kingdom and France, although this growth has followed dif-
ferent patterns. In the United Kingdom and Germany, the number of branches
increased substantially from 1985 and 1993 (by 50 per cent and almost 30 per
cent, respectively) and then started to fall. This trend perhaps reflects initial
strong growth over the period of major deregulation. In contrast, the branch
numbers continue to grow in Spain and Italy which experienced the strongest
branch restriction on all banks up to deregulation in 1992.
A second factor of change in many banking industries is that techno-
logical advances seem to have reduced transaction costs and information
asymmetry. Telephone banking can reduce transaction costs by 50 per cent
Table 2.1: Number of banks in European banking systems between 1985 and 2002
Number of joiners and leavers in the banking sector

Country Number of banks (January 1999 – December 2001)
Rate of Joiners Leavers Joiners Leavers

1985 2001 change (%) (no.) (no.) Net (%) (%)
Austria 1,241 838 –32.5 18 78 –60 2.1 9.3

Belgium 165 112 –32.1 16 27 –11 14.3 24.1
Finland 654 340 –48.0 6 14 –8 1.8 4.1
France 2,109 1,056 –49.9 72 242 –170 6.8 22.9
Germany 4,740 2,547 –46.3 45 736 –691 1.8 28.9
Greece 38 60 57.9 12 11 1 20.0 18.3
Ireland 58 88 51.7 21 11 10 23.9 12.5
Italy 1,192 844 –29.2 94 184 –90 11.1 21.8
Luxemburg 118 194 64.4 14 32 –18 7.2 16.5
Holland 2,055 565 –72.5 34 103 –69 6.0 18.2
Portugal 224 213 –4.9 18 32 –14 8.5 15.0
Spain 773 369 –52.3 37 70 –33 10.0 19.0
Euro area 13,367 7,226 –45.9 387 1,540 –1,153 5.4 21.3
Denmark 166 203 22.3 8 17 –9 3.9 8.4
Sweden 543 150 –72.4 18 16 2 12.0 10.7
United Kingdom 655 450 –31.3 33 104 –71 7.3 23.1
European Union 14,731 8,029 –45.5 446 1,677 –1,231 5.6 20.9
Sources: Affinito et al. (2003: 36–9) reporting data collected from ECB and national central banks.
31
Figure 2.5: Number of banks in European banking systems between 1985 and 2002
100
90 UK
80
Number of banks
Ger Italy
70
60
EU
50 France
Sp
40
30
Netherlands
20
1985 88 89 90 91 92 93 94 95 96 97 98 99 00 01
Year resources
Sources: Affinito et al. (2003: 42).
Figure 2.6: Number of branches in European banking systems between 1985 and
2002
250
230
Italy
210
Number of banks
190
170
150
Spain
130 UK
110 EU
France
90 Germany
Netherlands
70
50
1985 88 89 90 91 92 93 94 95 96 97 98 99 00
Year (1985=100)
Sources: Affinito et al. (2003: 43).
Figure 2.7: Estimated bank transaction costs (Euro per transaction)
80
No. of inhabitants
No. of web users
70 Web users over inhabitants (in %)
60
50
Milions
40
30
20
10
7.5
5
2.5
Irl Nor Fin Dan Swi Au Swe Por Bel Gre Aus Spa Ita UK Fra Ger
European countries
Source: author’s own using data from Tylecote and Tarhan (2000: 13) who report data from
Lehman Brothers (1999).
(compared with the traditional branch-based transactions), ATM transac-

tions are 75 per cent lower and internet transactions 85 per cent lower than
branch-based transactions.24 The impact of the internet is potentially higher
in some countries (such as Scandinavia and the United Kingdom) where the
diffusion of the internet is larger than in the rest of continental Europe
(Figure 2.7). As noted by Tylecote and Tarhan (2000: 10), ‘It will probably be
decades before the use of the internet becomes virtually universal: presum-
ably a large proportion of those now over 40 will be reluctant to adopt it –
and so there is a clear opening for the use of more familiar electronic media,
such as TV. Thus HSBC launched TV banking for their existing customers in
November 1999, in collaboration with Panasonic (a hardware supplier) and
B Sky B (a satellite broadcaster). This is an interactive digital multimedia
system: the loop between bank and customer is completed by the phone
line, which carries the customer’s messages to the bank.’
A third major trend in the banking industry relates to the changing produc-
tion process and the relationship with the workforce. If the importance of the
workforce is currently recognised in most service industries, it is still necessary
to highlight that human resources are crucial in banking. Most banking
services25 are supplied in real time directly by front office staff: a disgruntled
workforce will provide poor quality services creating unsatisfied customers.

Banking is an industry where success depends on the skills, the satisfaction and
motivation of the workforce; these conditions should apply not only to a few
individuals,26 but to the majority of employees. Several important banks have
solved the divergence between employees’ goals and shareholders’ goals by
means of an appropriate compensation system. As affirmed by Hatch,27 ‘at the
First Union, we have moved very much to performance-based compensation
for middle and upper management. For most of our middle and top managers,
only one-third of the total compensation package is based on salary. Another
third is stock-based compensation and the remaining third is incentive-based
compensation. So, we believe very strongly in turning managers into owners.’
In such market conditions (i.e. higher competition, lower transaction costs
and information more widely available to customers, plus the critical role of
human resources), the creation of sustained shareholder value seems to require
a ‘stakeholder symbiosis’ correlating stakeholders’ and shareholders’ satisfac-
tion. While shareholders can maximise their value at the expense of other
stakeholders in markets that are uncompetitive and with high transaction
costs, the shareholder value approach needs to be aligned or be acceptable to
all stakeholders in competitive markets. These remarks have been fortified by
evidence in a survey of European banks carried out in 1999 by Institut
International d’Etudes Bancaire,28 which found that ‘half of the respondents
say they communicate an explicit objective for shareholder value creation, and
more than 70 per cent find that the investor community expects banks to set
up shareholder value goals. The shareholder value concept has, according to
many respondents, received the right degree of attention in the business press,
but some find that this is not the case as regards the governing of companies’
(Dalborg 1999: 37). Shareholder-value maximisation has also been recognised
as a reasonable goal by regulators: ‘you may well wonder why a regulator is the
first speaker at a conference in which a major theme is maximising shareholder
value … regulators share with you the same objective of a strong and profitable
bank system’ (Greenspan 1996: 1).29
Is the shareholder value approach suitable for all banks?

Banks with different ownership features (private, public, mutual) operate in
many European systems and one may argue that these pursue different
objectives. Two crucial questions may be advanced: Is the shareholder
value approach credible for public banks? And is shareholder value maxi-
misation a reasonable goal for mutual banks?
Regarding the first question, the answer should be positive on theoretical
grounds as the ‘economicity’ of management (which leads to shareholder value
creation) is a proper goal for any company and, therefore, even for public
banks. In practice, however, government-owned banks have traditionally
benefited from lower pressures for profitability and shareholder value maximi-
sation has never represented a primary target for these banks. It is worthwhile
to note that this issue seems currently of less importance than in the past as
Table 2.2: Bank privatisation in Italy (1993–2001)
Bank Date of Share Gross proceed

operation transferred (in billions of
(%) ITF)
Credito Italiano Dec. 1993 58 1801

Banca Commerciale Feb. 1994 51.9 2891
IMI (1st tranche) Jan. 1994 36.5 2180
IMI (2nd tranche) Jul. 1995 19.0 1200
IMI (3rd tranche) Jul. 1996 6.9 501
Banco di Napoli (1st tranche) Jun. 1997 60.0 62
San Paolo di Torino Jun. 1997 6.2 529
Banca di Roma Dec. 1997 36.2 1893
BNL Dec. 1998 78.0 7731
Mediocredito Centrale Dec.1999 100.0 3944
Credito Industriale Sardo May 2000 53.2 42
Banco di Napoli (2nd tranche) Nov. 2000 16.2 984
Meliorbanca Jul. 2000 7.2 29
Mediocredito Lombardo Jul. 2000 3.4 75
Cofiri Feb. 2001 93.5 975
Mediocredito Fondiario Centro-Italia Mag 2002 3.4 11
Mediocredito Toscano Feb. 2002 6.51 34
Mediocredito Friuli Venezia Giulia Oct. 2003 34.01 119
Cassa Depositi e Prestiti Dec. 2003 30.00 2033
Coopcredito SPA Apr. 2004 14.42 30
Source: Adapted from Messori (2001: 28), Ministero dell’economia e delle finanze (2001, 2002,
2004).
most European countries have privatised many large government-owned com-

panies during the 1990s, including banks, and shareholder value maximisation
is currently a primary target for these banks. Table 2.2, for example, reports the
main bank privatisations in Italy between 1993 and 2001.
Concerning the other question, mutual and state-owned banks are
historically based on membership and their business model differs sig-
nificantly from those adopted by commercial banks so it is by no means
clear whether their main strategic forms should be based on shareholder
value maximisation. Firstly, there is no secondary market to trade shares
and no special price mechanism for supply and demand and, as such, real
shareholder wealth maximisation cannot take place. In the case of coopera-
tive banks, shareholders receive a different type of return since these banks
have to promote their financial status and redistribute extra returns to
members. However, as Schuster (2000) notes, mutual banks (such as co-
operative and savings banks) compete with private commercial banks and
so they are likely to adopt similar shareholder value strategies.
Another issue is whether the shareholder value approach is suitable for
continental European banks. The culture of a financial system affects the
goals of a bank. In this respect, it is possible to identify two models:

the Anglo-Saxon (or market-based) and the European continental (or
bank-based) models.
In the Anglo-Saxon (or market-based) model, such as in the US and UK,
equity is a key source of corporate financing. In these countries, institutional
investors traditionally hold a strong influence on corporate development, the
equity markets are developed and sustained by equity-financed pension funds
and mutual funds, information is easily available, trading is transparent and
the equity markets are liquid, large and efficient. In these markets, capital is
lured by equity growth (measured as return to investors) and shareholder value
creation becomes therefore a primary target for all parties (shareholders, com-
panies, institutional investors and individual investors). In continental
European (or bank-based) systems traditionally the banking sector has played a
more important role in financing industry. The origin of the bank-based
systems can be traced back to after the Second World War, when governments
took an active role in the rebuilding of European business by investing directly
in some sectors and giving certain advantages to increasing powerful national
investment groups (such as the Wallenbergs in Sweden, Daimler-Benz in
Germany and Fiat in Italy). Post-war many governments also introduced a wide
range of regulation to protect national companies. These measures and the
lack of incentives to save for retirement (as pension systems were developed
based on state-run funds) precluded the development of capital markets and
prevented them from becoming the main source of financing for companies.
In the aforementioned continental European countries, the return to
investors has not traditionally been the fundamental element to attract capital
and the tradition of shareholder value management is weaker than in Anglo-
Saxon countries. However, the situation is currently changing because various
forces are leading companies to focus more on shareholder value management,
such as deregulation, privatisation, globalisation, the growth of institutional
ownership, availability of information and lower transaction costs.
Conclusion
This chapter has analysed the economic objectives of banks and shown
that strategically focusing on managing sustainable shareholder value is a
credible and worthwhile goal. The shareholder model is a management
approach aiming to create long-term value for company owners. Generat-
ing stable shareholder value growth requires a focus on delivering benefits
to customers in the most efficient way, hiring and retaining a motivated
workforce, maintaining excellent supplier relationships, and being a good
corporate citizen in each of the company’s local communities. The service
value chain shows the long-term relationship between customer satis-
faction, customer loyalty, productive efficiency, employee motivation and
satisfaction and shareholder value. In a competitive market, shareholder
value is based on customer satisfaction and productive efficiency, which

requires employees’ and managers’ satisfaction.
The second half of this chapter briefly analysed the shareholder value
approach in banking. Three questions were addressed: Is long-term shareholder
value maximisation a proper goal for banks? Is this an appropriate goal for all
types of banks? Is this goal reasonable in European banking? While the answer
to the first question is certainly positive, the answer to the second question is
not straightforward because mutual banks and government-owned banks
have traditionally pursued different objectives in the past. Finally, we briefly
analysed the Anglo-Saxon model (traditionally shareholder-oriented) and the
continental Europe model (traditionally stakeholder-oriented). The discussion
also highlighted how banking markets in continental Europe have sub-
stantially changed over the last decade and suggested reasons as to why
the creation of shareholder value has become an increasingly important
strategic target for European banks. The next chapter outlines the literature on
shareholder value creation.
3
Shareholder Value: a Literature Review
Introduction
As noted in the previous chapter, managing to create sustainable share-
holder value is an important objective in European banking. While there is
a substantial literature dealing with shareholder value, there remains con-
fusion as to how to create value for shareholders and noticeable partiality
among supporters of various metrics for measuring shareholder value. Resti
(1999: 19) notes that ‘shareholder value in banking is becoming a sort of
mot de passe, a skeleton key for making all agree, a spice to give flavour
even to non-tasting projects and find the favour of all guests. Yet, despite
the unavoidable banalities implied in any formulas for corporate success,
the creation of shareholder value is an urgent issue in European banking.’
This chapter has three aims. The first is to define the concept of share-
holder value and briefly review the foundations of Shareholder Value
Theory; the second aim is to illustrate the theoretical issues relating to the
measurement of shareholder value, focusing, especially, on Economic
Value Added (EVA). The third aim is to review the empirical literature
dealing with shareholder value.
The concept of shareholder value

The concept of shareholder value is one of the oldest features of business
performance1 – a company creates value for the shareholders over a given
time period when the return on invested capital is greater than its oppor-
tunity cost, or than the rate that investors could earn by investing in other
securities with the same risk. As a consequence, the shareholder value
(SHV) created over the period (e.g. t–1, t) is obtained by multiplying this
abnormal return (i.e. shareholder total return minus the expected rate of
return over the period t–1, t) by the capital invested by shareholders, which
is expressed by the Equity Market Value of the company at time t–1. In
order to highlight that this concept of ‘value’ implies a comparison of the
38
Shareholder Value: a Literature Review 39
shareholder return with the opportunity cost, the word ‘added’ is often
joined with the term ‘value’ [SHV(added)].2 In detail:
SHV(Added)t–1,t = (Shareholder returnt–1,t – expected rate of (3.1)

returnt–1,t) × Equity Market Valuet–1
The shareholder return in a given time period (e.g. t–1, t) expresses the rate
of change of the shareholder wealth.3 This measure can be calculated by divid-
ing the shareholder wealth created in a given period (i.e. the change of the
shareholder wealth between the beginning and the end of the period) by the
market value of equity capital at the beginning of the period.
Created Shareholder wealtht–1,t
Shareholder returnt–1,t = (3.2)
Equity Market Valuet–1,t
If it is straightforward to define the shareholder wealth at a given point

of time (e.g. t) as the market value of the company’s equity capital, it is
more complex to define how shareholder wealth changes over time (this is
labelled ‘created shareholder wealth’). At first sight, the change of share-
holder wealth in a given time period may be simply obtained by looking at
the increase in equity market value. However, this is not entirely accurate
because it is also necessary to consider in the calculation: (1) the dividend
paid over the time period; (2) the other payments (e.g. for share buy-backs)
made to shareholders during the time period; (3) outlays for capital
increases; and (4) the conversion of a company’s convertible bonds. As
such, the shareholder value added can be accurately measured as:
Created Shareholder wealth t–1,t = equity market valuet – equity (3.3)

market valuet–1 + dividend
paid t–1,t + other payment to
shareholders t–1,t – outlay for
capital increases t–1,t –
conversion of convertible
company’s bond
As such, model (3.1) can be rearranged as follows:
SHV(Added)t–1,t = Created Shareholder wealtht–1,t – (expected rate (3.4)

of returnt–1,t × Equity Market Valuet–1)
= (equity market valuet – equity market value t–1
+ dividend paid t–1,t + other payment to
shareholders t–1,t – Outlay for capital
increasest–1,t – conversion of convertible
company’s bond) – (expected rate of return t–1,t
× Equity Market Valuet–1)
Although this is straightforward, the problem of measuring created

shareholder value is not completely solved since:
(1) This calculation procedure can be adopted only for publicly traded
companies. Since the number of listed banks is still small in continen-
tal Europe, there is a strong need to find which performance indicators
best capture created shareholder value so as to inform shareholder
value debate concerning non-listed firms.
(2) This measurement procedure provides an accurate ex-post assessment
of the bank’s ability to create shareholder value, but it has limited
use as an ex-ante assessment since future changes in stock prices, divi-
dends, other payments to shareholders, outlays for capital increases
and the conversion of convertible company’s bond are unknown.
(3) There are a number of competing bank performance measures and it
is by no means clear which is preferable in terms of measuring value cre-
ation. For example, several performance measures have been developed
over the 1990s and it is currently debated which is the best method for
assessing the value created by firms for their owners, as researchers and
practitioners grapple with different performance metrics.
Evolution of Shareholder Value Theory

The origin of Shareholder Value Theory (labelled also the Theory of
Business Value) can be traced back to the seminal works of Markowitz
(1952), Miller and Modigliani (1961), Sharpe (1964) and Fama (1976).
Shareholder Value Theory originates from the works undertaken by these
economists around the Capital Asset Pricing Model (CAPM), which affirms
that the returns expected by investors are related to the risk incurred by
owning particular financial assets.
This consideration enables us to clarify an initial confusion: profitability
is not the same as shareholder value. Although profitability and share-
holder value have a common root, the differences between these concepts
become evident when a bank defines its target in terms of return on equity
(ROE). The ROE is a simple financial ratio that expresses profits (after tax
and independently from any decision about its distribution as dividend) for
every 100 of equity capital invested. Only after analysing the functions of
equity capital is it possible to clarify the distinction between profitability
and shareholder value. From the regulator’s point of view, capital acts as a
protection for bank depositors in case of a substantial reduction in the
value of a bank’s assets. From the point of view of the bank, equity capital
is a productive factor, which is necessary to finance its business activities.
Like any productive factor, equity capital has a cost for a bank which is
given by the rate of return required by private investors (that can be mea-
sured with the CAPM framework). When one considers the rate of return
required by private investors, the difference between profitability and

shareholder value becomes evident. A bank can improve its profitability by
undertaking risky operations (e.g. by speculating on securities), but may
not create value since shareholders will require a higher rate of return. As a
rule, a bank (like any firm) creates shareholder value only when its rate
of return is higher than its cost of equity. When this condition cannot be
met, some banks prefer to give back redundant equity capital through
buy-backs or by paying extraordinary dividends rather than inadequately
remunerate this capital in the future.
This section briefly provides an overview of the most important contri-
butions to the Theory of Business Value outlined by Miller and Modigliani
(1961) and Porter (1980) as well as some more recent developments.
The work of Miller and Modigliani (1961) is widely considered to be the
origin of the Theory of Business Value research. They demonstrated the
irrelevance of dividend (distribution) policies in an ‘ideal’ environment,
where the value of a business is linked to two elements: (1) the present
value of expected dividends and (2) a company’s residual value. This divi-
sion is the starting point for most of the subsequent research on financial
evaluation methods. The Shareholder Value Approach is clearly spelled out
for the first time with the appearance of the Capital Asset Pricing Model
(CAPM) and its widespread use in business research.
Porter (1980) models economic sectors and the competitors within them
and highlights the importance of strategy. This author developed the ‘value
chain’ concept that is the key to understanding competitive advantage and
creating sustainable value for shareholders. Company activity is broken
down into strategically meaningful units to understand costs and other firm
characteristics. By supplying a complete method for strategy formulation,
Porter (1980) outlines the fundamental elements that need to be considered
when selecting a strategy aimed at creating value. These include such factors
as a sector’s attractiveness (long-term return factors) and relevant factors
affecting a company’s competitive standing in the industry. Porter identifies
five structural forces whose interaction determines long-term sector profit-
ability: (1) threat of new competitors; (2) threat of new substitute products;
(3) suppliers’ bargaining power; (4) buyers’ bargaining power; and (5) the
degree of competition among companies. These five forces establish a
complex competitive environment and determine a company’s ability to
achieve investment returns higher than the average cost of capital, i.e. the
starting point to create value for shareholders. In conclusion, Porter is
the first to have provided a complete framework for generating a compet-
itive advantage and, as such, creating sustainable shareholder value.
Among more recent studies, two studies deserve a special mention.
Rappaport (1998) explains effectively why shareholder value creation should
be the main corporate objective by providing not only a clear goal, but also
the tools necessary for its achievement: strategic planning, competitive
improvement and investment management. Copeland et al. (2000) show

how to effectively apply shareholder value concepts to a going concern sup-
posed to generate value over time. Managing company value is not an
elusive and mysterious art: it is simply the result of a different perspective,
emphasising long-term cash flow returns rather than quarter to quarter
earnings per share (EPS) changes. Copeland et al. (2000) stress the impor-
tance of a corporate-wide adoption of this mentality, which might even
require a first painful restructuring phase to ‘free’ captive business value. The
next step is to identify a value-based priority scale: planning behaviour,
result measurement and reward systems directly linked to shareholder value.
The last fundamental step implied is ‘value reporting’, to keep shareholders
aware of the wealth-creation progress.
The Miller–Modigliani framework

The starting point of studies on Shareholder Value Theory evolved from the
seminal work of Miller and Modigliani (1961) that assessed the impact of
companies’ financial decisions on their own value. Miller and Modigliani
approached the problem of financial choices in an original way, which
remains the foundation for modern theories on company value.4 Dividend
policy is an important issue for company managers, for investors who have
to structure their portfolios and for economists who attempt to understand
the way financial markets operate. Miller and Modigliani work on company
valuation focused on dividend policy and their initial starting point is an
ideal world where several limiting hypotheses are in place. The basic
assumptions are: (1) financial markets are perfect. In fact, it is assumed that
no single market participant is able to affect market prices through his
transactions; information is available to all participants without cost and
transaction costs are absent; tax rates applied to dividend distributions and
capital gains are the same; (2) investors have a rational behaviour. Investors
prefer more wealth to less; they consider capital appreciation and cash
payouts as equivalent; (3) each investor has perfect knowledge of each
company’s future investment plans; and (4) there is only one financial
instrument and, as such, stocks are the only securities available. No dis-
tinction is made at this point between stocks and bonds as a source of
capital.
According to Miller and Modigliani, when these assumptions hold, share
valuation is subject to the following principle: the market-required rate
of return for shares of companies in the same industry is the same. This
rate of return is obtained as the sum of dividends and capital gains.
Independently from the company considered, the market rate of return for
period t of a company can be expressed as follows:
dj(t) + pj(t–1) – pj(t)

rj(t) = (3.5)
Pj(t)
where dj(t) is dividend per share for company j in period t, pj(t) is the share
price, net of t–1’s dividend for company j at the beginning of period t and
rj(t) is market rate of return for period t (independent from j).
By adjusting this formula as follows:
rj(t) × pj(t) = dj(t) + pp(t–1) – pj(t)

rj(t) × pj(t) = dj(t) + pj(t) – dj(t) + pj(t+1)
pj(t) × pj(t) = (1 + rj(t)) = dj(t) + pj(t+1)
it is possible to show that the price of any company and in any specific
time can be calculated as:
dj(t) + pj(t–1)
pj(t) = (3.6)
1 + rj(t)
where dj(t) is dividend per share for company j in period t, pj(t) is share price,
net of t–1’s dividend for company j at the beginning of period t and rj(t) is
market rate of return for period t (independent from j).
These equations hold for any j and t, otherwise owners of shares with
low returns could increase their wealth by selling them and buying high-
return stocks. This process eventually affects share prices until they are
fairly priced (i.e. the returns are the same). This can be examined also con-
sidering the value of the whole company, rather than the value of its
shares. In this case, it is possible to calculate the value of the company as
follows:
1
V(t) = [D(r) + n(r) p(t+1)] (3.7)
1 + r(t)
where n(t) is number of shares at the beginning of period t, m(t+1) is number

of new shares (if any) sold during t at the ex-dividend closing price p(t+1) so
that n(t+1) = n(t) + m(t+1), V(t) is the total value of the company at the time t
[i.e. V(t) = n(t)* p(t)] and D(t) is total dividends paid during the period t to
holders of shares existing at beginning of t [i.e. D(t) = n(t) * d(t)].
This model expresses a company’s value as a function of two discounted
cash flows (CFs): distributed dividends (D(j)) and the company’s residual
value at the end of the chosen period (n(t)*p(t+1)). The model is especially
useful in pinpointing a company’s source of value and is at the cornerstone
of company valuation methods, including Discounted Cash Flow (DCF)
and Economic Value Added (EVA) approaches.
By adjusting this formula as follows:
1
V(t) = [D(t) + n(t) p(t+1) + V(t+1) – V(t+1))]
1 + r(t)
1
V(t) = [D(t) + n(t) p(t+1) + V(t+1) – n(t+1) p(t+1)]
1 + r(t)
1
V(t) = [D(t) + V(t+1) + (n(t) – n(t+1)) p(t+1)]
1 + r(t)
1
V(t) = [D(t) + V(t+1) + (n(t) – n(t) – m(t+1)) p(t+1)]
1 + r(t)
the value of any company and at any specific time can be calculated as:
1
V(t) = [D(t) + V(t+1) – m(t+1) p(t+1)] (3.8)
1 + r(t)
This model allows us to identify three different ways in which company

market value V(t) (or, equivalently, its share price P(t)) can be affected by
present dividend policy.
• The first relationship is direct: current dividends [D(t)] (i.e. the first term
in brackets) affect directly company market value [V(t)].
• The second way is indirect. In principle, company market value [V(t)]
may be influenced by the current dividend policy via the second term in
brackets [V(t+1) ], i.e. the new market value net of dividends. In theory,
the company’s new market value net of dividends depends on future
events and not on past ones; such could be the case, however, only if
simultaneously V(t+1) depends on future dividend policy, and D(t) pro-
vides information concerning future dividend policy not otherwise
available. This means that dividend policy is known for t+1 and all fol-
lowing periods, but is still independent from the present dividend level,
D(1). Thus V(t+1) does not depend on present dividend decisions, even
though it is affected by D(t+1).
• The equation’s third element [m(t+1) p(t+1)], i.e. the value of new shares
sold during the current period, can also influence company market value
[V(t)] since a higher payout ratio (i.e ratio between dividends paid and
earnings) implies a higher future need for external financing, for each
given level of company investments. This effect is clearly linked to the
‘shares only’ hypothesis, which excludes the possibility of using debt as
a source of financing.
As a result, dividend policy affects share prices in two opposite ways: posi-
tively through D(t+1), and negatively through –m(t+1)* p(t+1). If the company
aims to increase dividends given its current investment projects, it first
needs to determine how the increased payments are offset by the reduction
in its terminal value. In an ideal world where all the stated hypotheses
hold, the two opposite effects perfectly offset each other, so that dividend
policy will not influence share prices in t. In order to prove this, Miller and
Modigliani (1961) begin by rewriting m(t+1)* p(t+1) as a function of D(t) as
follows:
m(t+1) p(t+1) = I(t) (X(t) – D(t)) (3.9)
where I(t) is the level of company’s investment (or increase in its holding
physical assets) in the period t, and X(t) is the company’s total net profit for
the period t. By substituting model 3.9 in model 3.8 and assuming at the
same time that D(t) = X(t) – I(t), the company’s market value can be shown as:
1
V(t) ≡ n(t).p(t) = [X(t) – I(t) + V(t+1)] (3.10)
1 + r(t)
Since D(t) is not directly present and all the remaining variables are inde-
pendent of it, it follows that company value has to be independent from
current dividend decisions. Miller and Modigliani (1961: 414) affirm also
that ‘having established that V(t) is unaffected by the current dividend deci-
sion, it is easy to go on and to show that V(t) must also be unaffected by any
future dividend decision as well’.
Given a company’s investment policy the dividend payout ratio does not
affect the present share values, nor does it affect shareholders’ total returns.
Miller and Modigliani (1961) consider a dividend policy’s ineffectiveness as
an obvious corollary to the absence of ‘financial illusion’ in a perfect and
rational economic environment. In such an ideal world, company values
are affected only by company investment policy and potential for superior
returns, not by the way these returns are distributed to shareholders. Miller
and Modigliani had already stated the irrelevance of a company’s financial
structure, but in 1961 they reshaped the way companies conceived the
wealth creation process. Porter (1980) contributes to the literature by devel-
oping a framework to show how firms can create competitive advantage
and therefore generate sustainable shareholder value.
The Porter framework

Porter (1980) is the first (as far as we are aware) to provide a complete
framework for generating a competitive advantage in a competitive market
and thus enabling firms to create sustainable shareholder value. A strategic
analysis of any industry consists of two stages – strategy design and evalu-
ation. The first stage encompasses sector analysis, competition assessment
and the ability to affect these elements. The second involves a compari-
son of the different possible strategies, to identify the best way to create
economic value.
By supplying a complete method for strategy formulation, Porter outlines
two key elements that need to be considered when selecting a strategy:
(1) sector attractiveness (long-term return factors); (2) relevant factors

affecting a company’s competitive standing in the sector.
Both elements dynamically change over time. A sector’s attractiveness
changes over time and a company’s standing changes as competitive events
unfold. Strategy aims to establish a profitable and sustainable position,
against the backdrop of market forces affecting each specific sector. Strategy
is the product of a deep knowledge of a sector’s competitive rules. These are
rules to be followed, but also to be changed as much as possible to create
an advantage.
Porter identifies five structural forces whose interaction determines long-
term sector profitability. Figure 3.1 shows the framework advanced by
Porter and highlights the so-called five forces which include:
(1) Threat of new competitors.

(2) Threat of new substitute products.
(3) Suppliers’ bargaining power.
(4) Buyers’ bargaining power.
(5) Degree of competition among companies.
These five forces specify a complex competitive environment, and affect

prices, sales costs, investments and risks. They also determine a company’s
ability to achieve investment returns higher than the average cost of
capital.
When strategic planners correctly identify the company’s position in the
sector’s structure they can transform this knowledge into a competitive
advantage. As Porter (1980: 3) states, ‘Strategic advantage descends from a
business unit’s capability to create value for their customers, over the cost
required to produce it. Value is what customers are willing to pay for, and
the advantage arises when prices are lower, for a given value level; or when
special benefits provided justify higher prices. The two main strategic
advantages are cost superiority and differentiation.’ To be successful a
company needs to choose which kind of advantage it wants to have, and
the context in which it wishes to achieve it.
Regarding the first strategy, ‘cost leadership’ means that a company
intends to produce its goods at the lowest cost for its sector. This advantage
can have various sources: economies of scale, proprietary technology,
special access to resources, etc. This kind of company usually produces
highly standardised goods. Regarding the other strategy, ‘differentiation’
means the company chooses to be one of a kind in its sector, with respect
to a limited number of variables deemed important by customers. Once
these specific characteristics are identified the company sets out to provide
them in an unmatched way and this is the source of its competitive advan-
tage. As a consequence, competitive advantage cannot be understood by
looking at a business as a single entity since it is rather a collection of dif-
Figure 3.1: The Porter (1980) five forces model
Entry Barriers Competitive factors

Economies of scale Sector growth
Product specificity Fixed costs to value added ratio
Brands New Excess capacity
Conversion costs Competitors Product differentiation
Capital requirements Brand name
Access to distribution Conversion costs
Proprietary learning Threat of new competitors Concentration and equilibrium
Access to input Informational complexity
Low cost proprietary product design Variety among competitors
Government policies Stakes
Retaliation threat Exit Barriers
Existing
competitors
Suppliers' Buyers'
Suppliers bargaining power Buyers
bargaining power
Degree of rivalry
Buyers' bargaining power

Suppliers' bargaining power Threat from substitute products
Conversion costs for suppliers Transactional advantage Price sensitivity
Number of suppliers Quantities bought Price on lot purchase
Suppliers' focus on quantity sold Substitute Conversion costs Product differentiation
Cost of input relative to total supply cost products Buyers' information Brand recognition
Input relevance to cost or differentiation Upstream integration Quality & performance
Threat of supply market consolidation Substitute products Buyers' profits
Threat from substitute products Pull-through Decision-maker's incentives
Relative prices
Conversion costs
Customer perception of substitute product
Source: Adapted from Porter (1980: 3).

47
ferent activities (such as product design, packaging, production, sales and

customer assistance). Each one of these activities can be the source of
an advantage when it is carried out in a more efficient or economic way,
compared to company rivals.
Porter (1980) proposes the ‘value chain’ as a systematic approach to
understanding strategically relevant activities, cost trends, and existing and
potential sources of differentiation. The value chain shown in Figure 3.2 is
shaped by history, strategy and the way it is implemented, and by activity-
specific economic variables. The value chain approach should be used to
assess single business units, rather than an aggregate of them, to avoid a
loss of relevant information. Within this framework, in a competitive envi-
ronment, ‘value’ is what customers are willing to pay to obtain what the
company is offering. ‘Global value’ is thus given by total revenues or,
equivalently, quantity sold times prices. The company’s goal is then to
make sure that revenues exceed costs, allowing for a profit. Profit is the key
to competitive position analysis, since costs are often ‘managed’ to obtain a
higher price through differentiation.
The value chain has two elements: value-generating activities and
margins. Value-generating activities are physically and technologically
identifiable activities which allow the company to produce a viable good
for customers; they can be primary or secondary. Margin is the difference
between total value and total cost.
Figure 3.2: Porter’s value chain
Infrastrucutural activties Margins
Human Resources Management

Managerment
Managerment
R&D
Support activities
Resources purchases
Margins
Inbound Operations Out-bound Marketing Support

logistics logistics & Sales
Primary activities
Source: Adapted from Porter (1980: 40).

Primary activities, directly linked to product creation, can be classified as:
(1) Inbound logistics: input reception, stocking, and distribution.

(2) Operational activities: inputs are transformed and combined into the
final products (machining, assembly, plant management and main-
tenance).
(3) Outbound logistics: product collection, stocking, distribution, order flow
management, and shipping.
(4) Marketing and sales: providing a distribution apparatus, and inducing
customers to buy the product (advertising, pricing, special offers).
(5) Support: services provided to improve or maintain product value (instal-
lation, training, spare parts).
Each one of these activities can be crucial to obtaining a competitive

advantage, even though relative importance varies across sectors and
companies.
Value-generating activities are the building blocks of competitive advan-
tage, but the value chain is not a collection of independent and unrelated
activities, rather it is a complex system. Analysing the value chain allows
managers to focus on the areas most likely to generate a competitive advan-
tage, through costs or differentiation, and the way each activity is carried
out will affect costs and company position and also whether or not the
company’s products are perceived as unique by customers. Porter (1980)
dislikes the use of value added methods to identify competitive advantage,
since they use a somewhat artificial distinction between specific inputs
and other resources and he argues that cost trends cannot be fully
comprehended without a complete expenses examination.
Sector structure also affects the company value chain and it is at the
same time determined by the value chains of all the companies in that
sector. Sector structure shapes the relationships with customers and sup-
pliers, which is reflected in the way in which margins are distributed
among these players. Entry barriers affect a company’s ability to sustain the
configuration of its own value chain. Organisational chart design can be
used to aggregate activities in units. Chain value is a useful tool for this task
too, since it offers a systematic way to break down business activities and
reassemble them in (more) meaningful strategic units. By creating a com-
petitive advantage, a company is able to generate value for its shareholders.
Shareholder value metrics

Although the concept of shareholder value is by no means a new concept,
the problem of measuring created shareholder value is still open to debate
and controversial since there is no consensus as to the best method for
gauging the value that firms create for their owners. As noted in The
Economist (1997: 61), ‘inevitably the measures are also a big business for
consultants. Stern Stewart, the New York firm that developed Economic
Value Added (EVA), is the leader of the pack. But in recent years it has
faced competition from the Boston Consulting Group (BCG), Braxton
Associates, McKinsey and others. Many consultancies produce league
tables of value added and go to increasingly absurd lengths to protect their
particular brand.’5
The following sections discuss the main techniques proposed in the
literature to measure a firm’s performance.
Traditional accounting performance measures

Traditionally, company performance has been assessed using accounting
measures. The most commonly observed measure is net profit – the alge-
braic sum of all company incomes minus costs (including taxes). Since net
profit is generated both by core-business activities and extraordinary activi-
ties, firms’ performance is often assessed focusing on ‘operating profit (or
margin)’, which is obtained as the sum of all company incomes and costs
derived from core activities. In banking, two traditional measures of operat-
ing profit are the interest and intermediation margin. The first is defined as
the difference between interest revenues and costs, while the second is
obtained by adding net commission and fee profits (i.e. the difference
between incomes and costs due to commission and fees) to the interest
margin. Both margins, as well as company’s net income, assess corporate
performance in absolute terms. In order to have a relative measure, firm’s
performance can be assessed using profitability ratios: in this case, profit
and economic margins are standardised for a measure of the capital
invested within the company.
performance measure
Financial ratio = (3.11)
capital measure
The most common financial ratios for measuring firm performance include:
Return on Equity (ROE), Return on Assets (ROA) and Return on Invest-
ments (ROI). ROE is the ratio between the net income (after interest and
taxes) and the book value of stockholders’ equity. This index expresses the
overall rate of profitability of the company.
Net income
ROE = (3.12a)
Equity capital
In order to improve this measure, ROE has been modified in several ways. A
first set of adjustments focuses on the measure of company’s return: rather
than considering simply net profit, it is possible to use other measures such
as Earnings Before Interest and Tax (EBIT) and Net Operating Profit After
Tax (NOPAT). A second set of modifications aims to consider the effect of
inflation on ROE by deflating net profit by the inflation effect (labelled ‘real
ROE’) or by considering in net income the re-evaluation due to inflation of
fixed tangible assets (labelled ‘effective ROE’).
Net income – Inflation rate (Total assets – Tangible fixed asset)

Real ROE = (3.12b)
Equity capital
Net income + Inflation rate × Tangible fixed asset)

Effective ROE =
Equity capital
ROA is the ratio between net income (after interest and taxes) and the book
value of total assets. This index expresses the overall rate of profitability of
the company in terms of profit generated by assets (as opposed to equity
capital in the case of ROE).
Net income
ROA = (3.13)
Total asset
ROI is the ratio between the operating profit and the book value of assets
used in operating activities. This index enables one to assess the rate of
profitability of the company in its core activities: the company’s per-
formance is measured as operating profit and capital is measured as the
book value of the assets used in operating activities.
Operating profit
ROI = (3.14)
Capital invested
Profitability ratios are commonly employed to assess corporate perform-

ance following a decomposition technique. Figure 3.3 outlines one of the
most well-known ROE decompositions (originally proposed by Cole 1972),
where there are three decomposition stages. Firstly, ROE is decomposed
into ROA and the Equity Multiplier. Secondly, ROA is decomposed into
a ‘profit margin’ and ‘asset utilisation’. Thirdly, ‘profit margin’ and ‘asset
utilisation’ are once more decomposed to assess the cause of good or bad
performance uncovered by the first two stages.
Financial ratio analysis has been criticised for the following reasons. First,
these measures are built on accounting data rather than on an economic
base. As noted in Kimbal (1998: 36), ‘economists and accountants differ on
the proper definition of profit. To an accountant, profit is the excess of
revenues over expenses and taxes and is best measured by earnings. To an
economist, earnings fail to include an important expense item, the oppor-
tunity cost of the equity capital contributed by the shareholders of the
firm.’ In addition, accounting data are influenced by accounting standards
Figure 3.3: ROE decomposition in banking
52
Deposit
Total Income
minus
minus Non-deposit
Interest expenses
Net income minus Salaries and wages
Operating & other expenses plus
Profit margin minus Net occupancy expenses
divided Taxes plus
Provision for loan losses
Interest and fees on loans plus
ROA Operating income plus Other expenses
(Return over total assets) Interest on investments
divided
plus
Service charges Balance sheet
Operating income
plus
Off- balance sheet
ROE Other income
Return over divided Asset utilisation divided
equity Tangible
Cash and due
plus
plus Tax-exempt
Investments Commercial and industrial
Total assets plus plus
Loans Consumer
Leverage plus
(Total asset over equity) plus Real estate
plus
Other assets
Other loans
Decomposition I: Decomposition II: Decomposition III:

ROE = ROA x Leverage ROA = Profit Margin x Asset Profit Margin = Net income / Operating income
Utilisation Asset Utilisation = Operating income / Total Assets
Source: Adapted from Sinkey (1998: 91).

(such as GAAP) and may be manipulated by managers. As a consequence,

accounting ratios may not supply objective and accurate indications about
corporate performance (such as, for example, for Enron in the US or
Parmalat in Italy). Second, financial ratios do not consider the time factor
in the performance valuation: e.g. some of the value compared in financial
ratios are obtained at different times or may have a different nature (flows
over stock values). Third, financial ratios concentrate on the past and do
not consider the future prospects for firms. Fourth, financial ratios do not
take into account risk. Accounting ratios fail to consider risk and simply
focus on returns: in this way, analysts obtain a partial vision. As noted in
Uyemura et al. (1996: 98), ‘traditional performance measures (net income,
ROA; ROE, earning per share) do not properly reflect risk and therefore
reinforce behaviour that is either too aggressive (that is, aims to maximise
earnings) or too conservative (aims to prevent dilution of returns)’. This
limitation is particularly important in banking since risk assumes different
configurations (such as credit risk, interest rate risk, market risk, liquidity
risk, operational risk, etc.) and regulators aim to supervise banks in order to
create disincentives for excessive risk-taking activities. Finally, financial
ratios are not able to assess the true value created by companies. These per-
formance indicators do not consider the capital invested and do not
include the cost of capital invested and, as a result, financial ratios are
unable to assess shareholder value creation. For instance, the performance
of a company is judged positive (according to the financial ratio) if ROE is
positive. However, this firm may not have created value: if the cost of
equity is higher than ROE, the company has destroyed value.
In order to solve these limitations, three different approaches to corpor-
ate valuation have emerged in the literature. Firstly, company evaluation
has developed from an accounting perspective to a financial perspective.
The underlying idea is that the value of a company is given by the sum of
the present value of all future cash flows. Starting from the company value,
it is possible to assess company returns in terms of the difference between
the value of the company at the beginning and at the end of the period
considered. Within this literature, it is possible to identify the following
main methodologies: Discounted Cash Flow (DCF), Shareholder Value
Added (SVA) and Cash Flow Return on Investment (CFROI). DCF estimates
a company’s value by using future cash flows generated by the firm over its
life and a discount rate that reflects the collective risk of the firm’s assets.
SVA, originally developed by Rappaport (1986), is a more refined approach
to the DCF method. It aims to determine the Shareholder Value Added
created in a given period as the change in shareholder value that is mea-
sured as corporate value minus debts. CFROI measures the expected return
of the company by applying the concept of Internal Rate of Return (IRR) to
a company’s cash flows. In other words, CFROI is the discount rate that
makes the Net Present Value (NPV) of the gross cash flows and salvage
value equal to the gross investment, and can thus be viewed as a composite
internal rate of return, in current dollar terms.
Secondly, company evaluation has also evolved from a traditional
accounting view of company earnings to a more economic-oriented per-
spective. The underlying idea is that the value of a company is given by the
sum of the present value of all future economic flows. However, economic
flows should not be measured using accounting earnings that do not reflect
completely the true economic condition of a company since some costs
and expenses are unreal and may be influenced by accounting standards
and by managers’ subjective decisions. As noted by Al Ehrbar (1998: 70),
‘the economic model holds that investors care about only two things: the
cash that a business can be expected to generate over its life and the riski-
ness of the expected cash receipts’: consequently, investors have to filter
accounting earnings in order to consider only true economic costs and
incomes.
Finally, company evaluation has also evolved to include risk in corporate
valuation. Within this approach, there are three different forms of evalua-
tion: Risk Adjusted Performance Measures (RAPM), Market Value Added
(MVA) and Economic Value Added (EVA). RAPM are financial ratios calcu-
lated as economic earnings (i.e. adjusted accounting earnings) over the
capital invested: economic earnings and/or capital invested are explicitly
adjusted to account for company risk. MVA and EVA explicitly include a
capital charge in corporate evaluation by considering: (1) the cost of capital
invested (which is a function of firm risk); (2) the return required by share-
holders; and (3) firm’s financial structure. In detail, MVA6 is defined as the
current market value of all capital elements minus the historical amount of
capital invested in the company: this measure can be seen as the net
present value of all future market returns. EVA expresses the surplus value
created by a company in a given period, i.e. the firm’s profit net of the cost
of all capital. EVA is seen by its proponents as providing the most reliable
year-to-year indicator of MVA.7
Discounted cash flow

Building on a core principle in finance, the value of an asset can be viewed
as the net present value (NPV) of the expected cash flows on that asset.
N
Value of Asset = ∑ E(CFt)(1 + r)–t (3.15)
t=1
where r is the discount rate that reflects both the risk of the cash flows and
financing mix used to acquire it; N is the asset’s economic life and E (CFt) is
the expected cash flow at time t.
The value of any asset is a function of the cash flows generated by that
asset, the life of the asset, the expected growth in the cash flows and the
riskiness associated with the cash flows. Considering a firm as a collection
Table 3.1: Discounting cash flow valuation: equity vs. firm valuation
Model elements Equity valuation Firm valuation
Cash flow Expected Cash Flows Expected Cash Flows

discounted to Equity [E(CFE)] to the Firm [E(CFF)]
Discounting rate Cost of Equity (Ke) Weighted Average Cost of

Capital (WACC)
Model Value of Equity = Value of Equity =

N N
E(CFEt) (1 + Ke)–1 E(CFEt) (1 + WACC)–1
t=1 t=1
of assets, it is possible to define its value by using cash flows to the firm
over its life and by discounting them at the appropriate rate that reflects
both the risk of the cash flows and the financing mix used to acquire it.
The extension (from a single asset to a firm) of this approach is not
straightforward since some assets of a firm have already been made (i.e.
assets-in-place), but a significant component of firm value reflects expecta-
tions about future investments. It is necessary to consider cash flows from
all investments already made, but also to estimate the expected value from
future growth. This is the essence of the Discounted Cash Flow (DCF) method.
DCF can be applied to estimate both the value of equity stock and the
value of the entire firm. While both approaches discount cash flows, they
differ concerning the relevant cash flows and the choice of discount rates
employed. Table 3.1 summarises both approaches.8
The value of equity capital is obtained by discounting the residual cash
flows after meeting all expenses, tax obligations, interest and principal pay-
ments (i.e. labelled ‘expected cash flows to equity’) at the rate of return
required by equity investors in the firm (i.e. the ‘cost of equity’). Regarding
the calculation of the cash flows to equity (CFE), these can be measured as:
CFE = Net income + depreciation – Capital spending – ∆ working (3.16)

capital – principal repayments + new debt issues
In contrast, the value of the firm is obtained by discounting the residual

cash flows after meeting all operating expenses and taxes, but prior to debt
payments (i.e. labelled ‘expected cash flows to the firm’) at the cost of the
different components of financing used by the firm weighted by the market
value proportions (i.e. labelled ‘Weighted Average Cost of Capital’ –
WACC). Regarding the calculation of the Cash Flows to the Firm (CFF),
these can be measured using either Earnings before Interest and Tax or
Cash Flow to the Equity as a base for the calculation.
CFF = EBIT (1 – tax rate) + depreciation – Capital spending –

∆ working capital
CFF = CFE + interest expense (1 – tax rate) + principal (3.17)
repayments – new debt issues + preferred dividends
It is important to note that the correct application of these two approaches

leads to the same estimates of firm value. As affirmed in Damodaran (1994:
11), ‘while the two approaches use different definitions of cash flow and
discount rates, they will yield consistent estimates of value as long as the
same set of assumptions is used for both. The key error to avoid is mis-
matching cash flows and discount rates, since discounting cash flows to
equity at the weighted average cost of capital will lead to an upwardly
biased estimate of the value of equity, while discounting cash flows to the
firm at the cost of equity will yield a downgrade biased estimate of the
value of the firm.’
Shareholder value added

Shareholder value added (SVA) is a methodology based on the DCF method
and, as noted by Rappaport9 (1998: 32), ‘the shareholder value approach
estimates the economic value of an investment by discounting forecasted
cash flows by the cost of capital’.
According to the SVA framework, shareholder value is given by the cor-
porate value minus debts, where corporate value is the sum of three com-
ponents (model 3.18): namely, the present value of cash flows from
operations during the forecast period, the present value of the business
attributable to the period beyond the forecast period and the current value
of marketable securities and other investment that can be converted to
cash and are not essential to operating business.
SHV = Corporate value – Debt

SHV = (PV of CF in the forecast period + Residual value + (3.18)
Marketable securities) – debt
Regarding CF from operations during the forecast period, these are calcu-
lated as difference between operating cash inflows and outflows. Rappaport
(1998) proposes to estimate these as follows:
Cash flow = cash inflow – cash outflow

= [(Salest–1)*(1+Sales growth rate)*(operating profit
margin)*(1 – cash income tax rate) – (Incremental (3.19)
fixed plus working capital investment)]
These cash flows are discounted using the cost of capital, commonly mea-
sured as Weight Average Cost of Capital (WACC). Regarding the residual
value, this is often the largest portion of the value of the firm. The residual
value can be estimated using the perpetuity method, which assumes that,
after the forecast period, a company generates returns on new investment
which are equal to the cost of capital. In other words, it is assumed that if
the company would be able to generate returns higher than the cost of
capital a new competitor will be attracted into the market driving returns
to the minimum acceptable, i.e. the cost of capital. Using the perpetuity
model, residual value can be calculated as follows:
Cash flowt+1
Residual flow = (3.20)
Cost of capital
Shareholder value is the basis for estimates of SVA, the latter being ‘the
amount of value created by the forecast scenario. While shareholder value
characterises the absolute economic value resulting from the forecasted
scenario, SVA addresses the change in value over the forecast period’
(Rappaport 1998: 49). In period t (in the forecast period), SVA is calculated
as the difference between the shareholder value at the end of the period
and at the beginning of the period.
SVAt = SHVt+1 – SHVt (3.21)
SVA can be calculated using an alternative formulation, which brings the

same SVA estimation. In detail, SVA can be calculated by estimating the
difference between the NPV of capitalised NOPAT and the present value of
incremental investment.
SVA = NPV of Change in NOPAT – PV of incremental investment (3.22)
If the liquidation value of the business is greater than its discounted cash
flow value, it is accurate to use the liquidation value in the analysis. In this
case, SVA is given as:
SVA = cumulative PV of cash flows + PV of liquidation at the (3.23)

end of forecast period – Current liquidation value
Cash flow return on investment

Cash Flow Return on Investment (CFROI)10 represents a shift from an
accounting to a cash-based framework to calculate the economic rate of
return being earned from a company’s operations. CFROI measures the
expected return of the company by applying the concept of Internal Rate
of Return (IRR) to a company and, therefore, considers both its future cash
flows and the time value of money (see Figure 3.4).
Salvage
Gross cash flow value
Asset life
Gross
investment
Figure 3.4: Cash flow return on investment
CFROI is the internal rate of return of future company’s cash flows. In

other words, CFROI is the discount rate that makes the NPV of the gross
cash flows and salvage value equal to the gross investment, and can thus be
viewed as a composite internal rate of return, in current terms. In order to
calculate CFROI, four basic inputs are necessary: (1) the gross investment
that the firm has in assets in place; (2) gross cash flow earned in the current
year on that asset; (3) expected life of the assets in place; and (4) the
expected value of the assets at the end of this life.
Gross Investment (GI) that the firm has in assets in place cannot be
directly calculated using accounting values, and it is necessary to make
some adjustments (regarding goodwill, asset depreciation and inflation)
to convert accounting values into cash flows. GI is calculated as the sum
of the net asset value (net of non-debt liabilities and intangible assets, such
as goodwill), its depreciation and an adjustment to convert the gross
investment into a current value to reflect inflation.
GI = Net Asset Value + Cumulated Depreciation on Asset + (3.24)

Current Adjustment
Similarly to GI, it is necessary to make some adjustments (regarding asset

depreciation and operating leases) to convert accounting values into cash flow
to calculate Gross Cash Flow (GCF) earned in the current year on that asset.
GCF is calculated as the sum of the after-tax operating income and the non-
charges against earnings (e.g. depreciation and amortisation). The operating
income is adjusted for operating leases and any accounting effects, in much
the same way that it was adjusted to compute economic value added.
GCF = Adjusted EBIT (1–t) + Current year’s depreciation & (3.25)

amortisation
The expected life of the assets (n) in place, at the time of the original
investment, varies from sector to sector, but reflects the earning life of the
investments in question. The expected value of the assets (SV) at the end of
this life is usually assumed to be the portion of the initial investment, such
as land and buildings that is not adjusted to current money terms.
Once calculated, CFROI can be compared with the real cost of capital in
order to assess if a company creates shareholder value: to enhance its value,
a firm should attempt to increase the spread between its CFROI and cost of
capital (measured using the WACC calculated using market value weights
for debt and equity).
Risk adjusted performance measures

Return and risk are fundamental features involved in evaluating the perfor-
mance of any company. While traditional financial ratios do not consider
risk, DCF includes risk implicitly in the discount rate, and Risk Adjusted
Performance Measures (RAPM) explicitly consider risk factors by correcting
traditional performance indicators, i.e. financial ratios given as profits over
capital as shown in Figure 3.5.
RAPM are summarised in this section by looking at two main areas.
Firstly, how can risk be measured? Following the Value at Risk (VAR)
methodology applied to portfolios of securities, risk in a company (typi-
cally in a financial firm) can be measured as Capital at Risk (CAR). Both
CAR and VAR measure risk as the maximum potential loss on capital
invested within a defined confidence interval. Various statistical methods
have been developed to estimate CAR.11 Secondly, how can financial ratios
be adjusted for risk? The risk adjustment typically focuses on:
• Returns: this indicator is labelled ‘Risk Adjusted Return on Capital’

(RAROC). In RAROC, capital is taken from the balance sheet and return
Traditional Risk Adjusted

Performance Measures Performance Measures
Risk Adjusted Return
RAROC =
Capital Invested
Return
Return
Capital Invested RORAC =
Risk Adjusted Capital Invested
Risk Adjusted Return

RARORAC =
Risk Adjusted Capital Invested
Figure 3.5: Performance measures and risk adjusted performance measures

is usually adjusted by subtracting expected losses on bad debts from

accounting net profit. One might criticise this adjustment since this
type of cost should be already considered in the cost-income state-
ment.
• Capital invested: this indicator is labelled ‘Return on Risk Adjusted
Capital’ (RORAC). In RORAC, the capital invested is not calculated
looking at the balance sheet, but is given by the CAR, which express the
equity capital sufficient to cover company’s risk.
• Return and capital invested: this indicator is labelled ‘Risk Adjusted Return
on Risk Adjusted Capital’ (RARORAC). In RAROC, both return and
capital are adjusted: return is net of the expected losses on bad debts and
capital invested is given by the CAR.
Market value added

Market Value Added (MVA) is a measure of shareholder value which is
calculated as the current market value of all capital elements minus the
historical amount of capital invested in the company.
MVA = Market Value of Capital – Invested Capital (3.26)
Regarding the calculation of MVA, it is not accurate to use book value

data for the invested capital, but it is necessary to convert accounting book
value into economic book value because accounting standards often
deviate from economic reality.12
According to many studies,13 MVA is an accurate measure of wealth
creation. Firstly, MVA expresses the cumulative amount by which a
company has enhanced/diminished shareholder wealth (see Figure 3.6).
MVA is calculated as the difference between cash in (i.e. what shareholders
put into a company as capital) and cash out (i.e. what shareholders could
obtain selling their participation). Because stock market prices are purely
expectational,14 a company’s market value is the NPV of future profits dis-
Market Value of Capital – Invested Capital = MVA
If investors expect:
N ROC = COC ⇒ Market Value of Capital = Capital invested ⇒ SHV unchanged ⇒ MVA = 0
,,E(ROC x CI)
‡ ROC > COC ⇒ Market Value of Capital > Capital invested ⇒ SHV created ⇒ MVA > 0
t =1 (1+COC)t
ROC > COC ⇒ Market Value of Capital < Capital invested ⇒ SHV destroyed ⇒ MVA < 0
where :
ROC = Return on Capital
COC = Cost of Capital
Figure 3.6: Market value added and market value of capital

counted at the company’s cost of capital. If investors expect a company to

have a Return on Capital (ROC) greater than the Cost of Capital (COC), the
market value will be higher than capital, and, as consequence, managers
have created value for shareholders: MVA will be positive. If investors
expect a company to have returns lower than its cost of capital, the market
value will be smaller than capital, and, as consequence, managers have
destroyed value for shareholders: MVA will be negative. Managers do not
create or destroy value for shareholders when investors expect company’s
return to be equal to the cost of capital, and, in this case, MVA is zero.
According to all these benefits, shareholder value creation can be
achieved by maximising MVA. Although this is theoretically indisputable,
MVA has some practical shortcomings. First of all, MVA can be calculated
only for publicly traded companies;15 secondly, MVA can be calculated
only at the firm level, but not at the level of business units; thirdly, it is not
helpful in day-to-day decision-making. For these reasons, shareholders
cannot in practice use MVA to assess how managers are creating value.
In order to address these problems, many studies have investigated the
relationship between MVA and various performance measures (such as net
income, accounting financial ratios and EVA). According to some studies
(usually undertaken by commercial promoters of the EVA metric), EVA has
the strongest statistical correlation with MVA. For example, Stern Stewart16
found that EVA (calculated using standard adjustments to General Accepted
Accounting Principles – GAAP – for all companies) statistically explains
about half of the movement in a company’s MVA, while ROE explains
about 35 per cent, cash flow about 22 per cent and Earning Per Share (EPS)
less than 20 per cent. EVA’s superiority became larger when Stern Stewart
calculated EVA and MVA making specific adjustments according to the
industry: in this case, the correlation between MVA and EVA became more
than 70 per cent. However, some other studies17 (usually undertaken by
academics) do not observe this EVA superiority.
Economic value added

This section illustrates the main features of Economic Value Added (EVA).
EVA is probably the most well-known measure of shareholder value.
Although academics and managers started to become interested in EVA in
the early 1990s, it is based on long-established basic finance principles. As
noted by Drucker (1995), ‘EVA is based on something we have known for a
long time: what we call profits, the money left to service equity, is usually
not profit at all. Until a business returns a profit that is greater than its cost
of capital, it operates at a loss. Never mind that it pays taxes as if it had a
genuine profit. The enterprise still returns less to the economy than it
devours in resources … until then it does not create wealth; it destroys it.’
The origin of EVA can be traced back to the 1950s when General Electric
developed a performance measure, labelled ‘Residual income’, calculated as
the difference between the net operating profit after tax and a charge for
invested capital (measured in terms of book value). EVA is a trademarked
variant of residual income developed by the consulting firm of Stern Stewart
& Company. EVA can be defined as the surplus value created by an invest-
ment or a portfolio of investments (shown in model 3.27). This measure can
be computed as the product of the excess return (i.e. the difference between
the return made on the investment and the composite cost of financing that
investment) made on an investment and the capital invested.
EVA = Capital Invested * (Return on Capital Invested – Cost of (3.27)

Capital)
The EVA success

EVA is currently one of the most well-known company performance indica-
tors. Some reasons can be identified for this success. Firstly, EVA is a trade-
marked variant of residual income developed by the consulting firm of
Stern Stewart & Company and, as such, EVA has benefited from con-
siderable promotion in financial newspapers, academic journals, books and
publicly available statistics.
A second reason for the success of EVA is that this measure is more than
a performance indicator since it has several advantages for shareholders, as
affirmed by consultants supporting EVA. Al Ehrbar (1998: 5), for example,
notes that when companies employ EVA to the fullest, which is what they
must do to change behaviour, it becomes far more than just another way of
adding up costs and computing profits. It is:
• The corporate performance measure that is tied most directly, both

theoretically and empirically, to the creation of shareholder wealth;
put simply, managing for higher EVA is, by definition, managing for
a higher stock price;
• The only performance measure that always gives the ‘right’ answer, in
the sense that more EVA always is unambiguously better for share-
holders, which makes it the only genuine continuous improvement
metric; in contrast, actions that increase profit margins, EPS and even
rates of return sometimes destroy shareholder wealth;
• The framework underlying a comprehensive new system of corporate
financial management that guides every decision, from annual oper-
ating budgets to capital budgeting, strategic planning, and acquisi-
tion and divestitures;
• A simple but effective method for teaching business literacy to even
the least sophisticated workers;
• The key variable in a unique incentive compensation system that, for
the first time, truly aligns the interest of managers with those of
shareholders and causes managers to think like and act like owners;
• A framework that companies can use to communicate the goals and

achievements to investors, and that investors can use to identify
companies with superior performance prospects.
Most important, EVA is an internal system of corporate governance that
motivates all managers and employees to work cooperatively and enthusi-
astically to achieve the very best performance possible.
EVA definition
EVA as a performance indicator at the firm level simply expresses the
portion of corporate profits that exceed the cost of all capital used by the
company (to achieve its profit). In other words, EVA can be computed as
the difference between after-tax operating profits (labelled ‘NOPAT’) and an
appropriate capital charge for both debt and equity.
EVA = NOPAT – (Cost of Capital * Capital Invested) (3.28)
When EVA is used to assess company performance in a given period, capital

invested and NOPAT should not be calculated in the same period. As investors
expect to receive returns on the investment made at the beginning (and not
on the cumulative amount at the end of the period), shareholders compare
returns (i.e. NOPAT) earned over the period with the capital invested at the
beginning (and not at the end) of the period. For this reason, capital invested
is measured with a lag of one year and EVA is calculated as follows:
EVAt–1,t = NOPATt–1,t – (Capital Investedt–1 * Cost of Capitalt–1,t) (3.29)
where EVAt–1,t is measured over the period (t–1, t), NOPATt–1,t is calculated
for the period (t–1, t) and capital Investedt–1 is measured at the end of
period t–1.
The components of EVA

In order to calculate EVA, the following three variables are necessary:
NOPAT, capital invested and cost of capital invested. One might claim that,
while shareholder value is mainly derived from market value figures, EVA is
mainly derived from accounting figures. Book and market figures mainly
differ for at least the following two reasons. First of all, changes in an
asset’s book value differ from changes in its market value since accounts do
not reflect changes of an asset’s market value (e.g. because the company
values the asset at its historic cost). In other words, book values are
conservative: gains are not recognised until they are realised, while
accounts recognise gains that should properly belong to the previous
period. Secondly, balance sheets do not include any goodwill18 that is not
attributable to individual assets. If investors change their expectations
about the future profitability of a company, the book value of a company’s

assets do not change, while this changes the market price of the company’s
share (as well as goodwill and shareholder value). As a consequence,
NOPAT and capital invested cannot be simply calculated using accounting
data, but it is necessary to adjust accounting figures in order to calculate
NOPAT and capital invested on an economic basis. This is achieved by
making specific accounting adjustments.
Regarding NOPAT, this is defined in the EVA calculation as the after-tax
operating income made by assets in place. However, the determination of
NOPAT is not immediate since operating income needs to consider ‘only’
the operating income made by assets in place, while the accounting
definition of operating income also considers expenses designed to gen-
erate future growth and operating expenses that are disguised financial
expenses. By comparing NOPAT with capital invested, it is possible to
obtain the Return on Capital (ROC).
NOPATec = OIec (1 – T) (3.30)

ROC = OIec (1 – T)/Capital Invested
where NOPATec is the after-tax operating income made by assets in place

(i.e. economic definition of NOPAT), OIec is the operating income made by
assets in place, this is measured as Earnings before Interest and Taxes (EBIT)
and T is the company tax rate.
Regarding the calculation of the capital invested, while this is straight-
forward for a single investment, it becomes more complex at the level of
the firm. At the firm level, the capital invested is given by the market value
of assets in place and this measure is difficult to compute since it is possible
to use a variety of proxies for the market value of assets in place. A first
measure that may be suggested is the market value of the firm. At first
sight, this solution appears correct. However, the market value of the firm
includes capital invested not just in assets in place but in expected future
growth. For this reason, this solution may be an inappropriate proxy
possibly leading to unfair valuations.
Capital invested may be also measured focusing on book value of capital.
This alternative proxy solves the problems associated with the aforemen-
tioned measure, but it also has some limitations. First of all, book values
reflect not just the accounting choices made in the current period but also
accounting decisions made over time on how to depreciate assets, value
inventory and deal with acquisitions. Furthermore, book value is also
influenced by the accounting classification of expenses into operating and
capital expenditures, with only the latter showing up as part of capital. In
order to resolve these shortcomings, several accounting adjustments have
been proposed to make the book value of capital a better measure of capital
invested.
Regarding the estimation of the cost of the capital invested, it is neces-

sary to analyse how capital invested has been financed. A company can
fund its activity by two financing sources: equity (i.e. also labelled ‘risk
capital’) and debts (i.e. also labelled ‘external capital’). Since companies
usually use both sources to finance their operating activities, the cost of
capital invested has to be calculated taking into account both sources and
incorporating the extent to which the different sources of funds have been
employed. The measure commonly used for the cost of capital invested is
known as the Weighted Average Cost of Capital (WACC): this is the
average between the cost of equity and the cost of debt capital weighted for
the effective proportion of debt and equity employed in the company.
Equity
WACC = Cost of Equity (3.31)
Equity + Debt
Debt
+ Cost of Debt
Equity + Debt
The calculation of WACC requires three main inputs: cost of equity, cost
of debt and weights. Regarding cost of equity of a firm, this can be esti-
mated using several methods. The most common approach measures the
Cost of Equity of a firm in terms of the investor’s expected rate of return.
Since investors are usually assumed to be risk averse, investor’s expected
rate of return can be calculated by adding a risk premium, which is the
extra return required by investors to assume a risk, to the return of a risk-
free asset (Rf).
Cost of equity = Investor’s Expected Returns [E(R)] (3.32)

= Rf + Risk Premium
There are therefore two inputs for calculating a company’s cost of equity:
the Risk Free Rate and Risk Premium. The first can be estimated taking the rate
of return of a short-term government bond (e.g. the shortest Treasury Bill):
this security can be reasonably considered risk-free since it has a very low
credit risk, liquidity risk, market risk and interest rate risk. Regarding the cal-
culation of the risk premium, this is commonly calculated applying the
Capital Asset Pricing Model (CAPM). For any risky assets, investors expect to
receive a return which is proportional to the ‘market portfolio risk premium’
(i.e. the difference between the market portfolio return and the risk-free rate)
and to the systematic risk of the share (i.e. the share risk not eliminable by
portfolio diversification: this is measured using an index labelled ‘Beta’).
Risk Premium = β * market portfolio risk premium (or equity (3.33)

market premium)
= β (Rm – Rf)
Regarding the cost of debt of a firm, this expresses the cost to the firm of
borrowing funds to finance its projects. The cost of debt is given by the
average of the cost of all borrowed funds weighted to their current value.
Since this information is often not publicly disclosed, external analysts
cannot measure exactly this cost and it is necessary to use a proxy. The cost
of debt is influenced by three variables: the current level of risk-free interest
rates, the default risk of the company and the tax advantages associated
with debt. As such, the cost of debt is commonly estimated as the sum of
the risk-free rate and a default spread (often expressed in terms of a credit
rating) net of taxes.
Cost of debt = (RFR + Default spread) (1–T) (3.34)
where RFR is rate of return of a risk-free asset (i.e. Risk Free Rate), default
spread is the interest rate premium required by investors to purchase risky
bonds and T is the average taxation rate. This approach is straightforward
for all companies having an external credit rating. For other companies,
this approach becomes problematic and it is necessary to use another
proxy, such as (for example) the ratio between interest expenses and the
total amount of debts.
Regarding weights, there is debate as to the appropriate weights that
should be used for debt and equity and if these are to relate to market or
book value. The book value approach can be supported with the following
three arguments: (1) book value is more reliable than market value since it
is less volatile; (2) the adoption of book value is a more conservative
approach to estimating debt ratios since the return on capital and capital
invested are measures based on book value; and (3) lenders will not lend on
the basis of market value. This solution finds little support in the academic
literature, however: for example, Damodaran (2001: 46) notes that ‘The
weight assigned to equity and debt in calculating average cost of capital
have to be based on market value, not book value. This rationale rests on
the fact that the cost of capital measures the cost of issuing securities –
stocks as well as bonds – to finance projects, and these securities are issued
at market value, not at book value.’ Starting from this statement, it is possi-
ble to criticise (for a generic company) the book value approach for the fol-
lowing two reasons: (1) capital invested is given by the market value of
assets in place. In calculating its cost, the market value of assets should be
considered rather than the book value of capital. By using a book value cost
of capital, analysts implicitly assume that all debt is attributable to assets in
place, while all future growth comes from equity. In other words, if we
adopted this rationale in valuation, we would discount cash flows from
assets in place at the book cost of capital, and all cash flows from expected
future growth at the cost of equity; (2) the adoption of a book value
measure for the cost of capital for all economic value added estimates,
including the portion that comes from future growth, is not coherent with
the basis of the approach (i.e. maximising the present value of economic
value added over time is equivalent to maximising firm value). The adop-
tion of market value for weights for debt and equity is therefore generally
preferred. As noted by Rappaport (1998: 37), ‘there is widespread agree-
ment in finance texts about the conceptual superiority of market values,
despite their volatility, on the ground that to justify its valuation the firm
will have to earn competitive rates of return for debt holders and share-
holders on their respective current market values.’
The EVA adjustments

The calculation of EVA requires that NOPAT and capital invested are
converted to an economic basis. For this reason, advocates of EVA suggest
the following adjustments in order to: (1) avoid mixing operating and
financing decisions; (2) provide a long-term perspective; (3) avoid mixing
flows and stocks and to convert accounting standards accrual items to a
cash-flow basis or, in other cases, convert accounting standards cash-flow
items to additions to capital.
Uyemura et al. (1996: 101), for example, note that ‘To date, over 160
potential accounting adjustments have been identified and catalogued’ in
the context of EVA adjustments. Some of these are, however, difficult to
make by external analysts and may not be strictly necessary. In this regard,
Damodaran (1999a: 36) observes two points: ‘First, many of these adjust-
ments are at the margin and do not affect economic value added very
much. Second, many of these adjustments can be made only by someone
who either works with the management of the firm (as a consultant) or as
an insider. External analysts who choose to use EVA have to accept the
reality that their estimates of operating income can adjust only for
the variables (such as R&D and operating lease expenses) on which there is
Figure 3.7: EVA spectrum
Basic Disclosed True

EVA EVA EVA
Tailored
EVA
Source: Modified from Al Ehrbar (1998: 165).

public information.’ Similarly, Uyemura et al. (1996: 101) note that

‘For any single company, however, it is rare to make more than 10 adjust-
ments of GAAP accounting.’ Regarding this point, advocates of EVA recog-
nise that there may be several EVA results according to the number of
accounting adjustments. It is possible to identify a spectrum of EVA values
(shown in Figure 3.7), such as basic, disclosed, true and tailored EVA.
Basic EVA is a value obtained using unadjusted GAAP operating profit
and GAAP balance sheet capital. This measure is an improvement on tra-
ditional accounting measure (since it considers a capital charge), but it is
affected by accounting standard problems.
Disclosed EVA is a value obtained making some standard adjustments to
publicly available accounting data. This measure improves the basic EVA by
solving some main accounting standard problems. The disclosed EVA is
usually adopted by external analysts: for example, disclosed EVA is used by
Stern Stewart to develop its database (i.e. Stern Stewart 1000 performance
database) and, as seen earlier, this measure can explain about 50 per cent of
changes in MVA. When internal information is available, disclosed EVA
is not a satisfactory management tool since more precise information is
available.
True EVA is a value obtained using all internal data that reflect the true
economic condition of the company. The true EVA is the most theoreti-
cally correct measure of economic profits using internal data, but its cal-
culation would require detailed company information that is unavailable to
outside researchers, such as academics and/or analysts.
Tailored EVA is a value obtained using specific internal information (e.g.
organisation structure, business mix, strategies, accounting mix) to adjust
accounting figures. In this case, researchers use a part of available internal data
balancing the trade-off between simplicity and precision. Uyemura et al.
(1996: 101) note that the following four filter criteria should be used to select
the necessary EVA adjustments: ‘(1) Materiality: the adjustments should be
significant enough to be material to performance measurement; (2) Effect on
behaviour: it should drive value increasing behaviour. It should not cause dis-
tortions or promote manipulation; (3) Ease of understanding: it should be
clearly understood by management; (4) Data availability: the data needed to
make the adjustment should be accessible and available for reporting.’
In order to calculate the disclosed EVA, the standard adjustments are the
following:
(1) Eliminate (from a company’s NOPAT and capital invested) financial dis-
tortions to avoid mixing operating and financing decisions. In detail, it is
necessary to:
(a) remove the effects produced by corporate leverage on company’s
net income. As such, NOPAT and capital invested should be
adjusted as follows:
NOPAT = NOPATAcc + Interest paid on company’s debt (3.35)

Capital invested = Book value of Capital + total amount
of company’s debt
where NOPATAcc is the accounting NOPAT, that is defined as the

EBIT net of tax [i.e. NOPATAcc = EBIT (1– tax rate)]. By adding back
interest costs (net of tax shields) to accounting NOPAT, one obtains
the: (1) NOPAT that a company would achieve by financing all
assets using equity capital; (2) the Return on Capital (ROC), i.e. the
ratio between NOPAT and capital invested, if the company uses
only equity capital to finance its activities. These adjustments do
not imply that leverage is not important in assessing corporate per-
formance: however, since external debts (and therefore, interest
costs and tax shields) are considered in EVA by calculating the
opportunity cost of capital invested, it is accurate to delete all
financial distortions in capital invested and NOPAT.
(b) consider all operating expenses (e.g. operating lease expenses)
that are true disguised financial expenses. For example, to adjust
NOPAT and capital invested for operating expenses, it is necessary
to: (1) add back operating lease expenses to operating income;
(2) capitalise operating lease expenses: in other words, it is neces-
sary to estimate the present value of expected lease commitments
over time, and treat them as the equivalent of debt, which will
increase capital invested in the firm:
EBITOp. Leases = EBIT + Operating Lease Expenses

Capital invested = BV of Capital + Present Value of (3.36)
Operating Lease Expenses
(2) Eliminate (from a company’s NOPAT and capital invested) accounting stan-
dard distortions. Since NOPAT and capital invested are considered from
an economic perspective (rather than the accounting one), it is neces-
sary to remove all distortions introduced by conservative accounting
practices. First of all, it is necessary to: (1) add back to company’s
capital invested all ‘equity equivalent reserves’, such as deferred tax
reserves, general risks reserves, etc.; and (2) add back to company’s
NOPAT the provision for these equity equivalent reserves.
NOPAT = NOPATAcc + Provisions for equity equivalent

reserves
Capital invested = Book value of Capital + Equity (3.37)
equivalent reserves
The goal of these adjustments is to make NOPAT closer to the cash

flows generated by the company’s capital invested (i.e. the accurate
measure for assessing shareholder income). It is necessary to note that

most of these adjustments depend on the kind of company being
analysed. While specific adjustments regarding banks are analysed later
in this book, the following adjustments are common to all companies:
(a) Some expenses (e.g. research and development (R&D) expenses,

training expenses, etc.) are designed to generate future growth.
Assets in place do not benefit from these expenses and it would be
incorrect to reduce operating income due to these expenses. How-
ever, accounting standards require companies to treat all outlays
for R&D as operating expenses in the income statement. As a con-
sequence, it is appropriate to correct this accounting distortion by:
(1) considering NOPAT without these expenses; (2) adding back to
capital invested these capitalised expenses. In other words, these
expenses need to be capitalised and amortised over time: for
example, according to Stern Stewart’s statistics, five years is the
average useful life of R&D expenses.
EBITR&D = EBIT + R&D Expenses + training expenses (3.38)

Capital invested = Book value of Capital + Capitalised R&D
(b) Strategic investments whose payoff is not expected to come until

some point in the future need to be suspended. These investments
are not computed in capital invested, but are included in a special
‘suspension’ account. In the interim, the capital charge that would
have been required to be made to the suspension account is
omitted from the calculation of EVA. In this way, the balance of
the suspension account expresses the full cost (i.e. real cost plus the
cost opportunity) of the investment. The suspension account is
introduced back into the EVA calculation as soon as the invest-
ment is scheduled to produce operating profits.
In the interim
Capital invested = BV of Capital – Suspension account
Suspension account = Strategic investment + Capital charge
(over time when investment is scheduled to produce (3.39)
operating profits)
Capital invested = BV of Capital + Suspension account
(c) Some items (e.g. the amortisation of goodwill) may reduce, for
accounting reasons, the book value of capital, but does not have a
real impact on capital invested. This needs to be eliminated by
adding back these expenses to capital invested.
Capital invested = BV of Capital + Purely accounting (3.40)

reduction of Capital invested
A comparison among metrics

The previous sections have presented the main characteristics of various
innovative approaches that have been developed to assess companies’ per-
formance. These methods have been promoted mainly by consulting firms
to address the limitations of traditional performance measures in terms of
value creation for shareholders. There are three main issues to be consid-
ered in defining an accurate measure of shareholder value added: (1) share-
holder value can be only assessed considering economic (rather than
accounting) profit and, therefore, accounting figures need to be adjusted
for calculations relating to shareholder value; (2) performance cannot be
evaluated without considering risk as it would be dangerous to assess a
company’s returns without knowing its risk and; (3) as in any investment,
it is necessary to consider capital invested by shareholders.
All techniques analysed have attempted to improve traditional perfor-
mance measure in at least one of these three aspects as summarised in
Table 3.2. Among these techniques, DCF and EVA approaches are currently
well known and diffused among academics and practitioners. Regarding
Table 3.2: Innovation of performance measures

standard problems
Capital invested2
Capital charge4
Company Risk3
Completeness1
Accounting
Type
Performance
Measure
Operating margin No Yes No No No Absolute value

Net profit Yes Yes No No No Absolute value
ROE /ROA /ROI Yes Yes Yes No No Rate of return
DCF Yes No No Yes No Absolute value
SHV Yes No Yes Yes Yes Absolute value
CFROI Yes No Yes Yes No Rate of return
RAPM Yes No Yes Yes No Rate of return
EVA Yes No Yes Yes Yes Absolute value
1
Completeness: expresses if a performance measure considers both core and non-core business
activity
2
Capital invested expresses if a performance measure explicitly considers the amount of capital
invested in the company
3
Company risk expresses if a performance measure explicitly takes into account the company’s
risks
4
Capital invested expresses if a performance measure considers a capital charge based on the
capital invested in the company and the company’s risk
the first, the DCF approach is less homogeneous compared to other mea-
sures: while the others provide short-term indicators of corporate perfor-
mance, DCF aims to estimate a company’s value (rather than assessing its
corporate performance) by discounting all future cash flows. By applying
DCF, analysts can receive either:
• an ‘objective’ assessment of a company’s performance that, however,

does not accurately capture the shareholder value created in a single
period. This happens when a company’s performance in a given period
is measured as the difference between the company’s value at the end
and at the beginning of the period. In this case, performance is assessed
by focusing on cash flows over the period analysed (e.g. annually or
quarterly) and this may be inadequate for accurately assessing corporate
performance since some cash outflow may be generated by expenses
(e.g. R&D expenses, training expenses, but also fixed investment)
designed to generate future growth. For instance, in order to evaluate
the performance of a new company, all initial cash flows are negative,
but these cannot be considered as an indication of poor performance
(see example in Figure 3.8); or
• an ‘accurate’ assessment of a company’s performance that, however, is
available at the end of company’s life. As noted by Di Antonio (2002),
an accurate measure of a company’s performance can be obtained by
comparing the total end-of-lifetime company’s value with the initial
Assume one assessed a company's performance over the period (T, T+1), when there
are three negative cash flows to finance the company's core business. In the following
period cash flows are positive.
T T+1 T+2 T+3 Time

-Cf0--Cf1 -C2 +Cf3 +Cf4 +Cf5 +Cf6 +Cf7 +Cf8... ...
Cash flow
By applying the DCF model the company's performance is measured as follows:

Performance = Value at T+1 – Value at T =
n n 3
Cfi Cfi Cfi
= ‡” ‡” = ‡”
i=0 (1+WACC)i i=3 (1+WACC)i i=0 (1+WACC)i
Company's performance over the period (T, T+1) is given by the NPV of cash
flow over the period assessed. However, this value has little significance in terms of
performance assessment since these cash flows are not operating expenses,
but investment.
Figure 3.8: Assessing corporate performance using DCF

investment since DCF requires the knowledge of all cash flows over the
entire life of the company. The assessment obtained is an accurate indi-
cator of the company’s performance over the entire life of the company,
but it is available too late.
It is worthwhile to note that, although the DCF approach is theoretically

straightforward, its practical application has several difficulties. The most
relevant problem concerns the subjective forecasting of future company’s
cash flows.
Regarding EVA, this measure explicitly recognises the need of incorporat-
ing risk, capital invested and the economic view in a performance measure.
EVA is obtained subtracting from the economic company’s profit [Re]
(which is obtained adjusting the accounting NOPAT in order to proxy
company’s net operating cash flow) a capital charge [K], which is the
expected return required by all investors (debtholders and shareholders)
and is correlated to the company’s risk.
EVA – Re – K (3.41)
EVA is therefore a transparent measure of the value created by a company

for its shareholders in comparison to other investments with the same level
of risk. In order to show this result in relative terms (i.e. a sort of rate of
shareholder creation), it is necessary to compare EVA with capital invested
(CI). This indicator of shareholder value, which expresses the rate to which
a company has created value added for its shareholders over a given period
of time, can be formulated as follows:
EVA = Re – K
Shareholder value rate of return = (3.42)
CI CI
For the purpose of assessing a company’s performance, EVA seems to

be a preferred shareholder value indicator and its maximisation can
be taken as a primary corporate goal. However, one might claim that,
according to traditional corporate finance theory, the primary corporate
goal should be firm value maximisation. Damodaran (1999a: 44) notes that
‘a policy of maximising the present value of EVA over time is equivalent
to a policy of maximising firm value’ by comparing EVA and DCF. The
starting point is the definition of the company’s value, i.e. this is given
by the value of the assets in place and the expected value of the future
growth.
Firm value = Value of assets in place + Value of expected future (3.43)

growth
According to the DCF framework, the value of an asset can be viewed as the
net present value (NPV) of the expected cash flows on that asset. Firm value
can be rewritten as:
Firm value = Capital Invested + NPV Assts In Place + ∑ NPV (3.44)

Future Projects
It is now necessary to define the present value of an asset. Consider an

asset with the following characteristics: an initial investment of I, an
expected life of n years, a salvage value of SVn ; depreciation of Deprt in
the year t; the firm produces an EBITt at the marginal tax rate of t; the
company’s cost of equity is Ke . The NPV of this asset can be calculated as
follows:
t=n (EBITt (I – t) + Deptt) (SV)n

NPK = ∑ + +I (3.45)
t=1 (I + Kt)t (I + Ke)n
Consider now an investment in an alternative asset which requires an

initial investment of I, earns exactly the cost of equity and allows for the
entire investment to be salvaged at the end of the project, the NPV of the
project will be zero. Solving for I, it is possible to obtain:
t=n Ke(I) I
I=∑ t
+ (3.46)
t=1 (I + Kt) (I + Ke)n
Substituting this into the model 3.45, the NPV of the original project is:
t=n (EBITt (I – t) + Deptt) (SV)n t=n Ke(I) I

NPV = ∑ t
+ n
+ ∑ n
+ (3.47)
t=1 (I + Ke) (I + Ke) t=1 (I + Kt) (I + Ke)n
It is now possible to simplify this model by assuming: (1) the salvage value
of the project is zero; (2) the NPV of depreciation is equal to the NPV of the
initial investment, discounted back over the project life. The model can be
rewritten as:
t=n (EBITt (I – t) t=n Ke(I)

NPV = ∑ t
+ ∑ t
(3.48)
t=1 (I + Ke) t=1 (I + Ke)
(EBIT (I – t)
Since ROC = , the model becomes:
I
t=n (ROCt – Ke)(I) t=n EVAt

NPV = ∑ + ∑ (3.49)
t=1 (I + Ke)t t=1 (I + Ke)
t
As a result, the NPV of an asset is the present value of the EVA by that asset
over its life. However, this relationship is not observed when the salvage
value is large and/or the present value of depreciation tax benefits is greater
than or lesser than the present value of the capital invested. It is now pos-
sible to define firm value by restating model 3.44 substituting NPV of
the asset in place (NPVAsset in place) and the sum of all future projects
(∑ NPVFuture projects) with the present value of the EVA generated by these
assets over their life.
t=n EVAt, Assets in place

Firm Value = Capital invested + ∑ (3.50)
t=1 (I + Ke)t
t=n EVAt, Future projects
+ ∑
t=1 (I + Ke)t
As a result, the value of a company is the sum of three components: the

capital invested in assets in place, the present value of the EVA by these
assets and the expected present value of the EVA generated by future
investments. The above provides evidence that a policy of maximising the
present value of EVA over time is equivalent to a policy of maximising firm
value.
A literature review of empirical studies dealing with shareholder

value
This section summarises the main findings of previous empirical studies
dealing with shareholder value in banking. Analysing the most recent con-
tributions in the literature, it is possible to identify three branches of
empirical study:
(a) Some studies, usually proposed by consulting companies (such as Stern

Stewart MVA Performance ranking) or financial information providers
(such as the Financial Times (FT’s) European Company performance
survey), attempt to measure the value created by an industry or a
country by using their own proprietary shareholder value measures.
(b) Other studies, undertaken by consultants and academics, deal with the
value relevance of summary accounting data measures. Among these,19
some focus on the information content of traditional and innovative
performance measures in creating shareholder value by looking at their
statistical association with stock market returns. These studies propose
an empirical investigation without developing a formal model of the

relationship between market and accounting figures: their goal is to
compare different performance measures and establish the one with
the greatest ability in explaining the creation of shareholder value.
(c) Finally, there is a development in the academic literature that investi-
gates the relationship between share prices and accounting figures
(namely, residual income) by testing various models developed by
Ohlson (1995) and Feltham and Ohlson (1995).
The first two streams of literature are summarised in the following sections.
Shareholder value in European banking

Various empirical studies investigating shareholder value have attempted
to measure the value created by an industry by using specific shareholder
value performance measures. This type of study has usually been under-
taken by consulting companies or financial information providers.20 This
section summarises some of these findings made publicly available by the
Financial Times in 2001 and Stern Stewart & Co in 1998–2000.
Regarding the Financial Times European Company performance survey
(produced in conjunction with FTSE, the global equity index provider), this
ranks Europe’s biggest companies – those that are constituents of the FTSE
world index – according to their total returns to shareholders over the past
year and the past five years. Total Shareholder Return (TSR) is defined as
the percentage gain (or loss) received by a shareholder over the period,
assuming that all dividends distributed by the company are immediately
reinvested in its shares. At the time of writing (mid-2005) the latest pub-
licly available information is dated 31 March 2001. Table 3.3 reports the
average TSR by distinguishing for sectors and over two different time
periods (i.e. one and five years). In terms of TSR, banks are ranked the 8th
sector in terms of shareholder value creation, with a 16 per cent gain
received by shareholders over the period March 2000–1. Over a longer
period (namely, five years), the ranking among sectors changes: e.g.
tobacco is not the top value creator sector, but is ranked as 7th and banks
are ranked as 5th, with an overall 250 per cent gain received by sharehold-
ers over the period March 1996–2001, with a mean average annual growth
rate of about 50 per cent.
Focusing on European banking, UK banks achieved the highest TSR
over the period March 2000–1: four of the top ten banks are British, while
the remaining are from France, Denmark, Austria, Sweden, Ireland and
Norway. Among the German banks, the highest TSR is obtained by
Deutsche Bank (ranked 13th among European banks and 84th among
European companies), while the best Italian bank was Rolo Banca (ranked
14th among European banks and 86th among European companies).
Focusing on the four main European countries, among the fifty-five banks
Table 3.3: FT’s European company performance survey: best European sector performers
One year to 31 March 2001 Five years to 31 March 2001
Rank European Sector TSR % Rank Sector TSR %

1 Tobacco 38.5 1 Information technology/ 332.9
hardware
2 Beverages 28.9 2 Personal care & household 256.6
products
3 Personal care & household 23.8 3 Life assurance 255.9
products
4 Food producers & processors 21.2 4 Telecommunications services 255.8
5 Mining 21.2 5 Banks 250.9
6 Pharmaceuticals 19.9 6 Speciality & other finance 236.9
7 Real estate 17.9 7 Tobacco 234.1
8 Banks 12.6 8 Oil & gas 205.6
9 Health 11.0 9 Insurance 198.6
10 Food & drug retailers 9.9 10 Electronic & electric equipment 198.5
Source: www. http://specials.ft.com/europerformance, 25 July 2001.
ranked in the FT’s European company performance survey, ten are

British, eight are Italian, five are German and three are French. Table 3.4
summarises the situation by showing the rank distribution by quartile.
Analysing TSR over a longer period (namely, five years – March
1996–2001), the situation changes since Italian and British banks seem to
have achieved better results than French and German banks. Among the
top ten banks, three are Italian (namely, Unicredito is ranked first with an
overall gain for shareholders of 564 per cent, Banca Intesa is 4th with a TSR
of 443 per cent and San Paolo IMI is 10th with a TSR of 323 per cent ) and
three others are British (namely, Royal Bank of Scotland is ranked 2nd with
an overall gain for shareholders of 482 per cent, Barclays is 5th with a TSR
of 395 per cent and Bank of Scotland is 7th with a TSR of 361 per cent). No
German or French bank is listed in the top ten. The best performing
German bank is Dresdner Bank, ranked 24th with a TSR of 190 per cent,
while the top French bank is Société General, ranked 13th with a TSR of
301 per cent.
The discussion above focuses on the ability of European banks to create
shareholder value based on the TSR, which measures the percentage gain
(or loss) received by a shareholder over the period. Since this measure
enables one to assess shareholder value as rate of change, it is worthwhile
analysing European banks using an absolute measure of shareholder value,
such as EVA or MVA. This type of information is made available by Stern
Stewart & Co. that periodically publishes the ranking of the top value
creator companies focusing on EVA and MVA.
Table 3.4: FT’s European company performance survey: bank sector total shareholder return
78
Company Country Sector Overall TSR % Market Dividend % change Sector Overall TSR %
rank rank 1 year capital % yield 1 yr to rank rank 5 yrs
1 year 1 year 30/03/01 30/03/01 30/03/01 5 yrs 5 yrs
Royal Bank of Scotland UK 1 5 85.4 68,684.6 2.1 67.1 2 14 482.0

Danske Bank Denmark 2 12 68.7 13,620.1 3.3 –83.6 9 39 331.0
Northern Rock UK 3 20 60.6 3,434.3 3.1 54.0
Société Generale France 4 45 40.1 29,609.9 3.3 –66.4 13 51 301.5
Abbey National UK 5 50 38.1 25,812.6 4.1 31.4 18 73 242.0
DNB Hldgs Norway 6 56 34.8 4,003.5 5.4 28.1 23 110 191.2
Erste Bank Austria 7 60 33.7 2,936.0 2.1 29.6
Barclays UK 8 64 31.4 58,491.7 2.6 27.4 5 26 395.3
Svenska Handelsbanken Sweden 9 66 31.1 11,709.8 2.7 27.6 12 48 308.7
Bank of Ireland Ireland 10 70 30.5 9,366.3 2.7 27.2 8 37 337.0
Nordea Sweden 11 76 29.3 20,758.7 3.2 25.1 6 31 370.1
Jyske Bank Denmark 12 79 28.9 938.8 0.0 25.5 28 138 161.0
Deutsche Bank German 13 84 27.2 53,079.2 1.3 24.9 30 147 156.2
Rolo Banca Italy 14 86 26.8 8,945.4 4.4 20.6
UBS Switzerland 15 94 24.8 72,421.6 2.2 –40.5 25 117 186.4
Monte del Paschi di Siena Italy 16 103 23.7 9,619.3 2.0 21.0
Banca Intesa Italy 17 104 23.6 21,804.7 2.1 20.2 4 19 443.4
Dresdner Bank Germany 18 109 22.6 26,761.0 1.7 19.5 24 112 190.0
KBC Belguim 19 114 21.9 13,220.8 3.2 18.1 31 163 141.1
UniCredito Italiano Italy 20 119 21.3 24,512.2 2.6 17.8 1 3 564.4
Banco Popular Spain 21 124 20.4 8,110.7 3.1 16.7 26 128 173.0
BNP Paribas France 22 127 19.4 42,700.6 2.8 15.5 15 61 273.5
Dexia Belgium 23 131 18.6 16,255.7 2.6 15.2
Allied Irish Banks Ireland 24 136 17.0 10,053.0 3.4 13.1 17 65 258.1
Fortis* Belgium 25 150 14.2 38,391.7 3.0 11.0 20 79 226.3
HSBC UK 26 154 13.3 125,266.3 3.6 9.9 11 44 317.9
Table 3.4: FT’s European company performance survey: bank sector total shareholder return – continued
1 year 1 year 30/03/01 30/03/01 30/03/01 5 yrs 5 yrs
Alliance & Leicester UK 27 157 12.6 5,951.0 4.5 6.6

San Paolo-Imi Italy 28 166 10.8 21,458.6 3.7 6.8 10 41 323.4
Almanij Belgium 29 167 10.6 7,770.7 3.1 7.2 34 200 108.4
Credit Lyonnais France 30 170 9.3 14,280.9 1.8 7.0
Halifax UK 31 177 8.2 25,945.6 3.7 3.6
Banca di Roma Italy 32 197 5.7 6,550.4 1.1 4.0
Lloyds TSB UK 33 200 5.2 61,185.7 4.4 0.8 16 63 266.6
BBVA Spain 34 212 3.2 49,375.9 2.0 0.6 3 16 457.4
Banco Espirito Santo Portugal 35 217 2.4 3,360.0 3.0 –32.8
Banca Nazionale del Lavoro Italy 36 220 1.9 7,492.0 2.3 –0.3
Standard Chartered UK 37 231 –1.1 15,414.5 3.0 –3.8 32 182 124.1
Credit Suisse Switzerland 38 233 –1.2 59,979.3 2.3 –3.8 21 80 225.7
Bank of Scotland UK 39 239 –2.3 14,641.6 2.0 –4.9 7 32 361.7
BCP Portugal 40 245 –3.3 11,523.8 2.9 –8.0
Bayerische Hypo-und Vereinsbank Germany 41 247 –3.7 31,881.2 1.4 –5.5 22 105 194.6
Banco Santander Spain 42 257 –5.5 47,211.9 2.6 –7.8 14 53 295.7
Skandinaviska Enskilda Banken Sweden 43 265 –6.3 7,131.7 4.3 –9.9 27 130 172.4
Foreningssparbanken Sweden 44 267 –7.0 6,682.5 4.8 –10.9
ABN Amro Hldgs Netherlands 45 268 –7.4 31,132.1 4.3 –11.0 29 139 161.0
Commercial Bank of Greece Greece 46 316 –17.2 4,289.7 1.8 –18.9
Commerzbank Germany 47 320 –18.1 16,130.8 2.6 –20.5 33 198 111.5
National Bank of Greece Greece 48 353 –24.4 8,794.8 2.2 –47.4
EFG Eurobank Ergasias Greece 49 380 –32.0 5,194.1 1.3 –48.5
Alpha Bank Greece 50 396 –35.7 4,762.8 3.1 –58.3
79
80
Table 3.4: FT’s European company performance survey: bank sector total shareholder return – continued
1 year 1 year 30/03/01 30/03/01 30/03/01 5 yrs 5 yrs
BankGesellschaft Berlin Germany 51 412 –40.4 2,137.2 6.1 –42.7

Bankinter Spain 52 415 –41.5 2,813.9 2.3 –42.6 19 74 241.0
Piraeus Bank Greece 53 417 –41.5 2,096.1 5.4 –44.1
NIBID Greece 54 440 –50.4 496.5 25.7 –95.8
Bipop-Carire Italy 55 443 –50.8 10,314.1 1.3 –95.1
Source: www. http://specials.ft.com/europerformance, 25 July 2001.

Among the four largest European banking systems Stern Perfor-

mance rankings only list banks from Italy and the UK among the 100 top
companies. The latest publicly available statistics focus on the year
2000.
Regarding Italy Stern Stewart’s MVA Performance ranking shows that
there are twenty-six commercial banks among the top 100 shareholder
value creator companies in 2000. Focusing on the largest banks, the Stern
Stewart MVA Performance ranking seems to be consistent with the FT’s
European Company Performance Survey over five years. In detail:
• Unicredito Italiano, ranked 7th among the Italian companies with the
largest MVA in both 1999 and 2000, has the largest MVA among Italian
banks. Unicredito Italiano’s MVA was Euro 14,877 million in 1999 and
Euro 16,584 million in 2000. In both years, Unicredito Italiano created
substantial EVA, namely Euro 344 million in 1999 and Euro 968 million
in 2000.
• San Paolo-IMI improved its ranking between 1999 (12th among the
Italian companies with the largest MVA and 5th among the Italian
banks with the largest MVA) and 2000 (8th among the Italian com-
panies with the largest MVA and 4th among the Italian banks with the
largest MVA). San Paolo-IMI’s MVA increased from Euro 9,315 million in
1999 to Euro 14,235 million in 2002. While San Paolo-IMI destroyed
value in 1999 (its EVA is Euro –49 million since the return of capital
invested (i.e. 10.5 per cent) was smaller than the cost of capital invested
(i.e. 11.0 per cent)), San Paolo-IMI created substantial EVA (Euro
1,964 million) in 2000 that is the highest among Italian banks and the
second largest among Italian companies.
• Banca Intesa was ranked 20th among the Italian companies with the
largest MVA (7th among the Italian banks with the largest MVA) and
12th in 2000 (4th among banks). Banca Intesa’s MVA increased by
233 per cent: from Euro 4,734 million in 1999 to Euro 10,919 million
in 2000. Nevertheless, Banca Intesa destroyed value over 2000 with a
negative EVA (Euro –215 million).
• Banca Commerciale Italiana improved its ranking passing from 22nd to
18th among the Italian companies with the largest MVA. MVA increased
from Euro 3,810 million to Euro 6,636 million between 1999 and 2000.
The EVA created in 2000 was Euro 159 million.
• Monte Paschi Siena was ranked 23rd among the Italian companies with
the largest MVA both in 1999 and 2000 (respectively, Euro 3,790 million
and Euro 4,131 million). While Monte Paschi di Siena destroyed value in
1999 it obtained a negative (EVA was –27 Euro million since the return
of capital invested (i.e. 9.9 per cent) was smaller than the cost of capital
invested (i.e. 10.4 per cent)), this bank created substantial value in 2000
(EVA was Euro 769 million).
• Banca Nazionale del Lavoro’s (BNL) ranking worsened from 37th in

1999 to 39th in 2000 among the top MVA companies despite the fact
that its MVA increased by 11.4 per cent (from Euro 1,491 million to
Euro 1,660 million). It is important to note that BNL’s MVA was nega-
tive in 1998 and 1997, when it was ranked 93rd and 100th. Both in
1999 and in 2000, BNL’s EVA was negative (Euro –627 million in 1999
and Euro –173 million in 2000).
• Banca di Roma was ranked 100th among the Italian companies with the
largest MVA in 1999 and 99th in 2000. Both in 1999 and in 2000, Banca
di Roma destroyed substantial value (its EVA was Euro –163 million and
Euro –1,052 million, respectively) and its MVA was negative (Euro
–1,698 Euro million and Euro –2,717 Euro million, respectively).
Regarding the United Kingdom, the Stern Stewart MVA Performance

ranking21 shows that there are eight banks and building societies among
the top fifty shareholder value creators. Focusing on the main banks and
building societies:
• HSBC Holdings PLC created the 5th largest MVA in the UK and the first
among all banks: HSBC’s MVA was GBP 42,585 million and the bank
achieved an EVA of GBP 621 million, due to a return on capital invested
of 12.3 per cent and a cost of capital of 11.2 per cent.22 In 2000, HSBC
Holdings PLC MVA is USD 39,370 million and became the 6th largest
MVA creator in the UK (and it was still the first among all banks). The
EVA generated in 2000 was USD 830 million, due to a return on capital
invested of 11.0 per cent and a cost of capital of 10.0 per cent.
• Lloyds TSB Group PLC was the second largest bank MVA creator (the
10th out of all UK companies) with an MVA of GBP 26,713 million: it is
interesting to note that, in 1999, Lloyds TSB Group PLC created more
value for its shareholders than HSBC since it achieved a higher EVA
(GBP 1,027 million), due to a higher return on capital (17.1 per cent)
and a cost of capital of 11.2 per cent. In 2000, Lloyds TSB Group PLC’s
ranking declined, becoming the 15th largest MVA creator out of all
British companies with an MVA of USD 13,464 million: the EVA gener-
ated was only USD 13.3 million with a return on capital invested slightly
higher than its cost of capital invested (i.e. 10.0 per cent).
• The third largest MVA creator in the UK was Barclays PLC, ranked 15th
among British companies, with an MVA of GBP 14,412 million. This
bank created an EVA of GBP 512 million, which was lower than the EVA
created by HSBC: however, it is interesting to note that Barclays PLC
achieved a larger gap between return on the capital invested (i.e. 3.9 per
cent) than that of HSBC (i.e. 1.1 per cent). In 2000, Barclays PLC
improved its ranking becoming the 7th largest MVA creator out of all
British companies with an MVA of USD 27,391.6 million: the EVA gen-
erated was USD 1,672.7 million by further increasing the gap between
the return on capital invested (18.0 per cent) and a cost of capital
invested (10.0 per cent).
Focusing on these banks, it is interesting to note that Lloyds TSB Group

PLC seems to have created more shareholder value than the others as its
EVA and return on capital invested is the highest. HSBC Holdings PLC
obtained a better ranking because it had the largest amount of capital
invested (i.e. USD 71,410 million).
The value relevance literature and shareholder value creation

From the 1990s, the relationship between stock market values (or changes
in value) and accounting figures has been investigated by a number of
studies and this branch of literature is called ‘value relevance literature’.23
This section reviews this branch of literature as follows: firstly, the value
relevance studies are classified according to their research goals; secondly,
a summary of those studies explicitly dealing with shareholder value
creation is given and finally, two methodological problems (namely, the
scale-effect and the adoption of R2 in association studies) in value relevance
studies are analysed.
An introduction to the value relevance literature

According to the research questions investigated by value relevance
studies, the literature can be classified into relative, incremental and
marginal association studies.
Relative association studies compare the association between stock market
values (change in values) and alternative accounting measures. The investi-
gation methods are usually very similar. The ‘value relevance’ or ‘relative
information content’ of alternative accounting measures refers to its ability
to explain the variation of a measure expressing shareholder value. The
value relevance is usually assessed by looking at difference in the ex-
planatory power (R2 or adjusted R2)24 of regressions where share prices
(or raw share returns)25 are the dependent variable and an account-
ing measure is the independent variable. The accounting measures with
the higher explanatory power are described as being the more value
relevant.
Incremental association studies assess the contribution provided by an
accounting measure in explaining market value or returns given other
specified variables. An accounting measure is usually considered ‘value rele-
vant’ if the regression coefficient is different from zero and statistically
significant. A smaller number of studies make additional assumptions
about the relation between accounting numbers and inputs to a market
valuation model: the purpose is to forecast coefficients and/or assess
the ability of different accounting figures of measuring a valuation input
variable. Measurement errors in accounting numbers are usually detected

looking at differences between the estimated and predicted values.
Marginal information content studies analyse whether investors’ avail-
able information is increased by the release of particular accounting infor-
mation. In other words, the literature assesses whether value changes are
associated with the release of accounting data (conditional on other infor-
mation released). Table 3.5 reports a summary of the most recent studies in
the value relevance literature according to their research goals.
A review of empirical studies in the value relevance literature

As seen in Table 3.5, the number of studies in the value relevance literature
is rather large. Among these studies, our focus is on research dealing with
shareholder value and, therefore, on the studies that both: (1) explicitly
analyse the value relevance of performance measures in the light of creat-
ing shareholder value; and (2) assess the value-relevance of modern per-
formance measurement methods (especially EVA, the most popular) over
traditional accounting measures.
Adopting these criteria the studies that investigate this issue focusing on
the banking industry are somewhat limited. In detail:
• O’Byrne (1996) analyses industrial companies and found EVA to be a

superior measure in explaining shareholder value compared to other
performance measures by using a two-step analysis. In the first step,
company equity market value was regressed on EVA and then on earn-
ings (namely, NOPAT). O’Byrne (1996) found an adjusted R2 for EVA of
0.31 and of 0.33 for the NOPAT. In the second step of the analysis, a set
of adjustments were proposed: firstly, regression coefficients were
allowed to vary for positive and negative values of EVA; secondly, the
natural log of capital was introduced as a predictor in order to take into
account differences in the market value of firms of different sizes;
thirdly, fifty-seven dummy variables were introduced to consider poten-
tial industry effects. In this second stage, O’Byrne (1996) found an R2 for
EVA of 0.56, which enables him to conclude that EVA was superior to
earnings in explaining firm value.
• Peterson and Peterson (1996) analysed traditional and innovative mea-
sures of performance and compared them with stock returns. According
to their findings, traditional measures are not empirically less related to
stock returns than EVA measures. They conclude that traditional meas-
ures should not be eliminated as a means for evaluating firm perform-
ance. They affirm, however, that EVA measures are worthwhile since
because value added measures focus on economic rather than account-
ing profit they play an important role in evaluating performance
because managers will aim towards value creation rather than mere
manipulation of short-sighted accounting figures.
Table 3.5: Value relevance literature classified by methodology and motivation
Methodology Standard-setting motivation
Marginal
Relative Incremental Measure- info. Inter-
No. Author(s) Journala Year association association ment content temporal Explicit Implicit
1 Aboody JAE 1996 1 1

2 Aboody and Lev JAR 1998 1 1
3 Ahmed and Takeda JAE 1995 1 1
4 Alford, Jones,
Leftwich and Zmijewski JAR 1993 1 1
5 Amir AR 1993 1 1
6 Amir AR 1996 1 1
7 Amir, Harris and Venuti JAR 1993 1 1 1
8 Amir, Kirschenheiler
and Willard CAR 1997 1 1 1
9 Amir and Lev JAE 1996 1 1 1
10 Anthony and Petroni JAAF 1997 1 1
11 Ayers AR 1998 1 1
12 Ballas WP 1997 1 1 1
13 Balsam and Lipka AH 1998 1 1 1 1
14 Bandyopadhyay,
Hanna and Richardson JAR 1994 1 1 1
15 Barth AR 1991 1 1 1
16 Barth AR 1994 1 1 1
17 Barth, Beaver and Landsman JAE 1992 1 1
85
86
Table 3.5: Value relevance literature classified by methodology and motivation – continued
Marginal
18 Barth, Beaver and Landsman AR 1996 1 1

19 Barth, Beaver and Stinson AR 1991 1 1
20 Barth, Clement, RAS 1998 1 1 1
Foster and Kasznik
21 Barth and Clinch CAR 1996 1 1
22 Barth and Clinch JAR 1998 1 1
23 Barth and McNichols JAR 1994 1 1
24 Bartov CAR 1997 1 1
25 Beaver, Christie and Griffin JAE 1980 1 1
26 Beaver and Dukes AR 1972 1 1
27 Biddle, Bowen and Wallace JAE 1997 1 1 1
28 Black JFSA 1998 1 1
29 Bodnar and Weintrop JAE 1997 1 1 1
30 Chan and Seow JAE 1996 1 1
31 Chaney and Jeter JAAF 1994 1 1
32 Cheng, Liu and Schaefer AH 1997 1 1
33 Choi, Collins and Johnson AR 1997 1 1 1
34 D’Souza, Jacob and Soderstrom JAE 2000 1 1 1
35 Davis-Friday and Rivera AH 2000 1 1
36 Dhaliwal, Subramanyam
and Trezevant JAE 1999 1 1
Marginal
37 Eccher, Ramesh
and Thiagarajan JAE 1996 1 1 1
38 Ely and Waymire WP 1999 1 1
39 Fields, Rangan
and Thiagarajan RAS 1998 1 1 1
40 Francis and Schipper JAR 1999 1 1
41 Gheyara and Boatsman JAE 1980 1 1
42 Givoly and Hayn AR 1992 1 1
43 Gopalakrishnan RQFA 1994 1 1
44 Gopalakrishnan and Sugrue JBFA 1993 1 1 1
45 Graham, Lefanowicz and Petroni WP 1998 1 1
46 Harris, Lang and Moller JAR 1994 1 1 1
47 Harris and Muller JAE 1999 1 1 1 1
48 Harris and Ohlson AR 1987 1 1 1
49 Henning and Stock WP 1997 1 1
50 Hirschey, Richardson and Scholz WP 1998 1 1
51 Joos and Lang JAR 1994 1 1 1
52 Lev and Sougiannis JAE 1996 1 1
53 Nelson AR 1996 1 1 1 1
54 Petroni and Wahlen JRI 1995 1 1 1
55 Pope and Rees JIFMA 1993 1 1 1
56 Rees and Elgers JAR 1997 1 1
87
88
Marginal
57 Rees and Stott WP 1999 1 1 1

58 Shevlin AR 1991 1 1 1
59 Venkatachalam JAE 1996 1 1 1
60 Vincent JFSA 1997 1 1 1
61 Vincent JAE 1999 1 1 1 1
62 Whisenant WP 1998 1 1 1
Total 15 53 13 7 7 54 8
a
Journal abbreviations: JAE = Journal of Accounting and Economics
ABR = Accounting and Business Research JAL = Journal of Accounting Literature
AER = American Economic Review JAR = Journal of Accounting Research
AFE = Applied Financial Economics JBF = Journal of Banking and Finance
AF = Accounting and Finance JBFA = Journal of Business Finance and Accounting
AH = Accounting Horizons JEB = Journal of Economics and Business
AQAFA = Advances in Quantitative Analysis of Finance and Accounting JREPM = Journal of Real Estate Portfolio Management
AR = Accounting Review JFSA = Journal of Financial Statement Analysis
BAF = Bank Accounting and Finance JIFMA = Journal of International Financial Management and Accounting
CAR = Contemporary Accounting Research JRI = Journal of Risk and Insurance
FAJ = Financial Analyst Journal MA = Management Accounting
FASB = Financial Accounting Standards Board RAS = Review of Accounting Studies
IJA = International Journal of Accounting RQFA = Review of Quantitative Finance and Accounting
JAAF = Journal of Accounting, Auditing and Finance WP = Working Paper (included only if publication not found)
Source: Holthausen and Watts (2001: 8–9).

• Uyemura et al. (1996) analysed the largest 100 US bank holding com-
panies over a period of ten years (1986–95). By regressing changes in
standardised Market Value Added (MVA) against changes in standard-
ised EVA (defined as EVA divided by capital) and traditional perfor-
mance measures, EVA was found to have the highest correlation with
MVA.
• Lehen and Makhija (1997) assess which performance measure does the
best job of predicting the turnover of Chief Executive Officers (CEOs).
Focusing on the degree of correlation between different performance
measures and stock market returns, they found that correlation co-
efficients vary from 0.39 and 0.76. In detail, EVA and MVA are the most
highly correlated measure with stock market returns: 0.59 and 0.58
(respectively). The other performance measures have smaller correla-
tions: 0.455 for ROA, 0.455 ROE and Return on Sales (ROS) 0.388. It
is interesting to note that, similar to other studies, the measure most
correlated with MVA is EVA.
• Biddle et al. (1997) analysed a sample of 6174 firm-years over the period
1984–93 to investigate the following research questions: (1) Does EVA
and/or Residual Income (RI) dominate net income (NI) and operating
cash flow (CFO) in explaining contemporaneous annual stock returns?
(2) Do components unique to EVA and/or RI help explain contempora-
neous stock returns beyond that explained by CFO and NI? (3) Does
EVA dominate earnings in explaining firm values? Regarding the first
point, Biddle et al. compared adjusted R2 obtained by regressing abnor-
mal returns against EVA, Residual Income (RI), accounting earnings
(namely, Earning Before Extraordinary Item – EBEI) and CFO. According
to their results, EBEI has the highest adjusted R2 and EVA has a smaller
adjusted R2: these results do not support the hypothesis that EVA domi-
nates traditional performance measure in its association with stock
market returns. Regarding the second point, Biddle et al. found that
the additional information provided from EVA components (to the
information already available in net income) is poor. In other words,
components unique to EVA (i.e. the capital charge and accounting
adjustments) are often not significant in explaining contemporaneous
returns. Regarding the third question, Biddle et al. assessed the relation-
ship between performance measures and firm value by replicating
O’Byrne’s (1996) study with some adjustments. In order to level the
playing field, Biddle et al. (1999) extended the adjustment proposed in
the second stage of O’Byrne’s (1996) analysis to the regressions run
against NOPAT and, in this case, EVA’s superiority disappears. In fact,
according to their results, accounting earnings have the highest adjusted
R2 (0.53), EVA has an adjusted R2 of 0.50 and NOPAT has an adjusted R2
of 0.49. These results suggest that EVA does not dominate accounting
earnings in explaining firm values.
• Al Ehrbar (1998) reports that several empirical analyses have been

carried out by Stern Stewart using the Performance 1000 database.
According to the Stern Stewart findings, EVA explains half of the volatil-
ity in companies’ MVA, the highest correlation found.
• Garvey and Milbourn (2000) assessed the ‘declared’ EVA superiority by
focusing on EVA-based management compensation schemes. This paper
adds to the literature a different dimension since Garvey and Milbourn
(2000) initially criticise the investigation techniques used previously (i.e.
the statistical correlation with stock returns and/or firm value). They
suggest that a strong statistical correlation with stock returns does not
establish (a priori) that a performance measure adds value to a compen-
sation system. In order to define the criteria for judging alternative per-
formance measures, Garvey and Milbourn (2000) proposed both a
theoretical and an empirical analysis. The theoretical analysis develops a
standard agency model with a principal and one agent. They concluded
that it is irrelevant to investigate whether EVA beats earnings per se,
while it would be more accurate to investigate under what circumstances
EVA beats earnings (and for what reasons). In the empirical investiga-
tion, Garvey and Milbourn test the model in Paul (1992) to verify the
theoretical model. In detail, they analysed the benefit of adopting EVA
as a compensation scheme. Garvey and Milbourn (2000: 241) found that
the ‘accounting measures continue to explain changes in compensation
even when stock returns are used as a explanatory variable. This is con-
sistent with the Paul (1992) model in that firms do not use exactly the
same weights as the stock market in determining compensation … More
surprisingly, we show that the apparently simplistic idea of comparing
the relative ability of alternative measures to explain stock returns is
both theoretically defensible and a reasonable representation of practice.
Therefore, firms contemplating the adoption of EVA would be well
advised to begin with an examination of EVA’s R2 with its stock returns.’
• Acheampong and Wetzstein (2001) propose an innovative type of
analysis using parametric methods for estimating productive efficiency,
focusing on the food industry. It is interesting to note that they
conclude that: ‘the analysis showed that there are no significant differ-
ences between traditional and value added measures of performance’
(2001: 7).
On the basis of the aforementioned studies, it is possible to conclude that

the evidence surrounding the value relevance of performance measures in
the light of creating shareholder value is mixed. By classifying these studies
by the affiliation of authors, it is possible to identify two groups: studies
carried out by consultants and those undertaken by academics. Regarding
the practitioner literature, these studies usually find that EVA is superior to
traditional performance measures as it is found to better explain stock
returns and firm values. Garvey and Milbourn (2000: 211), for example,
observe that ‘Stern Stewart, Boston Consulting Group, and LEK/Alcar make
the claim that their proprietary performance measure correlates more closely
with stock returns than do either traditional accounting measures or the mea-
sures of rival firms, allegedly making it a more desirable compensation tool.’
Regarding the academic literature, the superiority of EVA is usually not
verified as most studies generally find that traditional measures are not empir-
ically less related to stock returns than EVA and other value added measures.
Most of the studies illustrated propose a relative association investiga-
tion: an Ordinary-Least Square (OLS) regression model is usually employed
to test the relationship between performance indicators (i.e. independent
variables) and a market measure expressing the shareholder value created
(i.e. the dependent variable). A smaller number of studies (e.g. Biddle et al.,
1997) carry out an incremental association investigation. Although these
studies adopt quite similar investigation techniques, the variables adopted
(as independent, but especially as dependent variables) are heterogeneous.
Some studies (such as O’Byrne (1996), Peterson and Peterson (1996) and
Biddle et al. (1997 and 1999)) attempt to evaluate different performance
measures, including accounting earnings and residual income measures
such as EVA, by examining their degree of correlation with raw or adjusted
market rate of returns on the ground that the best measure is the most
highly correlated with stock returns. Some other studies (Al Ehrbar (1998)
and Uyemura et al. (1996)) compare financial measures looking at the
degree of correlation with MVA, considered by EVA promoters the ‘ulti-
mate measure of shareholder wealth creation’. Regarding this latter point,
the definition of the stock market figure taken as the dependent variable
deserves great attention in the value relevance literature since this choice
determines the concept of ‘value relevance’ empirically tested. The selec-
tion of the dependent variable is fundamental to interpreting empirical
evidence since this variable needs to capture the created shareholder
value (added) over a given time period. According to the dependent vari-
able selected, it changes the concept of the ‘relative information content’
investigated. In detail, by adopting as a dependent variable:
• Market value of the equity capital’ (MVE), the relative information

content of a performance measure refers to its capability to explain the
variation of ‘overall shareholders’ wealth’;
• MVA, the relative information content of a performance indicator refers
to its capability to explain the variation of ‘net shareholders’ wealth’ (i.e.
overall shareholders’ wealth net of the historical amount of capital
invested by shareholders in the company);
• Raw market returns, the relative information content of a performance
refers to its capability to explain ‘variation of the created shareholders’
wealth’ over the period considered;
• Market Adjusted Returns (MAR), the information content of a perfor-

mance measure refers to its capability to explain the variation of ‘share-
holder value (added)’ created over the period analysed.
Methodological issues in the value relevance literature

This section illustrates a methodological problem in value-relevance
studies, namely, the distortion of the explanatory power (R2 or adjusted R2)
of regressions due to ‘scale effects’. Most studies investigating relative- and
incremental-association between accounting and market values assess the
ability of independent variables (i.e. accounting measures) to explain varia-
tion of the dependent variable (i.e. stock market values or change in these
values) by looking at the regression R2, i.e. a measure of the explanatory
power of the independent variables in linear regression. As such, the
change of value-relevance of accounting disclosure over time (or across dis-
closure regimes) has been assessed looking at inter-temporal (or cross-
sample) differences in R2.
Although R2 is intuitively appealing for helping to identify preferred
value-relevance indicators, several studies have found that R2 may be biased
by ‘scale effects’, which refers to the undue influence of large firms in the
regression analysis. In other words, R2 may express ‘not only’ the ability of
accounting measures to explain variation in market values but also scale
factors.
In order to exemplify the scale effects on R2, consider a linear bivariate
relationship:
z i = α + β wi + ε I (3.51)
where zi is the dependent variable, wi is the independent variable and n is

the number of stocks. The scale effect on R2 is avoided when the initial
value of the dependent variable is the same across all observation before
the experiment and all the variation in the dependent variable is due to the
experimental period. By assuming that the dependent variable (zi) is given
by stock returns (i.e. stock price over lagged stock price) and the inde-
pendent variable (wi) is given by EPS deflated by lagged stock price, the
equation compares 1 invested in each of the n stocks. Under these circum-
stances, R2 measures the explanatory power of the independent variables in
linear regression.
Under different assumptions, the variation of the response variable
derives from both the experiment and the initial condition and, as such, R2
refers also to the variation of the initial condition. In many studies, data do
not have constant initial scale. For example, using firm level data to
analyse the relationship between market and accounting data (such as
market value of equity, share prices, book value of equity, net income, etc.)
implies that the initial values are affected by a scale factor, since values
reflect firms’ size.26 The effect of the scale factor on R2 can be assessed by
introducing into model 3.51 a random scaling factor si = (s1, …, sn) as
follows:
s iz i = α s i + β s iw i + s iε i (3.52)
As such, sizi represents the stock price at the end of the period and siwi is
the EPS during the period. Independently from a priori reasons to prefer
one model over the other from a valuation perspective, the interpretation
of R2 changes: in model 3.52, the R2 varies both in terms of the strength of
the relation between the two variables (zi and wi) and the variation in
si. After a theoretical demonstration, Brown et al. (1999) conclude that:
(1) R2 is upwardly biased in value relevance studies using market share
prices (i.e. the market value of equity per share) as the dependent variable
and (2) the bias is increasing in the scale factor’s coefficient of variation. In
other words, ceteris paribus, if scale effects are large (small), the scale factor
contributes more (less) variation to the observed variables relative to the
amount contributed by the underlying variable of interest and, as such, R2
will be high (low). As a consequence, cross-sectional or time series compar-
isons of R2 should be considered unsound without controlling for differ-
ences in the coefficient of variation of the scale factor across the sample. In
order to correct for these scale effects, Brown et al. (1999) suggest two alter-
native approaches. The first is to introduce into the model a proxy of the
coefficient of variation of the unobservable scale factor si. Since the scale
factor is the size of an observation, Brown et al. (1999) note that the ‘eco-
nomic resources per share’ represent an accurate scale factor in per share
analysis: as such, the coefficient of variation of share price and book value
per share are taken as proxies of the coefficient of variation of the scale
factor. This approach has a major drawback, however, as it does allow one
to make cross-sectional comparison of R2. The second approach is to deflate
each observed variable by a proxy for the unobservable scale proxy: this
approach solves the problem of the previous approach and, moreover, the
R2 s obtained better reflect the explanatory power of the underlying vari-
ables of interest and not that of scale. This approach seems to have found a
larger application and at least five different deflators have been employed
in cross-sectional valuation models as a proxy for scale: sales,27 number of
shares,28 opening market values29 and closing book values30 and, more
recently, market value of equity.31 There is, however, no agreement about
the best deflator for correcting for scale effects. Since the scale factor is the
size of an observation, Brown et al. (1999) remark that, under the assump-
tion of market efficiency, the share price is an accurate proxy of the scale
factor since it captures the value of the underlying economic resources to
which the share has a claim and, as such, Brown et al. (1999) use the price
of shares one year before the valuation date. Easton and Sommers (2003)
affirm that market capitalisation is not a proxy of scale, but it is the

‘true’ measure of scale: the deflation of observed variables in their model is
made by market capitalisation via a weighted least square regression (WLS).
Using the valuation model proposed by Easton and Sommers (2003) on
UK data,32 Akbar and Stark (2003) compare the results obtained using OLS
versus WLS and using different scale deflators. Their findings show that:
(1) when the number of shares is used as a deflator, the use of OLS or WLS
is not effective in removing scale effect; (2) the presence and the degree of
the scale effect is influenced by the model specification; and (3) in cross-
sectional valuation equations, total equity market value does not appear to
be superior to other deflators in removing scale and/or heteroscedasticity
effects.
Conclusion
This chapter has outlined the foundations of shareholder value theory
focusing on the concept of shareholder value and wealth, the evolution of
shareholder value theory, measurement performance methodologies and
has also reviewed evidence from the empirical literature.
To recap, a company creates value for the shareholders over a given time
period when the return on invested capital is greater than its opportunity
cost. In order to highlight that this concept of ‘value’ implies a comparison
of the shareholder return with the opportunity cost, the word ‘added’ joins
the term ‘value’. Next, shareholders’ wealth at a given point of time is
defined as the market value of the company’s equity capital and sharehold-
ers’ wealth changes over time (i.e. labelled as ‘created shareholder wealth’)
obtained by considering the increase in equity market value, the dividend
paid over the time period, the other payments made to shareholders during
the time period, the outlays for capital increases and the conversion of
company’s convertible bonds.
This chapter also discussed the main techniques used to measure
company performance and the creation of shareholder value. We presented
a detailed review of the methodologies and findings of empirical studies
dealing with shareholder value. First, we discussed publicly available esti-
mates of shareholder value focusing both at the industry and firm levels.
Secondly, the value-relevance literature was summarised focusing on
studies dealing with modern performance measure (especially, EVA) and
the creation of shareholder value. The following chapter leads on from this
discussion and outlines how banks formulate strategy to create shareholder
value.
4
How Banks Create Shareholder Value
Introduction
This chapter deals with a fundamental topic for shareholder value-oriented
banks and how they can create shareholder value. In order to answer this ques-
tion, we start from Market Value Added (MVA), which is calculated as the dif-
ference between the current market value of all capital elements and the
historical amount of capital invested in the company. In order to create share-
holder value, companies have therefore to increase what shareholders could
obtain selling their participation. Since stock market prices are purely expecta-
tional the company’s market value is given by the net present value of all
future profits discounted at the company’s cost of capital. By applying the
Discounted Cash Flow (DCF) method to the Free Cash Flow to Equity (FCFE),
the value of equity capital (which expresses the shareholders’ interest) is
obtained by discounting the ‘expected cash flows to the equity’ at the Cost of
Equity. Banks (as any firms) that strategically target shareholder value creation
focus on the following decisions:
• Increasing the expected cash flows to equity, i.e. the residual cash flows after
meeting all expenses, tax obligations, interest and principal payments. For
banks, this can be achieved in several ways such as by increasing sales (e.g.
increasing deposits, loans, off-balance sheet activities, inter-bank operations,
etc.), increasing prices (e.g. increasing commissions, interest rates, etc.),
increasing returns on risk-activities (such as trading on security, derivatives,
etc.) and reducing operating costs (such as operating and administrative
costs) and financial costs (such as interest expenses).
• Reducing the hurdle rate (i.e. the cost of equity or capital). In order to reduce
the bank’s cost of equity, managers can only attempt to reduce the system-
atic (or market) risk of the bank, i.e. measured by beta (β ). Since β expresses
the relative correlation between the bank’s share returns with the market
portfolio returns,1 managers can increase shareholder value by reducing this
correlation. For instance, managers can diversify corporate activities abroad
95
and, therefore, lower the covariance between the company’s share price and
the market portfolio. Since the risk free rate (i.e. the rate of return of a risk
free asset) and the market portfolio returns (i.e. the rate of return of a portfo-
lio composed of all activities in the stock market) cannot be influenced,
these are exogenous variables out of the managers’ control.
• Matching as closely as possible bank’s financing sources with bank’s investments.
If a bank is able to employ financial sources with similar features (e.g. in
terms of maturity, credit and/or interest rate risks) to the assets being
financed, this bank will increase shareholder value since this reduces cash
outflow (due to the cost necessary to manage the company’s liquidity) and
reduces its overall risk (which influences negatively both the cost of equity
and cost of debt).
These features are illustrated in Figure 4.1.

One might claim that many of these actions are strictly related to each other
and they create true shareholder value only if achieved keeping constant other
conditions. For instance, a bank can create shareholder value by increasing its
volume of deposits and loans, but keeping constant prices and not increasing
risk. For this reason, bank managers need to consider the effect produced in all
these dimensions (i.e. expected cash flow, hurdle rate and bank’s financial
structure) before launching any action aiming to create shareholder value. In
order to achieve at least one of these results, banks have implemented several
strategies over the last decade, such as mergers and acquisitions, the adoption
of improved risk management systems, incentive plans for managers, etc.
The next two sub-sections describe these strategies by distinguishing between
endogenous channels (i.e. strategies implemented within the bank) and exo-
genous channels (i.e. strategies that create shareholder value by involving
external parties).
SHV drivers Actions Bank's common strategies
Increase volume Bank's common strategies

Expected Cash Increase prices Endogenous
Flows to Equity • Technical and allocative
Increase risky activities strategies
efficiency
Reduce operating costs
Reducing financial costs • Right incentives
• Risk management
SHV Cost of Equity techniques
Reduce Cost of Equity Exogenous
• Mergers and acquisitions strategies
Financial structure
Financing mix
Figure 4.1: The creation of shareholder value: drivers, actions, strategies

How Banks Create Shareholder Value 97
Final goal Endogenous Endogenous drivers Endogenous strategies

goal
° Customer satisfaction ° New management

Improve the techniques for
relationship ° Bank's efficiency managing
SHV between banking risks
creation shareholders ° Optimise bank's
and other financial structure ° Appropriate system of
stakeholders incentives for human
° Optimise the mix of resources
business activities
° New organisation
solutions
Figure 4.2: The shareholder value endogenous strategies
Endogenous channels
In order to analyse how commercial banks can create value for their share-
holders, it is opportune to start looking at the strategies to be implemented
within the bank as shown in Figure 4.2. In a perfect market, all stakehold-
ers are dependent on each other for their success and managing to create
sustained shareholder value is not a zero-sum game – creating stable share-
holder value requires an intense focus on delivering benefits to customers
in the most efficient way, hiring and retaining a motivated workforce,
maintaining excellent supplier relationships, and being a good corporate
citizen in each of the local communities where the company has a pres-
ence. For this reason, most of the endogenous channels to create a stable
and sustainable shareholder value focus on the optimal management of the
bank stakeholders.
An essential strategy to create shareholder value is related to customer satis-
faction. Customers are fundamental in creating shareholder value since they
supply sale proceeds, which are necessary to remunerate shareholders and, also,
all other stakeholders. Another three important drivers for creating shareholder
value relate to: (1) improving bank efficiency, which requires an optimal man-
agement of the other stakeholders; (2) optimising bank’s financial structure;
and (3) defining the optimal mix of business activities.
Achieving customer satisfaction

Among all stakeholders, customers have a crucial importance in creating
shareholder value since customer satisfaction is a necessary condition for
achieving sound and steady profits. In order to analyse this shareholder
value driver, the following questions are assessed:
(a) How is customer satisfaction defined?

(b) What is the relationship between shareholder value and customer satis-
faction?
(c) How can a commercial bank determine the value of a customer?
(d) How can a bank improve customer satisfaction?
(e) What evidence is found in the empirical literature on customer satisfaction
in banking?
These areas are addressed below.
Definition of customer satisfaction

A customer is satisfied when his/her perceptions about the service received
meet his/her expectations. In addition, customer satisfaction can be defined
considering explicitly the price of the service.2 In this case, a customer is
satisfied when the perceived value of the service received meets the value
expected where the value perceived (and expected) is defined as the ratio
between the quality perceived (respectively, expected) and the cost perceived
(respectively, expected). Although these definitions differ in their formulation,
they express exactly the same concept. In our opinion, the first definition is
more general considering the ex-ante and ex-post feelings of the customer,
while the second also considers an important dimension (i.e. the cost) in
which customers have expectations and perceptions. The first definition does
not consider explicitly any dimensions of customer satisfaction, but proposes a
simple rule, which applies to all dimensions. In other words, customer satisfac-
tion is the result of the comparison between the customer’s expectations and
perceptions regarding different aspects of the relationship between the bank
and the customer. The bank’s customers have expectations about the cost of
the services, but also about the bank’s branches (e.g. the layout of the branch,
the length of the queue, etc.), the availability of off-line services (e.g. phone
banking, e-banking, etc.), the front-office employees (e.g. their competence,
their aspects, etc.), the technical features of the bank’s services, the type of rela-
tionship (e.g. loyalty) and the communication techniques (e.g. rapidity, effec-
tiveness, etc.). In detail, we draw upon three studies that define service quality
(labelled as dimensions of customer satisfaction) in the service industry:
• Parasuraman et al. (1985) identified ten dimensions of customer satisfaction

in the service industry: reliability, responsiveness, competence, access, cour-
tesy, communication, credibility, security, understanding/knowing the cus-
tomers and tangibles. Since the last seven dimensions were found to be
highly correlated, Parasuraman et al. (1988) reduced the dimensions of cus-
tomer satisfaction to five by grouping the last seven into two larger dimen-
sions. The five dimensions include tangibles (i.e. the physical evidence of the
service, such as physical facilities, appearance of personnel, etc.), reliability
(i.e. the consistency of performance and dependability, e.g. the company
performs the service right the first time and honours its promises), respon-
siveness (i.e. the willingness or readiness of employees to provide services),
assurance (i.e. the knowledge and courtesy of employees and their ability
to inspire trust and confidence) and empathy (i.e. the care and individual
attention that a company provides its customers).
• Johnston (1995) provides a list of factors for the service industry: access, aes-
thetics, care/attention, tidiness, comfort, commitment, communication,
competence, courtesy, flexibility, friendliness, functionality, integrity, relia-
bility, ability to answer customers’ needs and safety. In a successive study,
Johnston (1997) focuses on retail banking and finds that these dimensions,
independently from their relative importance, have a different impact on
customer satisfaction. In detail, some dimensions are usually dissatisfying
factors (i.e. hygienic factors), while others usually represent satisfying factors
(i.e. motivating factors). In other words, Johnston (1997) notes that:
(i) dissatisfying factors (usually all tangible aspects, such as integrity, aes-
thetic, safety, tidiness and reliability) generate customer dissatisfaction
if not properly managed, but do not create customer satisfaction if
properly managed;
(ii) satisfying factors (usually all intangible aspects, such as commitment,
friendliness, care/attention, ability to answer customers’ needs) generate
customer satisfaction if properly managed, but do not create customer
dissatisfaction if inappropriately managed.
• Stafforf (1996) analysed the banking industry and provided a list of
seven dimensions of customer satisfaction (i.e. branch atmosphere, rela-
tionship with customers, rates and costs of the services, availability and
convenience of services, ATMs, reliability and fairness and tellers) organ-
ised in several sub-items.
Customer satisfaction and shareholder value

The relationship between the creation of shareholder value and customer satis-
faction may be shown as in Figure 4.3. When a bank is able to satisfy its cus-
tomers and increase their confidence in the bank, the bank’s image and
reputation improves, customers have a more collaborative feeling, become
more loyal and the length of the relationship between the bank and the cus-
tomer increases.3 In contrast, dissatisfied customers tend to have less collabora-
tive behaviour and become less loyal.
For instance, research undertaken by the Technical Assistance Research
Program (TARP)4 in the US provides evidence of this problem in service indus-
tries. In detail, it has been found that: (a) 45 per cent of customers experiencing
a problem with a relative low cost service do not complain: 45 per cent of these
will never come back; (2) 37 per cent of customers experiencing a problem
with expensive services do not complain and 50 per cent of dissatisfied cus-
tomers will never come back. As a consequence, it is suggested5 that developing
and maintaining a loyal customer base is the easiest and most reliable source of
Customer satisfaction
Higher confidence Improving bank's image

in the bank and reputation
Collaborative
behaviour
Attracting new
customers
More information
on customers
Customer loyalty
Cross-selling
Improving
competitive position
Improving bank's performance
Creating shareholder value

(ceteris paribus)
Figure 4.3: Customer satisfaction
long-term financial performance. Under these circumstances, banks attract new

customers, sell new products to existing customers (increasing the number of
customers’ needs served by the bank), while also reducing the amortisation
period of fixed investments and advertising expenses. All these benefits make
customers more profitable and subsequently the bank’s competitive position
and performance increases. It is necessary to note that the steps described
above are interrelated and mutually enforcing. If correctly managed, these steps
can become a virtuous circle leading to better economic performance and,
ceteris paribus, to increased shareholder value for a bank.
It is interesting to note that while the relationship between customer satis-
faction and a bank’s performance is theoretically straightforward, the existing
literature has provided (in some cases) conflicting empirical evidence. For
example, Ittner and Larcker (1998: 32) found ‘modest support for claims that
customer satisfaction measures are leading indicators of customer purchase
behaviour (retention, revenue and revenue growth), growth in the number of
customers, and accounting performance (business-unit revenues, profit
margins and return on sales). We also find some evidence that firm-level cus-
tomer satisfaction measures can be economically relevant to the stock market
but are not completely reflected in contemporaneous accounting book values.
However, some of the tests suggest that customer behaviour and financial
results are relatively constant over a broad range of customer satisfaction,
changing only after satisfaction moves through various “threshold” values and
diminishing at high satisfaction levels.’
The presence of mixed results regarding customer satisfaction and share-
holder value creation is not surprising since there are several methodological
problems regarding the study of such issues. These relate to: (1) data collection:
for example, survey and questionnaire approaches appear the most appropriate
method for collecting data on customer satisfaction. However, there are con-
cerns about the incentives of those surveyed to accurately convey their opinion
and about the comparability of customers’ answers; (2) the choice of the metric
for measuring customer satisfaction; and (3) the functional form of the rela-
tionship between customer satisfaction and financial numbers.
Value of the customer

Regarding the value of a customer (or Customer Life Time Value – CLTV) this
can be measured by a commercial bank by discounting all future income gen-
erated by customers (as for any generic asset). As such, CLTV can be estimated
for a single customer (e.g. i) as:
ACMi
CLTVi = (4.1)
1 + wk – [(1 + g)(1 – 1i)]
where ACMs is the annual contribution margin (i.e. gross annual return) for
customer i, wk is the bank’s cost of capital invested, gs is the expected profitabil-
ity growth rate for customer i, ls is the percentage of customers at risk of leaving
the bank.
Since a common practice in commercial banking is to segment customers
into different groups with homogeneous features (labelled ‘segments’), a bank
can estimate the value of all its customers as the sum of the CTLV of all its cus-
tomer segments (j= 1,…, n):
CLTV = ∑
j=1
n
( AACMj
l + wk – [(l + gj)(l – lj)] )
Nj (4.2)
where AACMs is the average annual contribution margin (i.e. Average Gross
Annual Return) for customer segment j, wk is the bank’s cost of capital invested,
gs is the expected profitability growth rate for customer segment j, ls is the per-
centage of customers at risk of leaving the bank for customer segment j, and Ni
is the number of customer in segment j.
The relationship between customer satisfaction and profitability can also be
assessed looking at the potential economic losses associated with customer de-
sertions. The bank’s overall annual economic loss derived from poor customer
satisfaction (labelled ‘Value at risk of competition’) can be estimated as follows:
VARC = Σ i PCRLi × ACMi × Ni (4.3)

where VARC is the value at risk of competition, PCRLi is the percentage of cus-
tomers at risk of leaving the bank in the class i, ACMi is the average contribu-
tion margin of the customers in the class i, Ni is the number of customer in
the class i. The bank’s overall value at risk of competition (VARC) at a given
moment of time can be estimated by summing the potential economic loss
for all classes of customers. The value at risk of competition for a single class
of customers is obtained by multiplying the average profit margin of the cus-
tomers in the class, the number of customers in the class and the percentage
of customers at risk of leaving the bank in the class. For example,6 let us
suppose that (1) there are 1,200 customers in a specific segment, (2) the
average annual economic contribution is 2,000 euro and (3) the percentage of
customers at risk of leaving the bank in the class is 5.01 per cent. In this case,
the value at risk of competition in this class of customers is VARC = 5.01 per
cent × 1200 Euro × 2,000 = Euro 120,240.
These models of CLTV and VARC enable us to identify some possible strate-
gies for increasing bank’s shareholder value:
• a bank can create shareholder value, ceteris paribus, by reducing the percent-
age of customers at risk of leaving the bank in any of its customer segments
since this will increase CLTV and reduce VARC. A crucial role is played by
the estimation of the percentage of customers at risk of leaving the bank. To
estimate this percentage, one can use backward-looking techniques such as
the Customer Loss Rate (CLR) over a past period (e.g. the last period) and
this can be used as a proxy for the percentage of customers at risk of leaving
the bank over a future period. Since the customer loss rate is mostly deter-
mined in the long term by customer satisfaction, this proxy for the percent-
age at risk of leaving the bank indirectly can be considered as an indirect
measure of customer satisfaction. Other estimation approaches could use a
survey on a sample of customers aiming to measure their satisfaction and
their intention of leaving the bank in the future (e.g. within the coming
period). In this case, it is important to assess the relationship between cus-
tomer loyalty and customer satisfaction. In this case, it is possible to define
for each level of customer satisfaction (e.g. highly unsatisfied, unsatisfied, as
expected, satisfied, highly satisfied) the customer’s probability of leaving the
bank. As such, the percentage of customers at risk of leaving the bank can be
obtained by multiplying the customer’s probability of leaving the bank and
the percentage of customer having a given level of customer satisfaction. For
example, the value of 5.01 per cent used in the example above as the per-
centage of customers at risk of leaving the bank in the class was obtained as
shown in Table 4.1.
• a bank can create shareholder value, ceteris paribus, by increasing the
expected profitability growth rate for any of its customer segments. In this
regard, the length of the bank–customer relationship can play a fundamen-
tal role. By increasing customer loyalty (and therefore the length of the rela-
Table 4.1: The percentage of customers at risk of leaving the bank
Customer satisfaction Probability that Percentage of

the customer will Percentage of customers at risk
leave in the year customer of leaving
(%) (%) (%)
Highly unsatisfied 22 5 1.10

Unsatisfied 11 12 1.32
As expected 5 43 2.15
Satisfied 2 22 0.44
Highly satisfied
(enthusiastic) 0 18 0.00
Total 100 5.01
Source: Munari (2000: 219).
tionship), banks benefit from more customers with increasing revenue

generation. In addition, the average consistency of the annual income
generated by customers increases over time (see Figure 4.4) since customer
Figure 4.4: Customer satisfaction, intangible assets and shareholder value
Basic profit Cross-selling Premium price

Referrals Cost-savings Acquisition costs
100
80
60
Annual customer profit
40
20
0
1 2 3 4 5 6 7
–20
–40
–60
Years
Source: Munari (2000: 200), modified from Reichheld (1996).
satisfaction reduces price elasticity and increases margins on sales to

existing customers.
• a bank can create shareholder value, ceteris paribus, by increasing the
number of customers since this will increase CLTV. Like any kind of com-
pany, a commercial bank has several strategies for attracting new customers.
Firstly, this may be achieved by advertising the bank’s products by mass
marketing (such as advertising in mass media) and/or direct marketing tech-
niques. Despite the fact that these actions may be expensive, these tech-
niques may not be successful in attracting new customers in the short term
(since the relationship between a bank and a customer holds over a time
period and it is based on reciprocal reliance), while they may be helpful in
constructing a bank’s images/brand. A second possible driver for attracting
new customers relates to cross-customer communications. When a bank is
able to satisfy its existing customers, these customers increase their
confidence in the bank and communicate to other people their satisfaction.
As such, the bank’s image and reputation improve and unsatisfied cus-
tomers of competing banks will have the incentive to change their bank in
favour of more customer-oriented banks.
• changing the mix of customer segments. Every customer segment has a dif-
ferent shareholder value creation capacity since they have varying charac-
teristics. Each segment will have a different contribution margin, percentage
of customers at risk of leaving the bank, expected profitability growth rate
and number of customers. As such, banks can create shareholder value by
focusing their business activities on those segments having a better combi-
nation of these elements or attempting to improve some of these elements.
For example, a bank may attempt to: (1) reduce the probability of leaving
the bank for customer segments with a higher contribution margin or with
a higher number of customers; and (2) improve the expected profitability
growth rate or the average annual contribution margin in the customer seg-
ments having a lower probability of leaving the bank and with a larger
number of customers.
Improving customer satisfaction

A bank can achieve an improved customer satisfaction by enhancing cus-
tomers’ perceptions and/or reducing customers’ expectations. Although theo-
retically possible, this latter option may be difficult to achieve in practice since
customers’ expectations mostly are the result of exogenous factors, such as
direct experiences, customers’ needs, cross-customer communication (i.e. indi-
rect experience) and quality awards or certification of bank features by third
parties.
A bank can enhance customer satisfaction by improving its perception in all
dimensions of the relationship with its customers. One might claim that
improving the customer’s perception requires better products and services,
which would need new investment and expense. Under these circumstances, a
bank would benefit from higher revenues, but it also would sustain higher cost
making uncertain the creation of shareholder value. This criticism is straight-
forward when a bank focuses on final (usually labelled ‘external’) customers,
but it has been argued that it may be ineffective if customer satisfaction is
achieved as the final result of a Total Quality Management (TQM) programme.7
When TQM is applied, the bank intends to guarantee the satisfaction of its
external customers focusing on every process within the bank.
Within this framework, the production process becomes a customer–supplier
chain and all banks’ resources are directed towards satisfying their customers.
By correctly applying this approach, a bank achieves higher revenues (due to
Figure 4.5: A customer satisfaction model for retail banking
Cross-customers' Customers' direct Quality-certifications,

Customers' needs
communications experience -awards, etc.
Dimensions of the
quality of commercial
Customer's
banking services expectations
1. Bank branch
2. Bank off-line services. Customer
(i.e. phone, internet, etc.) Satisfaction
2. Front office employees
3. Bank's services/products
4. Cost of the service Customer's
5. Dual and fair relationship perceptions
6. Communications
Bank's image
Customer
Bank Intermediaries External

communications
Gap 4
Effective quality
of the services
Gap 3 Gap 1
Definition of the products
and services standards
Gap 2
Management's
perceptions of the
customers' expectations
Source: Fiordelisi (1997: 12).

the increased customer satisfaction and loyalty) and reduces costs (by eliminat-
ing resources that destroy customer satisfaction).
While this approach may be straightforward from a theoretical point of view,
it may be difficult to apply in practice. In order to create shareholder value by
Figure 4.6: Some causes of poor customer satisfaction in commercial banking
Poor market orientation
Insufficient and/or inadequate communication

towards the top
Too many hierarchical levels GAP 1

Inadequate customer segmentation
Poor focus on customer satisfaction measurement
Inaccurate product/service design
Insufficient commitment and involvement of the

management towards quality
A sensation of impossibility of achieving corporate

goals and productive standards GAP 2
Insufficient or unclear standardisation of duties
Insufficient or unclear productive standards
Inaccurate development and management of the point of

interaction with customers (e.g. branches, web site, etc.) CUSTOMER
SATISFACTION
Poor attitude of human resources towards
the duties assigned to them
Ambiguity in assigning duties among human resources
Conflicting duties
GAP 3
Inadequate system of supervision and control
A sensation of lack of control
Poor orientation towards team-work
Insufficient horizontal communication
Excessive promises in advertising
Poor adoption of direct marketing instruments

GAP 4
Inaccurate management of the bank’s image
Poor management of the pro-sumer
Wrong selection of ’s intermediaries

(e.g. IT providers)
Source: Fiordelisi (1997: 14).

improving customer satisfaction, the key point is to tailor the production

process to customers’ expectations along the entire dimensions of the
bank–customer relationship (see Figure 4.5). Following the original idea of
Parasuraman et al. (1985), the first step is to accurately understand customer
expectations. If a bank’s managers do not exactly comprehend customer expec-
tations, all the following business activities (e.g. the development of products
and services, etc.) will lead to poor customer satisfaction, even if these activities
are done well.
The difference between customer expectations and manager perceptions of
customer expectations we call GAP 1. If this gap is different to zero, banks will
not create shareholder value since external customers will be unsatisfied with
the bank services. Under these circumstances, customer satisfaction is still
achievable, but it becomes incompatible with the creation of shareholder value
since customer satisfaction will require corrective actions (which imply new
expenses and unplanned investments). Figure 4.6 shows some possible causes
of this gap.
Before proceeding with our analysis of how to improve customer satisfaction
it is necessary to note that the framework outlined in Figure 4.6 is dynamic
since customers continuously change and, consequently, their expectations
vary over time. The second step is to translate the managers’ perceptions of
customer expectations into appropriate productive standards, that will ensure
product and services compatible (in all dimensions of quality) with customer
expectations. In other words, bank managers should identify a set of concrete
productive standards, which should capture all customer expectations and will
lead the bank to deliver services coherent with customer expectations in all
dimensions of the service offered. If these productive standards are misaligned
with customer expectations, a bank’s products and services will be unable to
satisfy customers. The difference between manager perceptions of customer
expectations and product standards we can label GAP 2. If this gap is non-null,
banks will not create shareholder value since external customers will be
unsatisfied with the bank’s products and services. Similarly to GAP 1, customer
satisfaction may be an achievable goal even with the presence of a non-null
gap, but this would require some corrective (and costly) actions, which can
destroy shareholder value. Figure 4.6 shows some possible causes of this gap.
In the third stage, productive standards set in the previous step should be
correctly met. All human resources (top managers, but especially middle man-
agers and front- and back-office staff) in the bank should work to these produc-
tive standards, satisfying their internal customers and acting to realise products
and services exactly as projected in the previous step. The difference between
the products and services projected in the previous step and those materially
realised is labelled GAP 3. If this gap is non-null, banks will not create share-
holder value since external customers will be unsatisfied with the bank’s prod-
ucts and services realised, even if accurately projected. Similarly to the previous
gaps, customer satisfaction may be an achievable goal even in the presence of a
non-null gap, but this would require high investment to reorganise the
production process. Figure 4.6 shows some possible causes of this gap.
The fourth step recognises that customer perceptions may not reflect the
‘true’ features of the product and services delivered to the customers. In other
words, some factors may create a bias between the reality (i.e. the features of
product and services delivered) and the customer perceptions. In this case, even
if GAP 1, 2 and 3 are null, customers may be unsatisfied, perceiving that bank
services do not meet their expectations. The bias between the ‘true’ features of
the product and services (i.e. labelled ‘objective quality’) and customer percep-
tion (i.e. labelled ‘subjective quality’) is labelled GAP 4. In our opinion, there
are three main factors that create this perception. Firstly, the ‘image’ can affect
the reality: a brand name or the nationality of the bank may influence the per-
ception of customers. A second factor is the bank’s communication, which
influences both customer expectations and perceptions. For example, mislead-
ing or unfair communications may generate excessive customer expectations
(creating unsatisfied customers that otherwise would have been satisfied) or
decrease customer perceptions. The bank’s services providers (i.e. external com-
panies that deliver bank services) may be a third source of such bias. This
problem is critical for e-banking services because commercial banks usually
outsource technical IT procedures (e.g. the website project and management,
the local area network (LAN) connection, the kind of server, etc.) to specialised
companies (e.g. IT providers). In this case, the quality of these technical ser-
vices can affect the quality of the bank services: for example, if customers expe-
rience a low connection with the website or some problem with the server,
they feel unsatisfied towards the bank rather than towards the IT provider.
Although less critical, the same problem exists in off-line banking services: for
example, if the post is inefficient or untrustworthy (e.g. letters from the bank
are lost), customers will probably be unsatisfied towards the bank rather than
towards the postal services. In the presence of GAP 4, customer satisfaction
can be achieved only by changing intermediaries and the communication
style.
Overall, the framework outlined above shows that external customer satis-
faction is always an achievable goal and it is coherent (and indeed is enforced)
with the creation of shareholder value especially when the aforementioned
gaps are minimised.
Performance and customer satisfaction

Various studies have empirically analysed the relationship between customer
satisfaction and organisational performance. As noted in Zeithaml (2000), the
link between customer satisfaction and profits is neither straightforward nor
simple. A direct relationship between customer satisfaction and organisational
performance is found by several studies in different kinds of services: e.g. Koska
(1990) and Nelson et al. (1992) found that hospitals with improved customer
satisfaction have higher profitability; Aaker and Jacobson (1994) found better
stock returns linked to improved quality perceptions; Anderson et al. (1994)

found a significant association between customer satisfaction and accounting
return on assets; Ittner and Larcker (1996) found that shareholder value is
highly elastic with respect to customer satisfaction. A second area of investiga-
tion is the relationship between customer satisfaction and profitability.
Several studies (e.g. Buzzell and Gale (1987), Jacobson and Aaker (1987), Gale
(1992), Fornell (1992) and van der Wiele and Brown (2002)) provide evidence
that higher customer satisfaction translates into higher than normal market
share growth, the ability to charge a higher price, improved customer loyalty
with a strong link to improved profitability, and lower transaction costs.
Customer satisfaction is also found to be strongly correlated with repurchase
intentions, the willingness to recommend the company, and to improved
cross-buying (see Zeithaml et al. (1996); Anderson and Sullivan (1993); and
Verhoef et al. (1999)).
Improving banks’ efficiency and productivity

In addition to improving customer satisfaction, banks have also focused on
improving their productivity and efficiency in order to create value for
their shareholders. The issues of efficiency and productivity have become
particularly relevant for European commercial banks over the 1990s. Until
the 1980s, many supervisory authorities implemented structural regula-
tions in their national banking systems (in order to guarantee industry sta-
bility) by imposing entry barriers, restricted licensing, restrictions on asset
holdings and capital requirements, and so on. While these controls ensured
(to a certain extent) industry stability, these policies limited competition
between banks and probably reduced productive efficiency and sacrificed
the quality of banking services. In contrast, since the early 1990s, banking
regulation has shifted in Europe and many structural regulations have been
reduced or abolished, encouraging banks to improve their efficiency and
competitiveness.
As the European market for banking seems to have become more com-
petitive, banks have had to focus (among other things) on improving
their efficiency and productivity. This should result, ultimately, in improved
performance and shareholder value creation.
Defining bank productivity and efficiency

Productivity and efficiency have been frequently used as synonymous
although these terms do not express precisely the same concept. As defined by
Lovell (1993), the productivity of a production unit is expressed as the ratio of
its outputs to its inputs and it is determined by the production technology, the
efficiency and the environment. The efficiency of production (or productive
efficiency) is therefore only a determinant of productivity, which can be
defined as the comparison between observed and optimal values of a firm’s
inputs and outputs.
Focusing on productive efficiency, two components can be identified: pure

technical (or physical) efficiency and allocative (or price) efficiency. Koopmans
(1951: 60) provides a formal definition of technical efficiency: ‘a firm is techni-
cally efficient if an increase in any output requires a reduction in at least one
other output or an increase in at least one other input and if a reduction in any
input requires an increase in at least one input or a reduction in at least one
output.’ Thus, Technical Efficiency (TE) expresses the ability of a firm to obtain
maximal outputs from a given set of inputs or of minimising inputs for a given
target of outputs: this component focuses only on physical quantities and
technical relationships. If information on prices is available and a behavioural
assumption (such as profit maximisation or cost minimisation) can be appro-
priately made, Allocative Efficiencies (AE) can be introduced. AE in input selec-
tion refers to the selection of that mix of inputs which produces a given
quantity of outputs at the minimum cost. In formal terms, Price Efficiency8
refers to the ability of using the input in optimal proportions, given their
respective prices and production technology. Farrell (1957) introduced a
measure which combines technical and allocative efficiency, known as Overall
Efficiency (OE) which expresses the ability of a firm to choose its input and/or
output levels and mix them to optimise its economic goal. The overall
efficiency is also called ‘X-efficiency’ (or ‘Economic Efficiency’): Berger et al.
(1993b: 228) define the term X-efficiency as ‘all technical and allocative
efficiency of individual firms, as distinguished from scale and scope effi-
ciencies’. A more precise definition is proposed in Bauer et al. (1997: 1):
‘X-efficiency measures the deviations in performance from that of best-practice
firms on the efficient frontier, holding constant a number of exogenous market
factors such as the process faced in the local market. That is, the frontier
efficiency of an institution measures how well it performs relative to the pre-
dicted performance of the best firm in the industry if these best firms were
facing the same market conditions.’
Improvements in bank productivity and efficiency are believed to be inextri-
cably linked to the creation of shareholder value.
Efficiency and shareholder value

The relationship between company’s efficiency and shareholder value is
straightforward:
• by improving technical efficiency, a bank is able to obtain a higher level of

outputs from a given set of inputs or of reducing inputs for a given target of
outputs. In both cases, the bank will obtain a higher net income (and, ceteris
paribus, shareholder value) since it will benefit from higher operating
income or lower operating costs;
• by improving allocative efficiency, a bank is able to use inputs in optimal
proportions, given their respective prices and production technology. As
such, the bank will obtain a higher NOPAT (and, ceteris paribus, shareholder
value) since it will benefit from higher operating income or lower operating
costs.
According to the relationship between company’s efficiency (in all types of

configuration) and shareholder value, any improvement in efficiency will
create shareholder value. Although this is undeniable in the short term, it may
be false in managing to create a sustainable shareholder value over a long term.
If the efficiency/productivity improvements are obtained regardless of cus-
tomer satisfaction, the shareholder value created by the efficiency improve-
ment will be likely offset by value destroyed from losing customers. As such,
both efficiency and customer satisfaction appear to be necessary but
insufficient conditions for the creation of a sustained and sustainable share-
holder value and, only when these are achieved together, will this lead to the
creation of sustainable shareholder value.
Bank efficiency – empirical evidence

The goal of improving a bank’s efficiency in all its aspects can be achieved by
implementing various strategies. In detail, a commercial bank can improve its
efficiency by implementing new management techniques, introducing appro-
priate systems of incentives for the bank’s human resources and by adopting
new organisational solutions.
Bank practitioners usually assess bank efficiency and productivity focusing
on standard financial ratios, such as cost-income or cost-asset ratios. How-
ever, these efficiency measures have the limitation of omitting to consider a
bank’s business mix and, consequently, these provide rather crude efficiency
measures. To face this problem, various methodologies have been developed
by academics to measure firm efficiency by identifying a set of best-practice
firms (by estimating an efficient frontier) and measuring the efficiency of other
companies in comparison to the most efficient firms.
A variety of empirical approaches are currently available to estimate the rela-
tive efficiency of firms and this literature can be roughly divided into two
groups, according to the method chosen to estimate the frontier production
function. These two methodologies are the parametric and non-parametric
approaches: the first identifies a specific form for the production function,
while non-parametric approaches do not specify any functional form (except
for the linear interpolation among data points). Over the last thirty years,
various empirical studies have been applied to analyse the efficiency of bank-
ing sectors. Berger and Humphrey (1997) summarise over 120 studies (listed
in Table 4.2) dealing with cost and profit efficiency in banking. These studies
are divided between those using parametric techniques and non-parametric
techniques: sixty-nine studies applied non-parametric techniques and sixty
adopted parametric approaches. Overall, banks were found to have an average
efficiency of 79 per cent (median 83 per cent): this shows that banks have on
average the chance to improve their production (by reducing their inputs,
112
Table 4.2: Studies of the efficiency of banks up to 1997
Application Country Method Authors
Inform government policy:

Deregulation, financial disruption Norway DEA Berg, Forsund and Jansen (1992)
US DEA Elyasiani and Mehdian (1995)
Japan DEA Fukuyama (1995)
Spain TFA Lozano (1995a)
Turkey DEA Zaire (1995)
US TFA Humphrey and Pulley (1997)
Spain DEA Grifell-Tatje and Lovell (forthcoming)
Institution failure, risk, problem US TFA Berger and Humphrey (1992a)
loans and management US SFA Cebenoyan et al. (1993)
US DEA Barr, Seiford and Siems (1994)
US DEA Elyasiani, Mehdian and Rezvanian (1994)
US DEA Hermalin and Wallace (1994)
US SFA Berger and De Young (1996)
US SFA Mester (1996)
US TFA De Young (1997c)
Market structure and concentration Norway TFA Berg and Kim (1994)
US DFA Berger (1995)
US DEA Devaney and Weber (1995)
Norway TFA Berg and Kim (1996)
Spain SFA Maudos (1996b)
US DFA Berger and Hannan (1997)
Table 4.2: Studies of the efficiency of banks up to 1997 – continued
Mergers Norway DEA Berg (1992)

US DFA Berger and Humphrey (1992b)
US IN Fixier and Zieschang (1993)
US DFA Akhavein, Berger and Humphrey (1997)
US TFA De Young (1997b)
US DFA Peristiani (1997)
Address Research Issues:
Confidence intervals Italy DEA Ferrier and Hirschberg (1994)
US DEA Wheelock and Wilson (1994)
Comparing different efficiency US DEA, SFA Ferrier and Lovell (1990)
techniques or assumptions Greece DEA, SFA Giokas (1991)
US SFA, DFA, TFA Bauer, Berger and Humphrey (1993)
US DEA, SFA Eiseinbeis, Ferrier and Kwan (1996)
Spain SFA Maudos (1996a)
Germany SFA Altunbas and Molineux (n.d.)
US SFA Zhu, Ellinger and Shumway (forthcoming)
Comparing different output Norway DEA Berg, Forsund and Jensen (1991)
measures Finland DEA Kuussaari (1993)
Italy DEA Favero and Papi (1995)
US DFA Hunter and Timme (1995)
Finland DEA Kuussaari and Vesala (1995)
113
114
Organisational form, corporate US DEA Rangan, Grabowski, Aly and Pasurka (1988)
control issues US DEA Aly, Grabowski, Pasurka and Rangan (1990)
US DEA Elyasiani and Mehdian (1992)
US SFA Cebenoyan, Cooperman, Register and Hudigins (1993)
US SFA Chang, Hasan and Hunter (1993)
US DEA Grabowski, Rangan and Rezvanian (1993)
US DFA Newman and Shrieves (1993)
US SFA Pi and Timme (1993)
US DFA De Young and Nolle (1996)
US TFA Mahajan, Rangan and Zardkochi (1996)
India DEA Battacharya, Lovell and Sahay (1997)
US SFA, TFA Hasan and Hunter (forthcoming)
General level of efficiency US DEA Elyasiani and Mehdian (1990a)
UK DEA Field (1990)
UK DEA Drake and Weyman-Jones (1992)
Tunisia SFA Chaffai (1993)
Japan DEA Fukuyama (1993)
Switzerland DEA Sheldon and Haegler (1993)
Denmark DEA Bukh (1994)
US SFA Kaparakis, Miller and Noulas (1994)
Spain DEA Perz and Quesada (1994)
Germany TFA Lang and Weizel (1995)
Italy DEA, SFA Resti (1995)
Germany DFA Lang and Weizel (1996)
US DEA Miller and Noulas (1996)
Inter-country comparisons Norway, Sweden, DEA Berg, Forsund, Hjalmarsson and Suominen (1993)
Finland
11 OECD countries SFA Fecher and Pestieau (1993)
8 developed countries DEA Pastor, Perez and Quesada (1994)
Norway, Sweden, DEA Bukh, Berg and Forsund (1995)
Finland, Denmark
15 developed countries TFA Ruthenberg and Elias (1996)
Methodology issues US DEA Charnes, Cooper, Huang and Sun (1990)
US TFA Berger and Humphrey (1991)
US DFA Berger (1993)
Belgium FDH Tulkens (1993)
US DEA Ferrier,Kersterns and Vanden Eeckaut (1994)
Spain DEA Grifell-Tajte and Lovell (1994)
Norway, Sweden, MOS Bergendahl (1995)
Finland, Denmark
US DFA Adams,Berger and Sickles (1996)
US DFA Akhavein,Swamy and Taubman (1997)
Notes: DEA = Data Envelopment Analysis; SFA = Stochastic Frontier Approach; TFA = Thick Frontier Approach; DFA = Distribution Free Approach.
Source: Berger and Humphrey (1997: table 1).
115
given output levels, or increasing their outputs, given input levels) by

21 per cent.
Altunbas et al. (1999) analysed a wide sample of European banks and
found that technical change has systematically reduced European banks’
total costs during the 1990s. Another study by Altunbas et al. (2000a)
focused on the German banking sector by investigating the effect bank
ownership has on bank efficiency. Using the Fourier Flexible functional
form, stochastic frontier and distribution free analysis, cost and alternative
profit efficiencies were estimated for over 1,800 German banks of different
ownership type between 1989 and 1996. According to them, there is little
evidence that privately owned banks are more efficient than their publicly
owned counterparts. Indeed, the public and mutual banks appear to have
slight cost and profit advantages because of a possible advantage in terms
of lower cost of funds. The pooled cost efficiency level for all banks is
83.8 per cent, whilst profit efficiency is 78.9 per cent.
Dietsch and Lozano-Vives (2000) focus on French and Spanish banks by
assessing the effects of environmental conditions on commercial and savings
banks cost efficiencies. Cost frontiers for French and Spanish banks were ini-
tially estimated in each country based on individual national frontier between
1988 and 1992 by using the standard translog functional form and distribution
free approach. Their findings suggest that the cost efficiency estimates for the
two countries are similar (i.e. 88.1 per cent and 88.3 per cent for French and
Spanish banks, respectively). However, when a common frontier is estimated,
the cost efficiency levels for the two sets of banks fall to 58 per cent and 93 per
cent for French banks and Spanish banks, respectively. According to Dietsch
and Lozano-Vives (2000), these latter findings provide evidence that country-
specific variables are an important factor in explaining efficiency differences. In
addition, the authors argue that common frontiers, which do not include
country-specific effects, overestimate inefficiency. Hence, they propose the
inclusion of country-specific environmental, technological and regulatory con-
ditions in the construction of a common frontier. The inclusion of country-
specific environmental variables produces similar levels of cost efficiency as the
separate frontier for the French banks (88.8 per cent), whilst the new efficiency
level for Spanish banks is 74.8 per cent. As a result, Dietsch and Lozano-Vivas
provide evidence that the inclusion of environmental variables into a common
frontier can neutralise efficiency differences between countries.
Carbo et al. (2000b) assess variable cost efficiency for a sample of European
savings banks between 1989 and 1996 using the Fourier Flexible functional
form and stochastic frontier. Mean X-efficiency is found to be around 78 per
cent improving steadily from 1991 to 1996. The most cost efficient savings
banks are found in Sweden, Austria, Germany and Denmark with the least
efficient in Finland and Portugal. Smaller savings banks were found to have a
slight size advantage, being relatively more efficient than their larger counter-
parts: in fact, larger savings banks benefit more from technical progress and
economies of scale than smaller savings banks. Economies of scale were found
to be in the region of 7–8 per cent.
Altunbas et al. (2000b) focused on Japanese banking over the period 1993–6
by analysing the impact of risk and quality factors on banks’ cost. Using
the Fourier Flexible form and the stochastic frontier approach, they found
X-inefficiency around 5–7 per cent and observed little impact of risk and asset
quality on efficiency estimates.
Battese et al. (2000) focus on Swedish banking in order to analyse the
efficiency of labour use. They employ the technical efficiency effects model
developed by Battese and Coelli (1995): this methodology enables inefficiencies
to be modelled in terms of firm-specific variables and time, and in a manner
that does not invoke violations of the assumptions of regression analysis. Using
the translog functional form and a sample of 156 banks between 1984 and
1995, Battese et al. (2000) find a mean operating cost efficiency of 88.3 per
cent. The inefficiency effects are modelled considering ownership type, the size
of a bank’s branch network, total inventories and time. According to their
findings, cooperative banks, savings banks and large commercial Handelsbanken
are more efficient in their labour usage than ‘other commercial’ banks, whilst
the large SE Banken is not.
Berger (2000) evaluates the efficiency effects of the integration of the
financial services industry over the period 1992–7 for US and European banks.
Both cost and profit efficiencies are estimated and cost X-efficiency was found
to be in the order of 80 per cent on average and profit X-efficiency was typi-
cally found to be around 50 per cent.
Glass and McKillop (2000) analyse the UK building society sector by
using linear programming techniques and Malmquist productivity indices to
estimate total productivity change. According to them, UK building societies
experienced a substantial productivity growth between 1989 and 1993 mainly
due to progressive shifts in technology. Small efficiency increases were also
generated from improvements in scale efficiency.
Berger and De Young (2001) investigate the effects of geographical expansion
on bank cost and alternative profit efficiency focusing on US commercial banks
between 1993 and 1998. They use the DFA methodology and also specify cost
and profit functions using the Fourier Flexible functional form. Berger and De
Young (2001) find mixed evidence of the relationship between geographic
scope and banks’ efficiency. For example, banks in organisations expanding in
nearby regions experience efficiency gains, while the efficiency of bank
affiliates tends to reduce as they move further away from the parents, especially
for small bank affiliates.
Altunbas et al. (2001) analyse cost efficiency for a large sample of
European banks between 1989 and 1997 using the stochastic frontier
approach and the Fourier Flexible functional form. According to them,
mean X-efficiency increased over time from 75.5 per cent in 1989 to
82.1 per cent in 1997. Furthermore, X-efficiency varies both across
countries and bank asset size classes. Very small banks and banks with
assets above Euro 500 million have X-efficiencies less than 80 per cent;
indeed, inefficiency increases with bank size for institutions with more
than Euro 500 million worth of assets. On an individual country basis,
Altunbas et al. (2001) find the lowest mean X-inefficiency in the Italian and
German banking markets (12.6 per cent and 13.5 per cent, respectively),
whereas in Belgium, Ireland and Luxembourg bank cost X-inefficiency is
over 30 per cent.
Chaffai et al. (2001) analyse the productivity changes in France, Germany,
Italy and Spain over the period 1993–7 using a parametric methodology and a
Malmquist type productivity index. The authors found that Spanish banks
have lower productivity than Italian, French and German banks. However, the
authors note that country frontier differences may also be due to environ-
mental factors and not entirely to differences in banking technologies. By
introducing various environmental factors, the authors found that the pro-
ductivity gap between Spanish banks and other banks become larger, while it
reduces the gap between Italian and French banks.
Berger and Mester (2001) assess productivity change in the US banking
industry by estimating both cost and profit efficiency. Concerning cost
efficiency, they found that in the period 1984–91, costs for the average bank
fell at an annual rate of 0.3 per cent, while in the period 1991–7 costs rose at an
annual rate of 2.7 per cent: the overall effect over the period 1984–97 is that
costs for the average bank rose at an annual rate of 1.1 per cent. Assessing the
three components of the cost changes, Berger and Mester (2001) found that
cost productivity worsened over both sub-intervals, while business conditions
as a whole reduced costs over both sub-intervals.
Sathye (2001) uses DEA in order to investigate technical and allocative
efficiency in the Australian banking sector. Substantial cost saving oppor-
tunities are found and the mean cost efficiency scores are around 58 per cent.
As a source of inefficiency, the technical component is more important
than the allocative component: the mean values are 67 per cent and 85 per
cent, respectively. This study is interesting for two reasons. Firstly, empirical
studies on efficiency in the Australian banking market are rare (e.g. Berger and
Humphrey, 1997 do not mention any study). Secondly, Sathye (2001) under-
takes a regression analysis to assess whether the technical, allocative and
cost efficiency are related to banks’ size (measured by total assets), market
power (measured by log deposits), ownership (expressed with a dummy
variable: 0 = domestic bank; 1 = foreign bank), use of technology (proxy by the
number of bank-owned ATMs) and cost per employee. Sathye found that
banks’ market power is statistically significant at least at the 5 per cent level
and influences negatively cost, technical and allocative efficiency of Australian
banks.
Alarm (2001) use the Malmquist productity index to assess productivity
changes in the US banking system over the 1980s. The study finds, on average,
a substantial TFP increase between 1983 and 1984, a TFP fall in 1985 and, since
then, an increasing trend.
Glass and McKillop (2002) analyse cost and scale efficiency of UK credit
unions in 1996 using Data Envelopment Analysis. The authors measured tech-
nical efficiency using both Farrell (1957) radial measures and Fare and Lovell
(1978) non-radial measures. Glass and McKillop (2002) found that UK credit
unions have substantial cost inefficiencies.
Beccalli et al. (2003) investigated the relationship between stock market
returns and bank efficiency focusing on five European countries (France,
Germany, Italy, Spain and United Kingdom). After estimating bank cost
efficiency using DEA and SFA (using a standard translog cost function), the
study assesses the relationship between stock market returns and efficiency
change performing a standard value-relevance study. Beccalli et al. find that
DEA efficiency estimates have a higher ability in explaining variation in market
returns than compared to SFA efficiency measures.
Casu and Girardone (2004) assess productivity, cost and profit efficiency of
financial conglomerates focusing on Italy between 1996 and 1999. They use
two parametric estimation methodologies for estimating cost and alternative
profit efficiencies (namely, SFA and DFA using a Fourier Flexible cost function)
and a non-parametric technique (DEA) for estimating technical, allocative and
scale efficiency. Casu and Girardone found that Italian banking groups did not
experience a clear improvement in cost efficiency and productivity between
1996 and 1999. However, Italian financial conglomerates were found to have
consistently improved their profit efficiency (confirmed also by the increase in
their ROA) over the same period that (they argue) may have been the result of
strategic choices carried out by Italian banking groups. The authors also note
that the trend towards conglomeration did not translate into scale efficiency
gains, while their scope economies results seem to give a positive indication of
the benefits of diversification.
Beccalli (2004) focused on Italian and UK investment firms over the period
1995–8 by comparing cost X-efficiency. Cost efficiency is measured using four
different SFA specifications using a translog cost function and applying a time-
varying model for technical inefficiency effects following Battese and Coelli
(1992). Beccalli (2004) firstly estimated two separate domestic frontiers and
found that investment firms have higher cost efficiency in Italy compared to
the UK. In a second step, Beccalli estimated a common frontier (integrating
structural factors with interpretational and institutional variables) and found
that UK investment firms have higher efficiency than Italian firms.
Carbo and Humphrey (2004) analyse the source of cost inefficiency using
parametric and non-parametric estimation methodologies (Distribution Free
Approach, using both a translog and a Fourier Flexible cost function, and Data
Envelopment Analysis, respectively) for a sample of Spanish banks between
1992 and 2001. Carbo and Humphrey (2004) found that Spanish bank
efficiency is related to service delivery methods and banking productivity, such
as the intensity of labour usage per branch office, the use of ATMs versus
branch offices to deliver certain banking services, the number of employees per
branch office and the level of deposits raised per office. By considering these
issues, the authors found unexplained inefficiency levels of 1–4 per cent that
are substantially lower than the levels (20–25 per cent) commonly reported in
the previous literature (e.g. Berger and Humphrey, 1997).
Finally, Casu et al. (2004) analyse productivity change in five European
countries (France, Germany, Italy, Spain, United Kingdom) by estimating the
Malmquist Total Factor Productivity (TFP) index over the period 1993–7. The
authors use both a parametric methodology (by decomposing cost changes,
estimated using a standard translog functional form, following Berger and
Mester (1999 and 2001)) and a non-parametric approach (DEA). Casu et al.
(2004) provide evidence that Italian and Spanish banks experienced productiv-
ity improvements, while French, German and UK banks did not. The authors
found that the two methodologies applied do not give markedly conflicting
results.
While this section provides a brief review of the empirical literature, it can be
seen that an extensive number of studies have sought to examine efficiency of
various banking systems, including various European countries. Most tend to
focus on the level of cost and profit efficiency, with less attention paid to
examining trends or the determinants of this efficiency. What is clear is that
for the European studies there is certainly no consensus that bank efficiency
improved during the 1990s, although there is some evidence supporting the
view that productivity (Malmquist TFP change) has improved. In any event,
systematic improvement in bank efficiency should feed through into system-
atic improvements in shareholder value creation, all other things being equal.
These issues will be considered in more detail later in this text.
Optimise banks’ financial structure

In addition to improving customer satisfaction and boosting efficiency, banks
can also optimise their financial structure in order to increase shareholder
value. According to a basic principle of corporate finance, a company’s
financial structure can be considered optimal when it ensures the minimum
cost of capital, considering both debt and equity capital. As such, the cost of
capital is usually measured as the Weighted Average Cost of Capital (WACC),
which combines both the cost of debt and the cost of equity according to their
relevance in financing the company’s assets.
D E
WACC = kd + ke (4.4)
K K
where WACC is Weighted Average Cost of Capital, Wd is the cost of debt, We is

the cost of common equity, D is the book value of company’s debts and E is
the book value of company’s equity.
WACC minimisation should be the goal of any company in defining its

optimal capital structure. Independently from the metrics adopted for measur-
ing the creation of shareholder value, a reduction in WACC is likely to lead to
an increase in shareholder value. Damodaran (1999a) identifies four possible
ways to reduce the WACC for any company:
(1) Reducing operating risk. Since both cost of equity and cost of debt are posi-
tively correlated to a company’s risks, WACC can be reduced by reducing
the risk of the business or businesses in which it operates.
(2) Reducing operating leverage. A company’s operating leverage expresses the
proportion of fixed costs among a firm’s total expenses. Since the com-
pany’s operating leverage is positively related to its risk, the volatility of a
company’s profit is positively related to its operating leverage. As a result,
managers can reduce the WACC by reducing the proportion of fixed costs
and, therefore, making the business less risky. This can be achieved by
outsourcing some non-core activities and improving cost efficiency.
(3) Changing the financing mix. In order to reduce WACC, a company can
modify the mix of debt and equity used.
(4) Changing the financing type. In choosing the financing type, bank managers
should ensure that cash flows on the debt match as closely as possible
those on assets since this reduces the liquidity risk and the probability of
default: as a result, WACC decreases and shareholder value increases.
Focusing on the selection of the optimal financial mix (or financial struc-
D
ture), it is possible to identify the debt ratio (i.e. ) that minimises a com-
K
pany’s WACC. Let us define the cost of equity using the CAPM (as shown
in 4.5).
ke = Rfr + β lev[Rm – Rfr] (4.5)
The cost of common equity (ke) of a bank is estimated as the sum a risk free
rate of return (Rfr) and a risk premium, that is calculated as a proportion (β lev) of
the market premium (i.e. the difference between the market rate of return (Rm)
and the risk free rate of returns. The coefficient β lev is labelled as leveraged beta,
i.e. the beta obtained for a company with external financing debts. This is esti-

 D 
mated as β lev =  β unlev(1 – (1 – tx) E  , where tx is the mean company’s tax
 
rate.9
The mean cost of company’s debts (Ed) is estimated as a function of the debt
ratio, the interest rate, the relationship between cost of debt and debt ratio and
the company’s tax rate, as described in equation 4.6.
D
kd = (i + γ ) (1 – tx) (4.6)
K
where i is non-leveraged interest rate for the company (i.e. the interest rate for
company with no debts), γ is the intensity of the reaction of cost of debt to
changes of the debt ratio (geometrically, γ expresses the inclination of the rela-
tionship linking cost of debt and the debt ratio) and tx is the mean company’s
tax rate. Once the company’s cost of equity and cost of debts are defined,
company’s WACC can be restated as:
D D
WACC = (i + γ )(1 – tx) + (4.7)
K K

D  E
Rfe +  β unlev (1 + (1 – tx)  [Rm – Rfr]
 E  K
Let us define: x = D ; 1 – x = E and D = x and substitute in model 4.7.

K K E 1–x
With the following arithmetic steps:

x 
WACC = {(i + γ x)(1 – tx)}x + Rfr +  β unlev (1 + (1 – tx) [R – Rfr] (1 – x)
 1 – x  m
1 – x + x – xtx
= (ix + γ x2)(1 – tx) + Rfr + β unlev [Rm – Rfr] (1 – x)
1–x
Rfr + xRfr + β unlev [Rm – Rfr] – β unlev [Rm – Rfr]xtx

= (ix – ixtx + γ (1 – tx)x2 + (1 – x)
1–x
= {γ (1 – tx)}x2 + {(i(1 – tx) – Rfr – β unlev [Rm – Rfr]tx}x + {Rfr + β unlev [Rm – Rfr]}
WACC can be modelled as a quadratic parabola (in its generic form y = ax2 +
bx + c ) where the independent variable is the debt-ratio:
WACC = {γ (1 – tx)} K
D
( ) 2
+ {(i(1 – tx) – Rfr – β unlev [Rm – Rfr]tx} (4.8)
D
K ( )
+ {Rfr + β unlev [Rm – Rfr]}
Since the term γ (1 – tx) is positive, the parabola slopes upwards and, conse-
quently, the minimum company’s WACC is obtained in the parabola’s vertice.
As such, it is possible to identify the debt ratio minimising, ceteris paribus, the
b
company’s WACC by determining the parabola vertice (i.e. – in the generic
2a
10
form of a parabola) as follows:
( )
D
K
R +β
= fr unlev [Rm – Rfr]tx – (i(1 – tx)
2{γ (1 – tx)}
(4.9)
This level of debt ratio expresses the company’s optimal financial structure
since it minimises, ceteris paribus, the cost of capital. This level depends mainly
on exogenous factors such as:
• (i) the non-leveraged interest rate for the company. As the non-leveraged
interest rate increases, financing company’s assets with debts becomes more
expensive and, as such, the optimal debt-ratio decreases.
• (β unlev) the unleveraged beta. Since unleveraged beta expresses the systematic
risk of the company, if this coefficient increases, the company is judged
more risky by investors and the cost of equity required to invest in equity,
ceteris paribus, increases becoming more expensive. As a consequence, the
company finds it more convenient to finance company’s investments with
debts and, as such, the optimal debt-ratio increases.
• (Rm – Rfr) the market risk premium. When the difference between market
return and the risk free rate of return increases, shareholders expect to
receive, ceteris paribus, a higher return on their investment. In other words,
the cost of equity increases and, for the same process described in the previ-
ous point, the optimal debt-ratio increases.
• (Rfr) the risk-free rate of return. When the risk-free rate of return increases,
shareholders expect to receive, ceteris paribus, a higher return on their risky
investment. In other words, the cost of equity increases and, for the same
process described above, the optimal debt-ratio increases.
• (tx) the tax rate. As the tax rate increases, the tax shield effects become
stronger and the optimal debt-ratio increases (since it becomes more conve-
nient to use debts for financing company’s activities).
• (γ ) the intensity of the reaction of cost of debt to changes in the debt
ratio. As the intensity of the reaction of cost of debt to changes in the
debt ratio increases, the cost of debt increases and the optimal debt-ratio
decreases.
Studies on optimal financial structure have gained considerable attention in

banking in recent years especially due to the growing number of operations of
equity capital increases and, especially, of equity buy-backs. For example,
almost all of the largest twenty-five US banks implemented equity buy-back
programmes in the late 1990s11 and some important European banks have
done the same (e.g. Barclays for GBP 1.5 billion, Morgan Stanley for USD
600 million and Nordbanken for more than Euro 500 million).12
While the above presents points valid for all kinds of companies, banks have
some peculiar features that complicate the task of establishing appropriate
levels of capital. The fundamental difference between the capital structure of
banks and non-financial companies is that banks have a higher leverage (i.e.
D/E) since these are financial intermediaries. Banking supervisory authorities
regulate equity capital in order to control bank risks and, since the first 1988
Basle Capital Accord, a minimum level of equity capital is required. As such,
the selection of the optimal financial structure is not completely free for banks,
but it is necessary to consider exogenous constraints. In this regard, Davis and
Lee (1997: 33) observe that ‘the process of developing an optimal capital struc-
ture for banks has three dimensions – three aspects or set of considerations that
we will refer to as economic risk, regulatory and practical’. Focusing
on this approach, banks should quantify their risk exposure and define the
level of Capital At Risk (CAR), which is the amount of economic capital
required as a cushion against those risks. This amount of capital represents a
lower bound on the optimal capital target. Next, banks need to compare CAR
with the regulatory capital requirement. It is interesting to note that banking
regulation (namely, the New Basle Capital Accord) has moved towards the
adoption of bank internal methods in order to determine capital requirements.
Davis and Lee (1997) note that other factors should also be considered such
as: (1) the situation of comparable banks (estimable by a peer group analysis):
this provides a sort of ‘reality check’; (2) the bank’s future prospects and needs:
in this regard, it is necessary to consider the internal capital generation capabil-
ity, the potential investment opportunities and the bank’s desire to achieve a
specific debt rating. Once the overall amount of capital is defined, the last step
is to determine its composition. In other words, it is necessary to assess the dif-
ferent forms of bank capital: common stock, preferred stock, loss reserves, con-
vertible and subordinated debt, and so on.
Defining the optimal mix of business activities

Banks can also boost shareholder value by moving towards an optimal mix of
their business activities. Two main approaches can be undertaken by bank
managers to create shareholder value focusing on the services offered to cus-
tomers: they can diversify the bank’s services and products or they can focus
on their core business. Regarding the first approach, bank managers may
attempt to create shareholder value by diversifying bank’s supply. As noted by
Resti (1999), diversification can be viewed as the development, within the
bank, of a ‘second’ bank which is able to offer new services and products.
Looking at the evolution of commercial banking over the last decade, it is
possible to identify some examples of diversification.
A first example concerns the development of asset management services by
many (especially continental) European banks over the 1990s. Up to the 1980s,
the traditional source of profit for commercial banks in Europe was the interest
margin, a result of deposit-loan financial intermediation. In other words, com-
mercial banks have traditionally achieved profits by applying to loans a higher
interest rate than those applied to deposits (saving- and current-account).
Another common form of investment for many private investors were govern-
ment and corporate bonds and, in this case, commercial banks earned a com-
mission for the securities custody service provided. Commercial banks also
earned a commission for brokering services. Up to the beginning of the 1990s,
commercial banks’ role in asset management services, pensions and other fee-
based services was mainly passive. Banks typically did not emphasise the
development of non-interest base of business: for example, in continental
Europe, mutual funds were initially offered to private investors by non-
commercial bank financial institutions. Commercial banks’ attitude to
introducing mutual funds was limited since they were worried about a pos-
sible cannibalisation of their traditional products (which were more profit-
able). However, during the 1990s all this has changed and there has been a
substantial growth in fee income. During the 1990s, interest margins have
fallen and banks have emphasised fee and commission income as a major
source of income. For example, ECB (2004: 10) notes that ‘Net interest
income continued to decline in 2003 for most of the 50 large banks in the
sample. Margins narrowed because of relatively low nominal interest rates
Figure 4.7: Non-interest income of European banks during the 1990s
45 45
40 40
35 35
30 30
25 25
20 20
1989 1990 1991 1992 1993 1994 1995 1995 1996 1997 1998
EURO weighted EU weighted EURO weighted EU weighted
Source: ECB (2000: 15).
Non-interest income expressed as a percentage of operating income

1980 1985 1990 1995 1996 1997 1998 %change %change %change
1996/95 1997/96 1998/97
LU 18 20 35 34 38 43 55 11 14 27
SE 46 26 30 35 41 53 16 17 29
FR 34 38 45 52 11 19 16
AT 42 39 41 43 47 5 5 10
IT 31 35 39 46 15 11 18
GR 45 47 45 45 4 –5 1
BE 29 32 34 37 43 7 9 15
FI 35 43 40 42 22 –5 4
PT 59 57 20 27 35 35 41 26 1 17
NL 26 26 29 33 37 40 40 9 8 2
IE 26 17 27 31 36 37 40 15 5 7
UK 39 43 39 39 40 –8 –1 2
DK 28 31 30 37 8 –1 23
ES 13 14 17 27 31 33 36 13 8 8
DE 25 25 29 33 3 13 14
EURO weighted 30 32 36 41 10 12 14
EU weighted 32 34 37 41 6 9 12
Source: National central banks and supervisory authorities represented in the Banking Supervision
Committee (see ECB 2000: 50).
in most of EU countries coupled with increased competition in some market

segments … Interim financial statements also indicate that net non-interest
income increased in the first half of 2004. This was largely due to increasing
fee and commission income’; and ECB (2004: 14) noted that ‘in 2003, the
share of non-interest income in total operating income increased slightly to
42 per cent, 1 percentage higher than in 2002’. Figure 4.7 reports the trend
of non-interest income over the 1990s.
Another example of diversification relates to the development of internet-
based banking services. In this case, the diversification is not related to the
development of new services, but by innovation in the distribution channel of
traditional banking services. Up to the mid-1990s, commercial banks delivered
their services in physical places (i.e. branches). With the development of the
internet and its diffusion among banking customers, there has been substantial
potential for developing new business areas within traditional banking activi-
ties, such as e-banking business. This solution is appealing for a specific target
of banks’ customers; however, the way of doing e-banking differs substantially
from traditional banking management practices. Although the services offered
are ‘technically’ the same, banking and e-banking are two different (but con-
tiguous) business areas. Recent research13 has found that most banks offer
e-banking services by applying strategies usually implemented in off-line
services, while e-banking services should require new approaches and strategies
tailored to the specific features of the internet.
In many cases, banks’ managers have explained the diversification of
banking activities with the goal of diversifying corporate risk, reducing risks
Current business services

SHV
added
SHV
destroyed
Closeness to core
Core business strategy competencies Diversification strategy
SHV SHV
added added
SHV SHV
destroyed destroyed
Closeness to core Closeness to core
competencies competencies
Figure 4.8: Bank’s supply: diversification vs. core business strategies

and obtaining an optimal risk-return combination. Although this aim is cer-

tainly considerable, one may claim that investors could cheaply achieve risk
diversification by privately diversifying their own portfolio (labelled as lateral
diversification). Starting from this observation, in our opinion, there are two
reasons which can justify the diversification of banking activities, namely, cost
and revenue synergies. Revenue synergies are achieved by cross-selling financial
products (e.g. mutual/pension funds, retail banking services, payment services,
insurances, etc.). Cost saving synergies are explained in terms of cost
X-efficiency gains and/or economies of scale and/or scope. According to this
approach, value is created by analysing all activities in terms of shareholder
value creation (and, secondarily, of the closeness to a bank’s core competen-
cies) and business mix can be chosen that boost returns to shareholders
(see Figure 4.8).
The diametrically opposite strategy to business mix diversification relates
to focusing on the company’s core business and the outsourcing of non-core
business bank activities. Since true cost and revenues efficiency are often
difficult to find through diversification, it is argued that commercial banks
should focus on their core business activities since they possess superior
competencies (such as client relationships, skills, product competencies, risk
management knowledge, legal competencies, etc.). As noted in Hörther
(2000), various banks tend to focus on their core business. For example,
Lloyds TSB concentrates on retail- and mortgage banking products in
selected (mainly domestic) markets and, from July 1989 to June 1997,
Lloyds more than doubled its share price every three years with an annual
growth rate of 34.95 per cent. This strategy has also been implemented by
other commercial banks, such as Barclays. A recent study (European Banking
Briefing14) of competition in European banking systems notes most major
European banks have attempted to downsize and outsource non-core busi-
ness processes in order to reduce operating costs. For instance, Abbey
National noted on 13 January 2004 that it was to close its insurance division
‘Scottish Provident’, reducing its workforce by 730 people.
In many cases, downsizing and outsourcing policy often imply a staff reduc-
tion and a decline in the number of branches in order to rationalise bank
distribution channels.15 For example, the Finnish bank Nordea announced on
9 January 2004 its decision to reduce staff numbers by 200–300 employees.
Deutsche Bank reduced the number of branches (from 1,500 at the end of the
1990s to 770 by 2004) and is currently planning to reduce its workforce by
14,400 (6,000 in Germany) within two to five years. Similarly, Commerzbank
reduced its number of branches (from 920 at the end of 2000 to 720
by June 2003) and cut 6,000 employees (after having increased its staff by
8.000 between 1997 and 2001), and is planning to reduce it even further
by 1,000 staff by the end of 2004. Table 4.3 shows the decline in the number of
branches and staff of European and other banking sectors between 1990 and
2003 highlighting the aforementioned trends.
Table 4.3: Restructuring of the banking sector1 128
Concentration2 Number of branches Employees3
1990 1997 20034 1990 1997 20034 var.5 1990 1997 20034 var.5
United States 13 21 24 72.8 77.3 84.8 0.0 1911 1847 2129 0.0
Japan 42 39 42 24.7 25.4 22.7 –11.8 593 561 447 –27.8
Germany 176 17 22 43.3 47.1 38.2 –22.3 696 751 732 –3.5
France 52 38 45 25.7 25.5 26.2 0.0 399 386 384 –3.8
United Kingdom 49 47 41 19.0 14.3 12.9 –32.4 423 360 360 –15.0
Italy 24 25 27 17.7 25.6 29.9 0.0 324 343 341 –4.3
Canada 83 87 87 8.7 9.4 10.4 0.0 211 264 279 0.0
Spain 38 47 55 35.2 37.6 39.4 0.0 252 242 239 –6.6
Australia 65 69 77 6.9 6.1 4.9 –31.2 357 308 344 –3.4
Netherlands 74 79 84 8.0 7.0 3.7 –54.1 123 120 140 –9.6
Belgium 48 57 83 8.3 7.4 5.6 –33.2 79 77 75 –5.6
Sweden 70 90 90 3.3 2.5 2.0 –37.2 45 43 42 –7.7
Austria 35 44 44 4.5 4.7 4.4 –6.2 75 75 75 –2.4
Switzerland 54 73 80 4.2 3.3 2.7 –35.9 120 107 100 –16.8
Norway 68 59 60 1.8 1.6 1.2 –32.9 31 24 22 –22.4
Finland 65 77 79 3.3 1.7 1.6 –55.8 50 30 27 –49.3
1 Deposit-taking institutions, generally including commercial, savings and various types of mutual and cooperative banks.
2 Top five banks’ assets as a percentage of all banks’ assets. 3 In thousands. 4 For Belgium, France, Germany, Italy, Japan,
Sweden and the United Kingdom, 2002. 5 Change in per cent from peak (since 1990) to most recent observation; 0.0 indicates that 2003 was the peak
year. 6 1995.
Source: Bank for International Settlement (2004: 131).

In conclusion, banks can aim to create sustainable shareholder value by

analysing their activities in terms of closeness to core competencies and also
by redefining their business portfolio so as to choose activities with a positive
shareholder value added. Overall, the broad strategy that has been adopted by
banks is to diversify their shareholder value creating revenue streams by
boosting fee and commission income (cross-selling financial services) and at
the same time focusing on core competencies. In addition, branch infrastruc-
ture and staff numbers are being reduced as cost-cutting (efficiency) measures,
as shown by ECB (2004: 14): ‘banks achieved cost-cutting mainly by reducing
staff and administrative costs’.
Exogenous channels: financial consolidation

The term ‘exogenous channels’ relates to all possible strategies that create
shareholder value involving external parties. Although these strategies are
analysed separately, it is worthwhile noting that exogenous strategies should
not be considered as an alternative to endogenous channels, but as com-
plementary. Since financial companies can combine with each other in sev-
eral ways, our analysis focuses on the primary external channels that relate
to mergers and acquisitions (M&As), joint-venture and strategic alliance
activities.
Recent trends in M&A in banking

Probably, the most common exogenous approach to create shareholder value is
through M&A, which combines independent companies under common
control. In addition, banks can engage in joint-ventures and strategic alliances.
By a joint-venture, two or more independent companies create a new company
and provide their business skills to this new company. With a strategic alliance,
two or more independent companies contractually agree to act jointly in some
part of their business.
All these forms of activity have been widespread in the financial sector, as
confirmed in a report of the Group of Ten (2001). According to this study,
which focuses on thirteen countries (including the US, most continental
European countries and Japan), consolidation in the financial sector increased
in the 1990s. M&A activities were substantial since the annual number of deals
increased threefold and the total value of deals increased almost tenfold. It is
interesting to note that, during the last few years of the 1990s, M&A activity
accelerated and the average value of transactions increased substantially: this
increasing trend was widespread across all nations. Amel et al. (2003) report
that more than 10,000 financial firms were acquired in the major industrial
countries from 1990 to 2001. The acquired firm had a market value greater
than USD 1 billion in 246 cases. From the mid-1990s, M&A activities increased:
from 1990 to 1996, deals worth more than USD 1 billion were 93, while this
number grew to 153 between 1997 and 2001. In Table 4.4, we report some data
Table 4.4: Merger and acquisition in the main industrial countries1 130
All Mergers and Acquisitions Mergers and Acquisitions in the Financial Sector2
Total Of Which: Total Of Which:

1990–1995 1996–2001 1990–95 Banks3 1996–2001 Banks3
Total Total Total Total Total Total

Number Value Number Value Number Value Number Value Number Value Number Value
% of % of
US$b GDP US$b GDP US$b US$b US$b US$b
Australia 628 29.5 1.5 1,423 91.7 4.0 136 4.5 53 2.4 268 25.2 91 13.2
Belgium 251 7.1 0.5 354 57.8 3.9 67 4.5 21 0.8 70 32.9 34 28.1
Canada 1,421 41.6 1.2 2,888 287.4 7.3 156 3.9 52 1.6 321 36.0 112 15.0
France 1,663 81.9 1.0 1,563 269.6 3.2 314 25.5 148 11.8 227 73.7 96 44.6
Germany 1,913 37.3 0.3 3,039 437.0 3.5 234 11.0 123 2.4 379 82.6 229 68.6
Italy 852 55.0 0.8 1,048 198.2 2.9 251 24.8 147 19.2 236 97.6 138 80.4
Japan 216 56.1 0.2 2,291 234.5 0.9 46 45.4 29 44.4 491 138.1 236 119.1
Netherlands 565 25.6 1.3 635 127.2 5.5 123 14.5 36 10.9 88 33.9 24 5.9
Spain 510 25.6 0.8 1,042 99.3 2.8 120 8.3 66 5.9 153 34.2 67 31.2
Sweden 473 33.8 2.4 793 126.0 8.9 84 4.1 44 2.8 81 21.2 38 16.9
Switzerland 412 14.6 1.0 485 85.9 5.6 111 4.9 81 3.3 87 35.2 43 24.2
United Kingdom 2,349 170.9 2.7 4,484 848.6 10.3 386 41.4 140 33.0 750 226.1 279 114.4
United States 8,743 811.2 2.1 14,102 5,272.3 9.7 2,341 205.3 1,691 156.6 2,902 1,138.2 1,796 754.9
T. M. Industrial
Countries4:
Euro Area 19,996 1,390.2 1.3 34,147 8,135.5 6.1 4,369 398.2 2,631 295.1 6,053 1,974.9 3,183 1,316.6
6,767 256.0 0.7 9,696 1,310.3 3.4 1,317 99.8 655 59.6 1,406 412.3 700 302.8
World 26,062 1,570.3 50,787 8,960.2 5,725 460.9 3,363 340.3 9,777 2,232.9 4,781 1,494.9
1 Mergers and acquisitions involving majority interests. 2 The sectors refer to that of the company being acquired. 3 Includes: Commercial Banks, Bank Holding Companies,
Saving and Loans, Mutual Savings Banks, Credit Institutions, Real Estate; Mortgage Bankers and Brokers. 4 G10 countries, Australia and Spain.
Source: Amel et al. (2003: 3).
showing the increasing number of M&As worldwide by distinguishing between

the whole market and the financial sector.
Most of the M&A activity involved banking firms; for instance, the Group of
Ten (2001) note that acquisition of banks accounted for 60 per cent of all
financial mergers and for 70 per cent of the value of those mergers. Similarly,
the number of joint ventures and strategic alliances increased over the 1990s,
especially in the last two years of the decade. As a consequence of this trend, a
significant number of large, and in some cases increasingly complex, financial
institutions have been created. Focusing on M&A activity, the report of the
Group of Ten (2001) noted that all types of M&As (whether within domestic
markets or cross-border and within one industry segment or across segments)
increased in frequency and value during the 1990s. Domestic M&As are the
most common form, while cross-border M&As are less frequent. Table 4.5
reports a list of the most important domestic mergers in European banking
markets.
Regarding domestic M&As, M&As among companies competing in the same
segment of the financial services industry are more frequent than those involv-
ing firms in different segments. Most domestic mergers involved banking
organisations, but cross-border deals were roughly evenly divided between
banks and insurance firms (Tables 4.6 and 4.7). Regarding cross-border M&As
deals between companies competing in the same segment of the financial
services industry are more frequent than those involving firms in different
segments.
Looking at the trend over the 1990s in the thirteen national domestic
financial sectors analysed in the Group of Ten (2001) report, most of the M&A
activity (i.e. 55 per cent) occurred in the US. The overall level of M&A activity
(measured as a percentage of GDP) varied across countries: high levels were
observed in Belgium, Switzerland, UK and US; low levels were registered in
Canada, Germany and Japan. Concerning the number and the size of M&As, a
little growth in the number of deals was observed in France, Netherlands and
Switzerland, while the number of transactions rapidly accelerated in Japan at
the end of 1990s. Regarding the average value of M&A activity, Belgium and
Switzerland observed particularly large increases in activity at the end of the
decade. Financial sector consolidation in Japan and the US focused more on
the domestic market, while financial firms in other countries (especially
Belgium) were more heavily involved in cross-border deals. Concerning the
segment of the financial services involved in M&A activity, in Japan and
the US, M&As among companies competing in the same segment of the
financial services industry were more frequent than those involving firms in
different segments: this can be explained in large part due to the presence of
various legal restrictions that were only removed in the mid-1990s (such as the
repeal of the Glass-Steagall Act16 in 1999). In addition, M&A activity concen-
trated more heavily in banking in the US, while Australia, Canada, Netherlands
and the UK have experienced a greater proportion of M&As in the insurance,
Table 4.5: A selection of major domestic bank mergers in Europe
Belgium 1992 CGER – AG (Fortis)

1995 Fortis – SNCI
1995 KB – Bank van Roeselaere
1997 BACOB – Paribas Belgium
CERA – Indosuez Belgium
1998 KBC (KB – CERA – ABB)
2001 Dexia – BACOB
Denmark 1990 Den Danske Bank
Unibank (Privatbanken, Sparekassen, Andelsbanken)
1999 Unibank – TrygBaltica
2000 Danske Bank – RealDanmark
Finland 1995 Merita Bank (KOP – Union Bank of Finland)
France 1996 Crédit Agricole – Indosuez
1999 BNP – Paribas
Germany 1997 Bayerische Vereinsbank – Bayerische Hypobank (HBV)
2001 Allianz – Dresdner
Italy 1992 Banco di Roma (Banco di Roma, Cassa di Risparmio
di Roma, Banco di Santo Spirito)
San Paolo – Crediop
1995 Credito Romagnolo (Rolo) – Credit Italiano
(UniCredito)
1997 Ambroveneto – Cariplo (Intesa)
1999 San Paolo – IMI
Intesa – BCI
SanPaoloIMI – Banca di Napoli
2000 Banca di Roma – Bipop (Capitalia)
Netherlands 1990 ABN – AMRO
1991 NMB – PostBank – ING
Portugal 1995 BCP – BPA
2000 BCP – BPSM
Spain 1988 BBV (Banco de Vizcaya – Banco de Bilbao)
1989 Caja de Barcelona – La Caixa
1992 Banco Central – Banco Hispano
1994 Santander – Banesto
1999 Santander – BCH
BBV – Argentaria (BBVA)
Sweden 1993 Nordbanken – Gota Bank
Switzerland 1993 CS – Volksbank – Winterthur
1997 SBC – UBS
United Kingdom 1995 Lloyds – C&G – TSB
2000 RBS – NatWest
Barclays – Woolwich
Abbey Nat. – Scottish Provident
2001 Halifax – Bank of Scotland
Source: Dermine (2002: 64–5).

Table 4.6: A selection of cross-border acquisitions of merchant banks
Buyer Target
Deutsche Bank Morgan Grenfell

ING Bank Barings
Swiss Bank Corp Warburg, O’Connor, Brison, Dillon Read
Dresdner Kleinwort Benson
ABN – AMRO Hoare Govett
UNIBANK ABB Aros
Merrill Lynch Smith New Court
FG (Spain), MAM
Morgan Stanley Dean Witter AB Asesores
CSFB BZW (equity part)
Société Générale Hambros
Citigroup Schroder
Chase Robert Fleming
ING Charterhouse Securities
Source: Dermine (2002: 66).
Table 4.7: A selection of cross-border acquisitions of commercial banks
Buyer Target
Dexia (B, F) Crédit Communal (B), Crédit Local (F), BIL (L),
Crediop (I), BACOB (B)
BACOB (B) Paribas (NL)
ING (NL) BBL (B), BHF (G)
GENERALE BANK (B) Crédit Lyonnais (NL), Hambros (UK, corporate)
FORTIS (B, NL) AMEV+Mees Pierson (NL) / CGER/SNCI (B)/
Generale Bank (B)
NORDBANKEN (S) Merita (F), Unidanmark (DK), Christiania (N)
BSCH (E) Champalimaud (P)
HSBC (UK) CCF (F)
Hypovereinsbank (D) Bank Austria – Creditanstalt (A)
Source: Dermine (2002: 66).
securities and other segments of the financial industry. It is worthwhile noting

that M&A activity (measured by the value of firms acquired) sharply acceler-
ated at the end of the 1990s: two-thirds of M&A activity occurred during the
decade’s last three years.
Regarding joint ventures and strategic alliances, US firms accounted for
nearly half of all joint ventures and strategic alliances, and these were mostly
domestic arrangements. In the remaining twelve countries, cross-border joint
ventures and strategic alliances were globally more common than domestic
deals: this is a remarkable difference from the trends in M&As.
An important question to address is to investigate when the acquisition of

another financial company or the merger between two banks creates value for
shareholders. The problem may be clarified using the example proposed by
Merton Miller after it was announced that he had won the Nobel prize. Miller
said ‘Let me give you a very simple analogy. Whether you cut a pizza into six
or eight slices doesn’t change the size of the pie, right? Well, it’s the same
with corporations.’ Adapting the example to consolidation activities, one
might ask when two pizzas are pasted together does the overall size of the pie
change? Regarding pizzas, the answer is clearly negative. However, the answer
is not straightforward for company M&As.
In order to establish when consolidation activities create shareholder value,
it is necessary to recall that shareholder value creation results from the follow-
ing decisions: increasing expected cash flows; reducing the hurdle rate; and
making financing type as close as possible to the assets being financed. Once
consolidation activity has achieved at least one of these results (keeping con-
stant the other conditions) the value of the new company should be greater
than the simple sum of the two companies’ values.
As a consequence, consolidation activities can be accurately evaluated only
on a case by case basis. However, without trying to be exhaustive, we believe
it is useful to briefly analyse the reasons usually proposed by managers to
justify consolidation actions. Regarding this point, the Group of Ten (2001)
investigated the motives for consolidation by a review of the literature and
interviews with forty-five selected industry participants.17 The economic
reasons put forward as to why M&As may create shareholder value relate to
possible cost savings, revenue enhancement and the advantages associated
with banks’ culture exchanges.
Realisable cost savings are usually the first reason put forward justifying
M&A activities in banking. M&As can lead to a reduction in banks’ costs if
there are significant economies of scale and/or scope and/or cost X-efficiency
gains. When these conditions are verified, M&As have a sound economic
rationale since these activities lead to a cost reduction which cannot be
achieved in other ways by shareholders. As a consequence, M&As can lead to
an increase in future cash flow (keeping constant other conditions) and the
value of the new company is greater than the mere sum of the two old com-
panies. Regarding the practical importance of economies of scale and scope,
Group of Ten (2001) found that:
(1) Economies of scale are a strong motivating factor in domestic consolidation

activities involving financial companies in the same market segment:
economies of scale are considered ‘very important’ by 80 per cent of the
respondents and ‘moderately important’ by 13.3 per cent of respondents.
These data contrast sharply with several empirical studies that have found
that economies of scale do not appear to be highly significant in the
banking sector (especially in the US, but also in Europe). Group of Ten
(2001: 69) explains this difference as follows: ‘econometric studies are

backward looking, making it difficult to achieve reliable estimates of scale
economies that can explain the current industry consolidation. Several
interviewees explained, for example, that the large investments required to
take advantage of the latest technological advances or to develop innova-
tive products could only be undertaken by very large organisations. Others
noted that mergers provide an opportunity to reduce staffing and elimi-
nate branches, thereby reducing costs.’ In domestic cross-segment M&As,
economies of scale are considered less important: economies of scale are
judged ‘very important’ by 12.5 per cent of the respondents and ‘moder-
ately important’ by 17.5 per cent of respondents. These findings reported
by the Group of Ten (2001) are expected from an economic standpoint
since scale gains are more easily achievable by M&A activity between two
similar companies (i.e. companies within the same market segment) rather
than between two different companies (such as companies operating in
two different segments of the financial market).
(2) Economies of scope are considered by respondents more important in
cross-segment M&As than within segment consolidation. Economies of
scope are judged ‘very important’ by 25 per cent of the respondents
and ‘moderately important’ by 30 per cent of respondents, while, in
the second case, the cost savings are judged ‘very important’ by 4.5 per
cent of the respondents and ‘moderately important’ by 25 per cent of
respondents.
The second economic reason put forward as to why M&As create share-
holder value is revenue enhancement. According to the Group of Ten (2001:
12) ‘consolidation can lead to increased revenues through its effect on the
firm size, firm scope (through either product or geographic diversification) or
market power. Research suggests that mergers may provide some opportuni-
ties for revenue enhancement either from efficiency gains or from increased
market power.’ Regarding the practical importance of these causes of revenue
enhancement, the Group of Ten (2001) reports that revenue enhancement is
due to the following: (1) increased size is considered slightly more important
for revenue enhancement in domestic within-segment consolidation than in
cross-segment deals; (2) product diversification is considered more important
in cross-segment consolidation than in within-segment (i.e. revenue enhance-
ment due to product diversification is considered ‘very important’ by 2.3 per
cent of respondents and ‘moderately important’ by 9.1 per cent of respon-
dents in domestic within-segment consolidation, while it is considered ‘very
important’ by 15 per cent of respondents and ‘moderately important’ by
20 per cent of respondents in domestic cross-segment segment consolidation).
This result could be largely expected since within-segment M&A does not lead
to product diversification; (3) a change in the organisational focus is consid-
ered more important in cross-segment consolidation than in within-segment
deals. Similarly to product diversification, this result could be largely

expected since within-segment M&As do not lead to a change in banks’
organisation; (4) increased market power is considered relevant by respon-
dents in both within- and cross-segment M&As, but it is considered more
important in within-segment consolidations.
These findings provide an interesting practical overview of the expected
cost and revenue features of M&As. However, one might note that M&A
activities have a sound economic rationale only when they lead to gains
that could not be achieved by shareholders in other ways. In other words,
while it is straightforward that M&As lead to an increase in future cash
flows (keeping constant other conditions) and the value of the new compa-
nies is greater than the mere sum of the two old companies, one might cast
doubts as to whether shareholders could achieve these gains in another
way (e.g. by diversifying their portfolio). According to Group of Ten (2001:
12), the motives advanced by practitioners for M&As are the following:
• ‘consolidation can lead to increased revenues through its effect on the firm
size, firm scope (through either product or geographic diversification)…’
According to this statement, revenue enhancement derives from the
revenue effect generated by economies of scale and scope. In the case of
scale efficiency, consolidation enables a bank to operate at optimal size
which leads presumably to lower financing costs and therefore higher
revenues. In our opinion, this revenue enhancement cannot be achieved
by shareholders in other ways and consequently M&As have a sound
economic rationale. In the case of scope economies, consolidation
enables a bank to diversify its product and/or its geographical activities
which can lead to superior risk-return combinations and, as a conse-
quence, to enhanced revenues. In our opinion, risk diversification does
not justify M&A activity: the same result (i.e. an improved risk-return
combination) can be achieved cheaply by shareholders acquiring:
(1) stock of companies working in other financial segments (i.e. equiva-
lent for bank’s shareholders to the bank’s product diversification);
(2) stock of banks operating in different banking markets (i.e. equivalent
for bank’s shareholders to the bank’s geographic diversification).
• ‘… or market power. Research suggests that mergers may provide some opportuni-
ties for revenue enhancement either from efficiency gains or from increased market
power.’ According to this statement, the relationship between market struc-
ture and firm profitability is positive. Although this result is generally
accepted in the literature,18 there is no agreement on the hypotheses which
generate it. Four hypotheses have been proposed to explain the positive
relationship between market structure and corporate profitability. These are
the traditional Structure-Conduct-Performance (SCP), the Relative-Market-
Power (RMP) and the Efficient-Structure (ES) hypotheses in the form of
X-efficiency or Scale Efficiency.
As Berger (1994) notes, the Traditional SCP hypothesis identifies a setting of

prices less favourable to customers in more concentrated markets as the cause
of the positive relationship between profitability and market structure.
According to this theory, companies have higher profits in more concentrated
markets because they can charge higher prices (than those set in competitive
markets) having a stronger market power. Similarly, the RMP hypothesis states
that companies with high market shares and well-differentiated products can
exploit their market power by setting prices to earn supernormal profits. A dif-
ferent explanation of the positive relation between profits and firm profitabil-
ity is supplied by the ES hypothesis. This hypothesis asserts that more efficient
companies have lower costs which directly increase profits: in this way, these
firms can increase their market share determining a higher market concentra-
tion. The ES hypothesis has usually been proposed in two different forms,
depending on the type of efficiency considered. In the X-efficiency form,
more efficient firms have lower costs, higher profits and larger market share
because they have a superior ability in minimising costs to produce any given
outputs. In the Scale Efficiency form, the same relationship described above is
due to the fact that more scale efficient firms produce closer to the minimum
average-cost point. In both forms of the ES hypothesis, the positive relation-
ship between profits and market structure is not direct but spurious, efficiency
being the main determinant of both profits and market share. In another
study, Berger and Hannan (1998) found strong evidence that US banks
in more concentrated markets show lower cost efficiency. According to
the authors, additional operating costs due to market concentration in
US banking are substantially larger (up to twenty times) than the social loss
attributable to the non-competitive pricing of bank outputs.
In terms of shareholder value creation, whether a positive relationship
between market structure and firm profitability is verified, M&A activities have
a sound economic reason. By increasing corporate size, a bank can achieve
revenue enhancement through increased market power and this is not achiev-
able in other ways by shareholders. It is interesting to note that this result is
independent from the reasons underlying the relationship between market-
structure and firm profitability (i.e. the traditional SCP, the RMP and the ES
hypotheses in the form of X-efficiency or scale efficiency), but it is subordinate
to the existence of a positive relationship.19
The third major economic justification put forward as to why M&As
create shareholder value relates to banks’ culture exchange. Beyond the
two reasons analysed above, banks can create shareholder value by M&A
when there is an exchange between two different companies’ cultures. The
exchange of corporate culture has various important features:
(1) Modality of this exchange. An exchange between corporate cultures can

imply the import or the export of a dominant culture. Without aiming to
be exhaustive and focusing on consolidation within the same financial
segment, the two most common forms of exchange between corporate

cultures are the following: (a) when a large bank acquires a smaller
bank, the former is likely to export its corporate culture by undertaking
a sort of ‘cultural colonisation’. Under these circumstances, managers
of the acquiring bank need to consider that the corporate culture is
likely to be rigid in the short term and, therefore, there will be some
resistance to changes from affected parties in the acquired bank; (b) a
large bank which acquires a smaller bank may import the latter’s cor-
porate culture. This approach involves a process of incorporation of the
new culture into the bank which would probably lead to a substantial
change of the existing corporate culture. Under these circumstances,
probably none of the two corporate cultures will be destroyed, but
these should be mixed in an optimal manner: the goal is to enrich the
existing culture with some particular features of the corporate culture
of the acquired bank. In this type of corporate culture exchange, the
main risk concerns a possible cannibalisation in the mid-term of the
acquired culture: if the corporate culture mixing process is not properly
managed, actors in the acquired banks will tend to colonise the culture
of the smaller bank. When mergers between similar sized banks occur
then it is up to the two banks to choose the best culture from the two
firms.
(2) The exchange of corporate cultures can lead to shareholder value cre-
ation. With this purpose, it is useful to recall that shareholder value
can be created by increasing the expected cash flows, reducing the
hurdle rate and making financing type as close as possible to the assets
being financed. As a consequence, culture exchanges can lead to share-
holder value creation when at least one of these conditions is met. In
our opinion, at least one of these conditions is met when one of the
corporate cultures involved in the consolidation is superior to the
other. When a large bank acquires a smaller bank and exports its corpo-
rate culture realising cultural colonisation, this can create shareholder
value as long as the culture of the acquiring bank is superior to the
acquired. Let us consider a bank which is oriented to shareholder value
creation (i.e. customer-oriented, cost efficient, with excellent manager-
ial skills in terms of ability of re-engineering business process or organ-
isational structure) and these features permeate all players (e.g. top
managers, middle managers, front-office and back-office) within the
bank and all the company’s relationship with the economic environ-
ment. This bank can create shareholder value by exporting its culture
into another bank with a culture not oriented to shareholder value cre-
ation (e.g. a bank cost inefficient and/or poorly oriented to customers
and/or with business process/organisational structure inefficiently
designed). In this case, M&A activity can lead to cost savings (e.g. due
to improvement in X-efficiency and/or business process re-engineering
and/or a more effective organisational structure) and revenue enhance-

ment (e.g. obtained by improving customer satisfaction). In this case,
shareholder value is created since future cash flows increase and the bank’s
risk (and consequently the hurdle rate) reduces: this result is obtained by
investing in the bank’s intangible resources (e.g. customer and employee
satisfaction) and it cannot be achieved by shareholders in any other way.
M&A and bank performance – empirical evidence

Empirical studies dealing with M&A in banking can be roughly divided
into two groups: studies investigating the overall performance of bank
mergers and studies assessing the factors that can explain M&A success.
Regarding the first branch of the literature, most studies dealing with
M&As in banking systems are in this group (as listed in Table 4.8) and
recent surveys can be found in Pilloff and Santomero (1998), Berger et al.
(1999) and Beitel and Schiereck (2001). Three different approaches have
usually been applied to investigate the performance of bank mergers: event
studies, dynamic efficiency studies and operating performance studies. In
event studies, the analysis focuses on abnormal returns of share prices over
a time interval (labelled as ‘window’) around the announcement of the
M&A. In other words, some analyses have considered a short window of
three or five days, which is constructed looking at one or two days before
and after the announcement, in order to look for abnormal returns by esti-
mating Cumulative Abnormal Returns (CuAR).20 This method enables one
to isolate the return of a share i at the time t (Ri,t), from the expected return
of the share i at the time t (i.e. market model) or the return of the market at
the same time (market adjusted model). Dynamic efficiency studies21 apply
parametric and non-parametric models for estimating efficiency to investi-
gate if the efficiency of merging and acquiring banks improved. Operating
performance studies apply mean-difference tests using accounting data to
assess significant changes in profitability and other ratios (mainly, ROE and
ROA) from a time period prior to the M&A transaction to a period there-
after (e.g. three to five years). Pilloff and Santomero (1998) note that the
event study methodology is the most direct approach for assessing value
implications of bank M&A activity.
The evidence found in the previous literature is mixed. In their survey of
more than 100 studies, Beitel and Schiereck (2001) report that about a
quarter of studies found significantly positive cumulative abnormal returns
(CuAR) in the event windows scrutinised for both target and bidder share-
holders. Instead, many other studies find mixed or negative results,
showing negative CuARs especially for bidding banks and observe that only
a subset of M&As create (or destroy) value across the analysed sub-samples
as shown in Table 4.9.
The other branch of the literature attempts to identify the factors that
can explain M&A success, such as the relative asset size of target banks in
Table 4.8: Overview of the banking-related M&A research over the last twenty years
Authors Study Authors Study
Lobue (1984) Ev Peristiani (1993b) Ef

Desai and Stover (1985) Ev Rhoades (1993) Ef
Dubofsky and Fraser (1985) Ev Schranz (1993) Su
Pettway and Trifts (1985) Ev Shaffer (1993) Ef
Rhoades (1986) Pe Tebroke (1993) Pe
De and Duplicham (1987) Ev Spindt and Tarhan (1993) Pe
James and Wier (1987) Ev Houston and Ryngaert (1994) Ev
James and Wier (1987b) Ev Madura and Wiant (1994) Ev
Neely (1987) Ev Palia (1994) Ev
Neely and Rochester (1987) Pe Haun (1996) Pe
Rose (1987a) Pe Lausberg and Rose (1995) Su
Rose (1987b) Pe Zhang (1995) Ev
Trifts and Scanlon (1987) Ev Seidel (1995) Ev
Boyd and Graham (1988) Pe Altunbas, Gardener et al. (1996) Ef
Cossio, Trifts et al. (1988) Ev Hudgins and Seifert (1996) Ev
Hoshino (1988) Pe Keeton (1996) Ku
Rose (1988) Pe Molineux, Altunbas et al. (1996) Ef
Sushka and Bendeck(1988) Ev Pilloff (1996) Ev
Bertin, Ghazanfari et al. (1989) Ev Siems (1996) Ev
Dubofsky and Fraser (1989) Ev Strahan and Weston (1996) Ku
Hannan and Wolken (1989) Ev Vander Vennet (1996) Pe
Kaen and Tehranian (1989) Ev Houston and Ryngaert (1996) Ev
Wall and Gup (1989) Ev Akhavein, Berger et al. (1997) Ef
Baradwaj, Fraser et al. (1990) Ev Akhavein, Berger et al. (1997) Ku
Hawawini and Swary (1990) Ev Berger and Mester (1997) Ef
Rhoades (1990) Pe De Young (1997) Ef
Allen and Cebenoyan (1991) Ev Houston and Ryngaert (1997) Ev
Cornett and De (1991) Ev Kolari and Zardkoohi (1997) Ku
Hoshino (1991) Pe Lang and Welzel (1997) Ef
Savage (1991) Ef Peristiani (1997) Ef
Spindt and Tarhan (1991) Pe Subrahmanyam, Rangan Ev
et al. (1997)
Baradwaj, Dubofsky et al. (1991) Ev Walraven (1997) Ku
Berg (1992) Ef Berger (1998) Ef
Berger and Humphrey (1992) Ef Berger, Saanders et al. (1998) Ku
Cornett and Tehranian (1992) Ev Banerjee and Cooperman (2000) Ev
Linder and Crane (1992) Pe Chamberlain (1998) Pe
Murphy (1992) Ef Peek and Rosengren (1998) Ku
O’Keefe (1992) Pe Prager and Hannan (1998) Ku
Rose (1992) Pe Resti (1998) Ef
Shaffer (1992) Ef Rhoades (1998) Ef
Spong and Shoenhair (1992) Pe Singh and Zollo (1998a) Su
Srinivasan (1992) Pe Strahan and Weston (1998) Ku
Srinivasan and Wall (1992) Ef Toyne (1998) Ev
Toevs (1992) Pe DeLong (1999) Ev
Boyd, Graham et al. (1993) Pe Avery, Bostic et al. (1999) Ku
Table 4.8: Overview of the banking-related M&A research over the last twenty years –
continued
Authors Study Authors Study
Crane and Linder (1993) Pe Cyree and DeGennaro (1999) Ev

De Young (1993) Ef Fried, Lovell et al. (1999) Ef
Fixler and Zieschang (1993) Ef Haynes and Thompson (1999) Ef
Peristiani (1993a) Ef Hughes, Lang et al. (1999) Ef
Kwan and Eisenbeis (1999) Ev
Where:
Ev = event study
Ef = dynamic efficiency study
Pe = performance study
Su = other survey
Ku = customer focused research
Source: Beitel and Schiereck (2001: 38).
comparison to the bidder banks, the profitability of bidder and target

banks, the geographic focus of M&A transactions, the growth focus of M&A
transactions, the risk reduction potential derived from M&A activities, the
cost efficiency of target and bidder banks, the method of payment, etc.
Among these studies, we focus on some recent studies (mostly on US
banking), such as Hawawini and Swary (1990), Houston and Ryngaert
(1994), Pilloff (1996), De Young (1997), Houston and Ryngaert (1997),
Becher (1999), Banerjee and Cooperman (2000), Zollo and Leshchinkskii
(2000), De Long (2001), Cornett et al. (2003) and Beitel et al. (2004).
Hawawini and Swary (1990) analyse the impact of various factors (such
as the size of the target bank, potential risk reduction, the profitability of
the bidder bank and management quality of the target bank) that impact
on success of bank M&A by analysing a sample of 123 mergers between
1972 and 1987 in the US. Regarding the impact of the size of a target on
M&A success, Hawawini and Swary (1990) found that: (1) bidder banks can
be more favourable M&A transactions if target banks have a small size in
comparison to bidders; (2) smaller bidder banks tend to be more successful
than larger bidder banks. Concerning the second issue, Hawawini and
Swary (1990) test the correlation between stock prices of the target and
bidder banks and found that transactions involving highly correlated part-
ners create more value for the bidding banks. In investigating the impact of
the profitability of the bidder bank they found that bidders that are rela-
tively more profitable than their targets do create significantly more value
in bank M&As. Finally, the study used the stock performance of the target
bank as a proxy for the management quality of the target company (follow-
ing Jensen and Ruback, 1983). The authors found that bidder banks achieve
a higher value by acquiring target companies with poor stock performance
Table 4.9: Cumulated abnormal returns for event studies of the last ten years 142
Event CuAR CuAR CuAR
Regional Years window bidder target combined
Study Year focus studied N in days (%) (%) entity
Hawawini, Swary 1990 USA 72–87 123 [0;+5] –1.7 +11.5 +3.1
Baradwaj et al. 1990 USA 80–87 53 [–60;+60] n.s. 25.9–30.3 N/A
Allen, Cebenoyan 1991 USA 79–86 138 [–5;+5] n.s. N/A N/A
Cornett, De 1991 USA 82–86 152 [–15;+15] n.s. +9.7 N/A
Cornett, Tehranian 1992 USA 82–87 30 [–1;0] –0.8 +8 N/A
Baradwaj et al. 1992 USA 81–87 108 [–5;+5] –2.6 N/A N/A
Houston, Ryngaert 1994 USA 85–91 153 [–4L; +1A] –2.3 +14.8 +0.5
Madura, Wyant 1994 USA 83–87 152 [0;36M] –27.1 N/A N/A
Palia 1994 USA 84–87 48 [–5;+5] –1.5 N/A N/A
Seidel 1995 USA 89–91 123 [–20;+20] +1.8 N/A N/A
Zhang 1995 USA 80–90 107 [–5;+5] n.s. +6.9 +7.3
Hudging. Seifert 1996 USA 70–89 160 [–1;+1] n.s. +7.8 N/A
Siems 1996 USA 1995 19 [–1;+1] –2 +13 N/A
Pilloff 1996 USA 82–91 48 [–10;0] N/A N/A +1.4
Houston. Ryngaert 1997 USA 85–92 209 [–4L; +1A] –2.4 +20.4 N/A
Subrahmanyam et al. 1997 USA 82–87 263 [–1;+1] –0.9 N/A N/A
Banerjee et al. 1998 USA 90–95 92 [–1;0] –1.3 +13.1 N/A
Toyne, Tripp 1998 USA 91–95 68 [–1;0] –2.2 +10.9 –0.7
Cyree, DeGennaro 1999 USA 89–95 132 [–1;0] n.s. N/A N/A
Kwan, Eisenbeis 1999 USA 89–96 3844 [–1;0] –2.2 +10.9 –0.7
De Long 1999 USA 88–95 280 [–10;1] –1.7 +16.6 n.s.
Tourani–Rad et al. 1999 Europe 89–96 17;56 [–40;+40] n.s. +5.7 N/A
Cornett et al. 2000 USA 88–95 423 [–1;+1] –0.78 N/A N/A
Cybo-Ottone et al. 2000 Europe 87–98 46 [–10;0] n.s. +16.1 +4
Becher 2000 USA 80–97 558 [–30;+5] –0.1 +22.6 +3.0
Brewer et al. 2000 USA 90–98 327 [0;+1] N/A +8.3–14 N/A
Houston et al. 2000 USA 85–96 64 [–4L; +1A] N/A N/A +3.1
Kane 2000 USA 91–98 110 [0] –1.5 +11.4 N/A
Karceski et al. 2000 Norway 83–96 39 [–7;0] n.s. +8.4 N/A
Schiereck, Strauss 2000 USA/Ger 98–99 1 [–20;+20] n.s. +30.1 N/A
Zollo et al. 2000 USA 77–98 579 [–10;+10] N/A N/A N/A
Source: Beitel and Schiereck (2001: 4).

prior to the transaction rather than acquiring target banks with better stock
performance.
Houston and Ryngaert (1994) investigate the impact of profitability in M&A
activity using a sample of 153 US bank M&As over the period 1972–87 and
focus on the combined bidders and targets. The authors found that bidders
that are relatively more profitable than their targets do create significantly
more value in bank M&As.
Pilloff (1996) investigates the role of cost efficiency in bank M&A deals by
focusing on the combined entity’s returns using a sample of forty-eight
mergers over the period 1982–91. The author found that cost efficiency
improvements after the transaction are positively correlated to value creation
in M&A transactions. De Young (1997) investigated the impact of the bidder
experience (measured by the frequency of conducting M&A transactions) on
M&A success by undertaking a dynamic efficiency study of 348 US bank deals
between 1987 and 1988. The author found that banks that undertook M&A
activity had a positive impact on overall performance.
Houston and Ryngaert (1997) investigated the impact of the geographic
focus of M&A transactions considering a sample of 209 US banking deals
between 1985 and 1992. According to their findings, geographic focus seems to
have a positive impact on M&A success. Becher (1999) investigates the impact
of the method of payment in US bank M&A and found that M&A activity
creates more shareholder wealth in banking for the bidder’s shareholders in
cash transactions as compared to stock transactions.
Banerjee and Cooperman (2000) focus on the profitability differences
between target and bidder banks for thirty bidders and sixty-two target banks
in the US between 1990 and 1995. They find that bidding banks are more
successful when they are more profitable than their targets. Zollo and
Leshchinkskii (2000) use a sample of 579 US deals over the period 1977–98 to
investigate the impact of size and the experience (in M&A activity) of the
bidder bank on M&A success. The authors found that the bidder bank size has
a substantial negative impact and report a substantial positive correlation
between the bidder’s experience and its CuAR.
De Long (2001) analyses a sample of 280 transactions in US banking over the
period 1988–95 and finds that deals that focus primarily on growth are
significantly more value creating than transactions that focus or diversify in
only one direction. In addition, De Long (2001) analysed the impact of stock
performance of target banks prior to the M&A transaction and found that
M&A activities create more value if target banks show poor stock performances,
compared to their peer-group.
Cornett et al. (2003) assess the impact of the growth of bank M&A deals
from a bidder viewpoint and examine a sample of 423 transactions in US
banking between 1988 and 1995. The authors found that product/activity
focus has a significantly positive impact on the value creation in US bank
M&As.
Beitel and Schiereck (2004) identified thirteen variables with explanatory

power and test their impact on M&A success (from the viewpoint of target and
bidder shareholders and of the combined entity of bidder and target) of
European bank M&As between 1985 and 2000 using comparative statistics
with mean-difference tests and multivariate cross-sectional regression analysis.
They found contrary results for the viewpoint of target and bidder sharehold-
ers. The authors found that successful bidders can be identified by investigating
their choice of a target since successful bidders chose smaller and faster
growing targets with poor relative efficiency. The authors found that large dif-
ferences in cost efficiency between target and bidder banks, as well as a poor
stock performance of the target prior to the transaction, are significant factors
in creating value for the targets’ shareholders.
In conclusion, empirical studies do not provide clear evidence whether M&A
activities create value for shareholders. In contrast, various studies identify
some factors having a clear positive impact on the overall results of the merger
(such as the accurate choice of the bank target, the bidder’s experience, the
bidding bank’s profitability etc.) and others with a negative impact (such as the
size of bidder and type of target bank).
Conclusion
This chapter has analysed how commercial banks can create shareholder value.
From an analytical point of view, shareholder value can be created by increas-
ing expected cash flows, reducing the hurdle rate and making financing type as
close as possible to the assets being financed. All strategies achieving at least
one of these effects will lead to the creation of sustainable shareholder value. As
such, this chapter outlined a set of strategies that appear able to improve, ceteris
paribus, at least one of the shareholder value drivers. These strategies have been
analysed by distinguishing between endogenous channels (i.e. strategies imple-
mented within the bank) and exogenous channels (i.e. strategies that create
shareholder value by involving external parties).
Regarding the endogenous channels, most of these focus on the optimal
management of bank stakeholders since creating stable shareholder value
requires an intense focus on delivering benefits to other stakeholders: for
example, by satisfying customers in the most efficient way, hiring and retain-
ing a motivated workforce, maintaining excellent supplier relationships and
being a good corporate citizen in each of the local communities where the
company has a presence.
The first major endogenous channel relates to customer satisfaction.
Customers are fundamental in creating shareholder value since they supply
sale proceeds, which are necessary to remunerate shareholders and, also, all
other stakeholders. The second strategy for creating shareholder value con-
cerns the improvement in banks’ efficiency. In order to improve banks’
operating efficiency, it is necessary to improve the management of the
other stakeholders, especially human resources. The third endogenous

strategy to improve shareholder value relates to improving the business
mix. In this regard, banks can focus on diversification of their services and
products or alternatively they can focus on a core business strategy. In
reality, they seem to do both. The final endogenous strategy discussed
concerns achieving optimal capital structures. There are four possible ways
to reduce the WACC for any company: reducing operating risk, reducing
operating leverage, changing the financing mix and changing the financing
type. Focusing on commercial banking, we presented a practical approach
to the definition of optimal capital structure.
Another possible way to create shareholder value is through restructuring via
M&A, joint-ventures, out-sourcing and other types of financial reorganisation/
consolidation. Financial companies can create shareholder value by combining
with each other with the aim of improving cost efficiency, economies of scale,
economies of scope, product or geographic diversification and increasing
market power.
A large number of studies have attempted to assess if banks create share-
holder value through M&A. The evidence found in this literature is mixed.
Around a quarter of studies find significantly positive Cumulative
Abnormal Returns in the event windows scrutinised for both target and
bidder shareholders, but many other studies find mixed or negative results,
showing negative Cumulative Abnormal Returns especially for bidding
banks. Another branch of literature investigates the impact of different
factors (such as the relative asset size of target banks in comparison to
bidder banks, the profitability of bidder and target banks, the geographic
focus of M&A transactions, the growth focus of M&A transactions, the risk
reduction potential from M&A activities, the cost efficiency of target and
bidder banks and the method of payment, etc.) on M&A success. Various
studies identified some factors having a positive impact on the overall
results of the merger (such as the accurate choice of the bank target, the
bidder’s experience, the bidding bank’s profitability etc.) and others with a
negative impact (such as the size of bidder and target banks).
5
Bank Performance Measures and
Shareholder Value
Introduction
Building on the discussion in previous chapters here we present an empiri-
cal investigation to assess which type of performance measure best explains
shareholder value creation in banking. Our analysis is undertaken on a
range of European listed banks and measures shareholder value created
across European banking sectors over the same period.
The primary aim of the chapter is to measure and empirically evaluate
the information content of the most common performance measures used
in the banking industry. As discussed in Chapter 3, it is a question of
debate as to which is the best method for assessing the value created by
firms for their owners, and as a consequence researchers and practitioners
have grappled with various performance metrics. In the first part of the
chapter, we undertake an empirical investigation of the relative informa-
tion content (i.e. the association between stock market values and alterna-
tive accounting measures) by looking at difference in explanatory power
(R2) of regressions where share prices (or raw share returns1) are the depend-
ent variables and common bank performance measures are the independ-
ent variables. We analyse the most traditional bank performance measures
(such as interest margin, intermediation margin, net income, ROE and
ROA) and two more shareholder value oriented measures (namely, Residual
Income and Economic Value Added – EVA). Our sample comprises French,
German, Italian and UK listed banks over the period 1995–2002.
The second part of the chapter undertakes an empirical investigation of
shareholder value (measured by Economic Value Added) focusing on four
primary European banking industries: France, Germany, Italy and the
United Kingdom. While publicly available statistics usually consider the
major companies (and therefore, both financial and non-financial firms)
listed in a country, our analysis considers only banks, but here we consider
both listed and non-listed banks. As a consequence, our results provide
novel evidence (as far as we are aware) about the ability of both listed and
146
Bank Performance Measures and Shareholder Value 147
non-listed banks to create sustainable shareholder value over a period of time.

Moreover, our analysis considers a range of bank ownership types (such as
commercial banks, savings banks and cooperative banks) in order to assess if
there are substantial differences in shareholder value creation for different
types of banking firms.
Bank performance measures and shareholder value – methodology

This chapter aims to assess the information content of a broad range of per-
formance measures in the context of creating shareholder value. Our focus is
on European banking sectors over an eight-year period (1995–2002). As seen
in Chapter 3, various studies have analysed the information content of perfor-
mance measures, but few have focused on banking and the information
content of performance measures oriented to shareholder value (such as EVA)
has rarely been tested. Our study intends to extend the existing literature by
running a standard relative and incremental association study with various
departures to take into account banks’ peculiar features. Firstly, we illustrate
the performance measures assessed in the empirical investigation and then we
analyse the econometric model to test the value relevance of these perfor-
mance measures. We prefer to illustrate the definition of bank performance
measure before the description of the empirical model since an accurate
definition of the variables expressing shareholder value created and the cal-
culation of performance measure seems to be the most critical issue in value-
relevance analysis, while the model follows a standard regression approach.
The definition of bank performance measures

Defining the variable expressing shareholder value created over a period is a
particularly interesting issue since the solutions adopted in earlier studies are
heterogeneous. Previous studies have generally adopted Market Value Added
(MVA), Market Raw Returns (RAWR) or Market Adjusted Returns (MAR) as
measures of shareholder value as the dependent variable in regression models.
In detail, some studies (such as O’Byrne (1996), Peterson and Peterson (1996)
and Biddle et al. (1997, 1999)) use equity market returns to express share-
holder value created and, consequently, evaluate firm performance measures
(such as accounting earnings, residual income and EVA) by examining their
correlation with stock returns on the ground that the best measure is the most
highly correlated with stock returns. Other studies (e.g. Al Ehrbar (1998) and
Uyemura et al. (1996)) compare firm performance measures looking at the
degree of correlation with the Market Value Added.
The selection of the variable capturing shareholder value (added) created
over a given time period is particularly important since results need to be
interpreted according to the variable selected. Although all these measures
(MVA, RAWR and MAR) provide useful information, our goal is to investi-
gate shareholder value (added) over a given time period and, as such, MAR is
more suitable. In fact, for our analysis, we define MAR as the increment of
equity market value (MVE), calculated considering a twelve-month non-
overlapping period ending four months after the firm’s fiscal year,2 and div-
idend per share paid in this period (DIV), both standardised by market value
of equity at the beginning of the period and net of expected rate of return.
MVE t + DIVt–1,t - MVEt–1

MARt–1,t = – E(Rt–1,t) (5.1)
MVEt–1
Regarding the second issue (i.e. the bank performance measures to be

analysed), we consider the most popular bank performance profit measures
(namely, net income, interest- and intermediation margins), two other
popular profit ratios (namely, ROE and ROA) and two well-known value-
oriented performance measures (namely, residual income and EVA). The
procedure for calculating these measures is described below.
Interest margin (IM) is defined as the difference between a bank’s interest
revenues (IR) and interest costs (IC) obtained over a year.
IM = IR – IC (5.2)
Intermediation margin (IDM) is obtained by subtracting the sum of a bank’s

interest costs (IC) and fee costs (FC) from the sum of a bank’s interest revenues
(IR) and fee revenues (FR) obtained over a year (i.e. equivalent to adding net fee
incomes to interest margin).
IDM = (IR + FR) – (IC + FC) = IM + (FR – FC) (5.3)
Net income is calculated as the difference between all bank income and costs
(including tax and financial expenses) obtained over a year.
NI = (IR + FR) – (IC + FC) = IM + (FR – FC) (5.4)
ROE and ROA are financial ratios obtained by dividing Net Income (NI)
obtained over a year over bank’s Total Equity (TE) and T
NI
ROE = (5.5)
TE
NI
ROA = (5.6)
TA
Residual income (RI) is calculated as the difference between the Net Oper-
ating Profit After Tax (NOPAT) and a charge for invested capital, i.e. measured
as the product of the cost of equity (ke) and the Book Value of Equity (BVE).
RI = NOPAT – Ke * BVE (5.7)

EVA expresses the surplus value created by a company in a given period,

i.e. the firm’s profit net of the cost of all capital. This measure is computed
as the product of the difference between the return on invested capital
(ROIC) and its composite financing cost (i.e. cost of capital – CC) and the
capital invested (CI).
EVA = CI * (ROIC– CC) = (CI* ROIC) – (CI* CC) = NOPAT – (CI* CC) (5.8)
As noted in Velez-Pareja (2000), when EVA is used to assess company per-

formance in a given period, capital invested and NOPAT should not be cal-
culated in the same period. As investors expect to receive returns on the
investment made at the beginning (and not on the cumulative amount at
the end of the period), shareholders compare returns (i.e. NOPAT) earned
over a period (e.g. t-1, t) with the capital invested at the beginning of the
period (t-1). As such, EVA is calculated as follows:
EVAt–1,t = NOPATt–1,t – (CIt–1 * CCt–1,t) (5.9)
One might claim that, while shareholder value is mainly derived from
market value figures, EVA is mainly derived from accounting figures. Book
and market figures mainly differ for at least the following two reasons. First
of all, changes in an asset’s book value differ from changes in its market
value since accounts do not reflect changes of an asset’s market value (e.g.
because the company values the asset at its historic cost). In other words,
book values are conservative. Gains are not recognised until they are
realised, while accounts recognise gains that should properly belong to the
previous period. Secondly, balance sheets do not include any goodwill
(‘goodwill refers to the capacity of the company to generate abnormal
returns in the future, because it represents the expected earnings power of a
collection of assets, combined with a given set of managerial skills’, Barker
(2001: 135)). So if investors change their expectations about the future
profitability of a company, the book value of company’s assets does not
change, while it alters the market price of the company’s share (as well as
goodwill and shareholder value). As a consequence, NOPAT and capital
invested cannot be simply calculated using accounting data, but it is neces-
sary to adjust accounting figures in order to calculate NOPAT and capital
invested on an economic basis. Advocates of EVA3 have identified more
than 160 accounting adjustments, but it is unrealistic even to think of
making all these adjustments for any single company. In the empirical
investigation, we calculate a ‘disclosed EVA’4 by making some standard
adjustments to publicly available accounting data. Two procedures have
been applied to calculate EVA. The first is a standard procedure for non-
financial companies (labelled EVAstd) and the second is a procedure tailored
for banking peculiarities (EVAbkg). By adopting a double set of EVA values,
it is possible to assess the importance of accounting for peculiar characteris-

tics of banking business.
In the standard calculation of EVAstd (e.g. Velez-Pareja, 2000 or
Santorum, 2002), capital invested is estimated using total assets (or the sum
of interest-bearing liabilities and equity capital) and, consequently, meas-
ures the cost of invested capital as the Weighted Average Cost of Capital
(WACC). In order to move the book values closer to their economic values,
two accounting adjustments (standard for all kinds of companies) are
made.5 The first adjustment concerns research and development (R&D)
costs and training costs: these expenses are designed to generate future
growth and, therefore, represent intangible investments. Current assets do
not benefit from these expenses and it would be incorrect to reduce operat-
ing income by the amount of these expenses. However, accounting stan-
dards require companies to treat all outlays for R&D as operating expenses
in the income statement. As a consequence (see equations 5.10 and 5.11),
this accounting distortion can be corrected by: (a) adding back these
expenses in calculating NOPAT; (b) capitalising all R&D expenses and train-
ing costs in capital invested; (c) amortising these capitalised expenses over
an appropriate period: for example, according to Stern Stewart’s statistics,6
five years is the average useful life of R&D expenses. As such, investments
in intangibles are treated in the same manner as investments in tangible
assets. The second adjustment concerns operating lease expenses and these
costs are usually considered as operating costs in companies’ cost-income
statement. However, operating leases are disguised financial expenses since
companies acquire a productive asset (and, therefore, finance their future
production) by paying periodic rent (i.e. operating leases expenses). In
order to face this conservative accounting practice (see equations 5.10 and
5.11), it is opportune to: (d) capitalise any operating lease expenses in cal-
culating NOPAT; (e) treat the present value of expected lease commitments
over time as capital invested in the firm; (f) amortise these capitalised
expenses over an appropriate period.7
As a result, the adjustment we make to NOPAT and Capital invested in
order to calculate EVAstd is obtained as follows:
NOPATstd = EBIT (1 – tax rate) (5.10)

+ R&D expenses (a)
+ Training expenses (a)
+ Operating lease expenses (d)
Capital investedstd = Total assets (5.11)

+ Capitalised R&D expenses (b)
+ Training expenses (b)
– Proxy for amortised R&D expenses (c)
– Proxy for amortised training expenses (c)
+ Proxy for the present value of expected

lease commitments over time (e)
– Proxy for amortised operating lease
commitments (f)
As such, EVAstd is obtained as follows:
EVAstdt–1,t = NOPATt–1,t – (TA t–1 * WACCt–1,t) (5.12)
In addition, we also calculate a second EVA measure (EVAbkg) that takes

account of the specific features of banking business, where capital invested
is not measured using total assets and the cost of invested capital is not
estimated as the Weighted Average Cost of Capital (WACC). Since financial
intermediation is the core business for banks, debts should be considered as
a productive input in banking rather than a financing source (as for other
companies). As such, interest expenses represent the cost for acquiring this
input and, consequently, should be considered as an operating cost rather
than a financial cost (as for other companies). As a consequence, if the
capital charge is calculated following a standard procedure (i.e. applying
WACC on total assets), EVA will be biased since it will double count the
charge on debt. As such, the charge on debt should be firstly subtracted
from NOPAT (the capital charge is calculated on the overall capital – i.e.
equity and debt – invested in the bank and, consequently, it includes the
charge on debt) and, secondly, it would be subtracted from operating pro-
ceeds in calculating NOPAT: interest expenses (i.e. the charge on debt
capital) are in fact subtracted from operating revenues. In the case of banks,
it seems reasonable to calculate the capital invested (and, consequently, the
capital charge) focusing on equity capital and measure the capital invested
in the bank as the book value of shareholder equity.8 Regarding the cost
of capital, the capital charge cannot be obtained applying the bank’s
WACC on the capital invested because the latter is given by the equity
capital and not by the overall capital (debt and equity). Consequently, a
commercial bank’s cost of capital invested should be measured by the
cost of equity.9 In this regard, Sironi (2004: 353) notes that ‘while the cost
of capital for a non-financial company is generally measured as the
weighted average of the costs of different sources of funds and the cost of
equity is just one of the different variables involved, in the case of a
banking firm the cost of equity plays a key role within the global cost of
capital’. Sironi (2004: 355) also notes that ‘with a capital structure exo-
genously determined by regulators, a marginal cost of debt close to that
obtainable from the inter-bank market, and relatively similar to that of
all other major banks, and an array of products that do not need any
debt financing, banks should look at their cost of equity capital as a key
variable’.
The cost of equity (ke) is estimated using the Capital Asset Pricing Model
(CAPM) looking at investors’ expected return. In this framework, there are
three inputs for estimating the cost of equity: (1) Risk Free Rate: following a
standard procedure10 and this is estimated taking the annual rate of return
of a long-term government bond; (2) Equity Risk Premium: the modified
historical approach proposed by Damodaran (1999c) has been applied.11
This is obtained by adding up a country premium to the case premium for
mature equity markets, such as in the US.12 The country premium is
obtained by adjusting13 the country bond spreads: this latter spread has
been obtained by comparing European government bond rates (namely,
the French, Italian, German and UK J. P. Morgan government bond return
indices) with the US J. P. Morgan government bond return indices over the
period analysed (January 1995–June 2003); (3) Beta: these coefficients have
been estimated using daily data on an annual basis by regressing the bank’s
share returns against stock market returns (namely, the CAC 40 for France,
DAX 30 for Germany, MIBTEL storico for Italy and FTSE100 for the UK).
These regression Betas have been successively adjusted following the
Bloomberg procedure.14
In order to move the book values closer to their economic values, we
undertake another five additional adjustments (specific for banks) to
move the book value of banks closer to their economic value. The first
adjustment concerns loan loss provisions and loan loss reserves. The
loan loss reserve is a reserve aiming to cover any future loan losses
and for this reason it should be equal to the net present value of all future
loan losses. In any single period, this reserve is reduced by net charge-offs
(i.e. the current period losses due to credit risk) and replenished by loan
loss provisions (i.e. the provision made in the current period to adjust
the reserves both for pre-existing loans and for estimated future loan
losses related to newly originated loans). This convention is certainly
commendable from a management perspective since it implies that all
loan losses are pre-funded out of current earnings. However, loan loss pro-
visions are commonly used to manage earnings: if a bank achieves high
operating returns, bank managers tend to overestimate this provision,
while they are inclined to underestimate it if operating earnings are poor.
This accounting practice introduces an important distortion in analysing
bank performance since it smoothes earnings. Business is risky, and the
volatility of profits is a manifestation of this risk: for purposes of economic
performance evaluation, smoothing earnings is inappropriate. In order
to face this conservative accounting practice, as shown in equations 5.13
and 5.14, it is opportune to: (g) add back loan loss provisions and deduct
the net charge-offs in calculating NOPAT; (h) capitalise the net loan
loss reserve as capital invested. These adjustments are intended to
reduce the opportunities open to management to smooth accounting
profits.
The second set of adjustments regard taxes. Many banks show significant
and persistent differences between book tax provisions and cash tax pay-
ments. Since these differences are quasi-permanent, deferred taxes should
be considered as capital and, similarly to loan loss provisions, taxes need to
be considered as current period expenses for purposes of economic perfor-
mance evaluation. As shown in equations 5.13 and 5.14, this accounting
conservatism can be faced by: (i) adding back loan book tax provisions and
deducting the cash operating tax; (j) capitalising the deferred tax credits
(net of the deferred tax debits) as capital invested. Similarly to the adjust-
ments for loan loss provisions, taxation adjustments are intended to
reduce the opportunities open to management to smooth accounting
profits.
The third set of adjustments concern restructuring charges. Over the last
decade, many banks have carried out restructuring plans in order to
improve their operating efficiency. To the extent that such restructuring
charges represent disinvestments, these costs should be treated as a capital
reduction rather than costs (and therefore reduce NOPAT). Data availability
limitations do not allow us to evaluate the extent of real disinvestments
due to restructuring charges; these costs are omitted when adjusting
NOPAT and capital invested.
The fourth set of adjustments concern security accounting. In many
countries (such as in the US, Italy, France and the UK), ‘Available for Sale
Securities’ (ASS) are marked to market through the capital account. From an
economic perspective, however, one might claim that selling a security,
with a coupon below (or above) the current market yield, and using the
proceeds to replace it with a current market yield security is a zero-sum
game. In evaluating the economic performance of a bank, it is therefore
more accurate to remove from NOPAT the effect due to gains and losses on
sale of ASS: these gains and losses should be amortised against NOPAT over
the remaining lives of the securities. However, since data on the remaining
lives of the securities are not available and a reasonable assumption cannot
be made, these costs are omitted in our adjustment of NOPAT. Capital
gains and losses generated marking to market ASS (rather than past capital
gains and losses amortised in the period t) are therefore considered as a part
of NOPAT.
The fifth adjustment concerns banks’ general risk reserves. This adjust-
ment aims to correct for distortions derived from the ‘general risk reserve’,
a standard feature in Italian banking. This provision is a reserve that
covers a bank’s future generic loan-losses. In any single period, this reserve
is reduced by net charge-offs (i.e. the current period losses) and replen-
ished by general risk provisions (i.e. the provision made in the current
period to adjust the reserve according to the bank’s risks). Similarly to the
loan-loss reserve, this convention is certainly commendable from a
manage-ment perspective, but it is used in an opportunistic manner. This
accounting practice introduces an important distortion in analysing

banks’ performance since it smoothes earnings. In order to face this con-
servative accounting practice (as shown in equations 5.13 and 5.14), it is
opportune to: (k) add back general risk provisions and deduct the net
charge-offs in calculating NOPAT; (l) capitalise the general risk reserve as
capital invested. These adjustments aim to reduce the opportunities open
to management to smooth accounting profits.
As a result of all the aforementioned adjustments, NOPAT and Capital
invested necessary to calculate EVAbkg are obtained as follows:
NOPATbkg = NOPATstd + (5.13)

+ Loan loss provisions – Net charge-off (g)
+ Book tax provisions – Cash operating tax (i)
+ General risk provisions – Net charge-off (k)
Capital investedbkg = Book value of equity (5.14)

+ Capitalised R&D expenses (b)
+ Training expenses (b)
– Proxy for amortised R&D expenses (c)
– Proxy for amortised training expenses (c)
+ Proxy for the present value of expected (e)
lease commitments over time
– Proxy for amortised operating lease (f)
commitments
+ Net loan loss reserve (h)
+ Deferred tax credits (j)
– Deferred tax debits (j)
+ General Risk Reserve (l)
As such, EVAbkg is obtained as follows:
EVAbkg(t–1,t) = NOPATt–1,t – (CIt–1 * Ket–1,t) (5.15)
The model
This section deals with the methods used in the empirical investigation
to test the value relevance of traditional and innovative performance
measures used in banking. Three main issues are analysed:
(a) methodology for assessing the information content of bank perfor-

mance measures;
(b) selection of the performance measures that are most relevant for
testing;
(c) selection of appropriate variable transformations to achieve robust

results.
Regarding the first issue (a), our aim is to assess the information content
of traditional bank performance measures by distinguishing between the
relative and incremental information content by assuming that equity
markets are efficient in a semi-strong way and forward looking. Regarding
the relative information content, this is assessed looking at the statistical
significance in the following Ordinary-Least-Squares (OLS) regression
model:
FEXt
Dt = b0 + b1 + et (5.16)
Deft–1
where Dt is the Market Adjusted Returns (MAR) for the time t, FEXt is the
unexpected realisation (or forecast error) for a given accounting measure X,
Def t–1 is a variable used to scale FEXt and et is a random disturbance term.
In order to represent the forecast error we follow the approach proposed
in Biddle and Seow (1991), Biddle et al. (1995) and Biddle et al. (1997).
Here the forecast error is expressed as the difference between the realised
value of a performance measure for the time t (Xt) and the market’s expec-
tation for the time t (E(Xt)):
FEXt = Xt – E(Xt) (5.17)
The market expectation of a performance measure is assumed to follow a

discrete linear stochastic process (in autoregressive form):
E(Xt) = ϕ + ψ1Xt–1 + ψ2Xt–2 + ψ3Xt–3 + … (5.18)
where ϕ is a constant and ψ are the autoregressive coefficients.

Rearranging 5.16, we obtain:
Xt – (ϕ + ψ1Xt–1 + ψ2Xt–2 + ψ3Xt–3 + …)
Dt = b0 + b1 + et (5.19)
Deft–1
Xt Xt–1 Xt–2 Xt–3

= b’0 + b’1 + b’2 + b’3 + b’4 + … + et
Deft–1 Deft–1 Deft–1 Deft–1
where Dt is the Market Adjusted Returns for the time t, Xt is the performance
measure at time t , Def t–1 is a variable used to scale Xt and et is a random
disturbance term, where E(b′0) = b0–b1 ϕ; E(b′1) = b1; E(b′i) = bi ψi–1 for i>1.
Equation 5.19 links extra-ordinary returns with lagged performance mea-
sures, encompassing a range of alternative specifications for market expec-
tations such as random walk, ARIMA, constant stock price multiples and
combined ‘levels of changes’ specifications. Although these two models allow

one to consider any order of lags, we limit the model to one lag (following
Biddle et al., 1997) since it is reasonable to assume that abnormal returns should
not be influenced by the value of performance measures older than one year.
With one lag,15 model 5.19 can be simply restated as:
Xt Xt–1
Dt = b′0 + b′1 + b′2 + et (5.20)
Deft–1 Deft–1
Following Biddle et al. (1997), this model is also run by allowing the regres-
sion coefficients to vary in response to positive and negative performances.16
As noted in O’Byrne (1996), the information content of performance measure
can vary according to their signs: for this reason, all performance measures
tested are partitioned in positive and negative values.17
The new equation can be stated as:
Xt,pos Xt,neg Xt–1,pos Xt–1,neg
Dt = b′0 + b′1 + b′2 + b′3 + b′4 + et (5.21)
Deft–1 Deft–1 Deft–1 Deft–1
In order to investigate the incremental information content, our focus is

on EVAbkg and the following one-lag specification is proposed, generalised
to the performance measures composing EVA (detailed in Figure 5.1) scaled
for bank’s MVE calculated four months after the beginning of the fiscal
year.18 This test is undertaken in five steps: firstly, we assess the value rele-
vance of interest margin (INTM), then we introduce in the model the Net
Commission and Fee Income (NetCFI), then we introduce the costs and
incomes necessary to arrive at the accounting NOPAT (CINOPAT ), then we
introduce the Capital Charge and, finally, we run the model introducing
also the Accounting adjustment made to NOPAT and Capital invested to
calculate EVA (AccAdj). The models run are:
INTMt INTMt–1
Dt = b′0 + b′1 + b′2 + et (5.22)
MVEt–1 MVEt–1
INTMt INTMt–1 NetCFIt NetCFIt–1
Dt = b′0 + b′1 + b′2 + b′3 + b′4 + et (5.23)
MVEt–1 MVEt-1 MVEt-1 MVEt–1

Dt = b′0 + b′1 + b′2 + b′3 + b′4
MVEt-1 MVEt–1 MVEt–1 MVEt–1
CINopatt CINopatt–1 (5.24)
+ b′5 + b′6 + et
MVEt–1 MVEt–1

Dt = b′0 +b′1 + b′2 + b′3 + b′4
MVEt–1 MVEt–1 MVEt–1 MVEt–1 (5.25)
CINopatt CINopatt–1 CapChgt CapChgt–1

+ b′5 + b’6 + b′7 + b′8 + et
MVEt–1 MVEt–1 MVEt–1 MVEt–1

Dt = b′0 + b’1 + b′2 + b′3 + b′4
CINopatt CINopatt–1 CapChgt CapChgt–1

+ b′5 + b′6 + b′7 + b′8 (5.26)
AccAdjt AccAdjt–1
+ b′9 + b′8 + et
MVEt–1 MVEt–1
The incremental information content is assessed using the t-tests on individ-

ual coefficients and F-tests of the joint null hypotheses:
H0X: b1 = b2 (or c1 = c2 ; d1 = d2 ; f1 = f2)
H0Y: b3 = b4 (or c3 = c4 ; d3 = d4 ; f3 = f4)
Regarding the choice of the preferred measure of bank performance,

market adjusted returns (calculated considering a twelve-month non-
EVA = Int.M. + Net CFI + CI AcNOPAT – CapChg + AccAdj
Intermediation Margin
Book value of NOPAT
Residual Income
EVA
Where:
Int.M. = Interest Margin
Net CFI = Net Commission and Fee Income
CINOPAT = Costs and incomes necessary to arrive at the accounting NOPAT
CapChg = Capital Charge (Cost of Equity * Book value of Capital)
AccAdj = Accounting adjustment made to NOPAT and Capital invested to calculate
EVA (i.e. EVA – Residual Income)
Figure 5.1: The incremental information content: EVA and its components
overlapping period ending four months after the firm’s fiscal year) are used
as dependent variable since these measures enable us to investigate the
information content of performance measures with regards to a bank’s
ability to create shareholder value and wealth, respectively, over a given
time period. Regarding the selection of the independent variables, the fol-
lowing performance measures in banking are considered: interest margin,
intermediation margin, net income, ROE, ROA residual income, EVAstd and
EVAbkg. Regarding these measures, their information content is assessed in
terms of relative information content, while the investigation of incremen-
tal information content focuses on EVAbkg and, consequently, on the mar-
ginal contribution provided by accounting measures composing this
measure (and shown in Figure 5.1).
The selection of a proper deflator is important in value-relevance studies
to minimise heteroscedasticity and scale effects, especially when the share
market price is used as the dependent variable. Since variables selected have
a different nature, it is necessary to use various deflators (labelled DEF in
equations from 5.16 to 5.21) to properly scale each variable. Namely,
bank’s MVE (four months after the beginning of the fiscal year) is used to
deflate interest margin, intermediation margin, net income, residual
income and EVA: for these variables, the scale effect is likely to bias the
regression R2. In the case of ROE and ROA, these variables do not suffer
from major scale effects as they are ratios and it is not necessary to deflate
them (i.e. DEF = 1). By considering ROE, ROA and net income (standard-
ised by MVE) among the independent variables, we have also the possibil-
ity to assess the impact of using three different deflators on net income: the
MVE, the book value of equity (since ROE is net income divided by book
value of equity) and the book value of total assets (since ROA is net income
divided by book value of assets).
In order to make our results more robust, some other models are esti-
mated. Firstly, the models used to test the relative and incremental infor-
mation content were run using raw market returns (RAWR) as a dependent
variable. This changes the meaning of the relative information content as
follows: (1) when MAR is used as the dependent variable, the relative infor-
mation content of a performance measure refers to its capability to explain
the variation of ‘shareholder value (added)’ created over the period
analysed; (2) when variable RAWR is used as the dependent variable, the
relative information content of a performance refers to its capability to
explain the ‘variation of the created shareholder wealth’ over the period
considered.
Secondly, the following variable transformations are carried out in order
to make the results more robust since previous studies (especially when
abnormal returns are used as the dependent variable) undertaken in the
value relevance literature have been often characterised by low explanatory
power (R2): (1) equations from 5.20 and 5.21 are estimated by taking the
natural log of the dependent variable (except ROE and ROA) rather than
deflating the data. Regression estimates may be interpreted as the strength
of the relationship between a percentage variation of a performance mea-
sure (independent variable) and the shareholder value (added) (dependent
variable); (2) independent variables are taken in terms of rate of change.
From a theoretical point of view, one might observe that a low R2 is due to
the fact that the dependent variables express flows, percentages and resid-
ual values (i.e. shareholder rate of returns net of the investment cost oppor-
tunity), while the independent variables express simply flow measures (i.e. the
level of net income/residual income/EVA/interest margin/intermediation
margin generated over a period). From the point of view of investors,
market share prices (and therefore investor expectations) are influenced by
the percentage variation of accounting performance measures, rather than
the absolute value of the measure. As such, the ability of the absolute value
of a performance measure of explaining variation in shareholder value
(added) is likely to be low. To face this problem, the independent variable
is transformed to capture the rate of variation of a performance measure
over the period (see Figure 5.2). These models assess the relationship
between percentage changes in a performance measure, rather than the
absolute value of the performance measure, and the shareholder value
(added).
Static variable Transformed dynamic variable (rate of change)
net incomet –net incomet–1

Net Incomet (NIt)
net incomet
ROEt – ROEt–1
Return on Equityt (ROEt)
ROEt
ROA t – ROA t–1

Return on Assetst (ROA t)
ROA t
residual incomet – residual incomet–1

Residual Incomet (RIt)
residual incomet
EVA t – EVA t–1

Economic Value Added t (EVA t)
EVA t
interest margint – interest margint–1

Interest Margint (INTMt)
interest margint
intermediation margin t – intermediation margin t–1

Intermediation Margint (IMt)
intermediation margint
Figure 5.2: Independent variables: flow and stock values

Sample description
Data employed were collected from two main sources. Financial statement
information was obtained from the Bankscope database and capital markets
information from Datastream. It is worthwhile noting that some data
(namely, share prices, risk free rates and equity market indices) are directly
taken from this latter source, while other data (namely, Beta indices,
bond and equity spreads) have to be calculated using the Datastream
data.
The sample adopted to assess the value-relevance analysis of the perfor-
mance measures considers only publicly listed banks since this analysis
requires rates of return from equity markets. The sample of listed banks
adopted comprises seventy-one banks (i.e. twenty French banks, thirteen
German Banks, twenty-eight Italian banks and ten UK banks) over a period
of seven years (from 1 January 1996 to 31 December 2002). One might
claim that equity markets were marked by a strongly positive trend in share
prices in the late 1990s and a strong reduction in the early 2000s and this
may influence the results. Although this is certainly true, it is worthwhile
noting that we attempted to minimise the effect caused by the volatility in
stock prices over this period by: (1) increasing the time period: observations
in the sample over a period of seven years (1996–2002); and (2) inserting
dummy variables controlling for time effects. In addition, in order to avoid
the undue effect of unrepresentative data (i.e. influential observations),
firm observations more than two standard deviations from the mean
MAR were deleted. As result, the sample of listed banks adopted is an
unbalanced panel data of 404 year-observations (172 for Italy, 121 for
France, 60 for Germany and 51 for the UK). Table 5.1 reports some
figures describing the sample. In addition we also found that the
correlation between different performance measures was generally
positive, although these were statistically significant in only a few
cases.
Once we identify the bank performance measure with the highest value-
relevance, we undertake an empirical investigation to measure shareholder
value created in European banking focusing on both listed and non-listed
European banks. As such, we select a second sample including both pub-
licly listed and non-listed banks. The decision to include unlisted banks in
the sample creates some methodological problems (such as the estimation
of the cost of equity), but this enables us to consider a larger sample and,
consequently, achieve results of wider interest to the banking industry. For
example, by considering unlisted banks, it is possible to assess the ability of
small and regional banks (usually not assessed in previous studies) to create
shareholder value (added) and enables a comparison with larger banks.
Table 5.2 reports the composition of the sample, by distinguishing among
countries and bank ownership types.
Table 5.1: Descriptive statistics – sample of European listed banks 1996–2002
Panel A General information*
N Minimum Maximum Mean Std. Deviation
Total Deposits 404 62.50 548371.00 46205.30 88266.47

Total Loans 404 19.00 352321.00 34224.37 58434.12
Net Income 404 –66.90 7537.00 386.52 1014.31
Market Raw Returns 404 –71.06% 216.46% 2.76% 41.35%
Total Assets 404 81.10 825288.00 72221.00 135706.92
* Data for MAR is expressed in percentage. All other items are in Euro million.
Panel B Variables used for assessing the relative information content*
Market Adjusted Returns 404 –71.06% 216.46% 2.76% 41.35%

Interest Margin 404 2.10 15738.00 1192.99 2200.541
Intermediation Margin 404 6.90 23562.00 1826.86 3406.79
ROE 404 –11.85% 37.53% 9.31% 6.12%
ROA 404 –0.25% 17.03% 0.73% 1.18%
Net Income 404 –66.90 7537.00 386.52 1014.31
Residual Income 404 1681.67 2985.00 –49.45 414.77
EVAbkg 404 –1837.40 4102.00 4.661 454.08
EVAstd 404 –1409.92 –.032 –66.48 162.55
* Data for MAR, ROE and ROA are expressed in percentage. All other items are in Euro million.
161
162
Table 5.1: Descriptive statistics – sample of European listed banks 1996–2002 – continued
Panel C Variables used for assessing the incremental information content*
Market Adjusted Returns 405 –71.06% 216.46% 2.76% 41.35%

Interest Margin 405 2.10 15738.00 1192.99 2200.541
Net Commission and Fee Incomes 405 –11.00 7824.00 619.81 1232.87
Costs and Incomes necessary to
calculate NOPAT 405 –16446.00 60.00 –1399.80 2494.76
Accounting Adjustment for EVAbkg 405 –2605.00 2246.00 42.97 391.44
Accounting Adjustment for EVAstd 405 –49525.63 2805.98 –188.2936 2679.70
Capital Charge for EVAbkg 405 –9446.00 –0.23 –484.78 1127.29
Capital Charge for EVAstd 405 –118763.66 –0.77 –9102.48 17375.89
* Data for MAR is expressed in percentage. All other items are in Euro million.
Source: Bankscope and Datastream.

Table 5.2: The number of listed and non-listed banks in the sample by bank
ownership
Panel A France 1995 1996 1997 1998 1999 2000 2001 2002
Commercial Banks 167 186 193 196 211 212 225 218
Cooperative Banks 80 89 92 94 98 100 112 106
Savings Banks 22 22 25 25 25 29 31 29
Total 297 310 315 334 341 368 353 297
Panel B Germany 1995 1996 1997 1998 1999 2000 2001 2002
Savings Banks 514 549 569 588 601 609 615 615
Total 1229 1407 1561 1673 1892 2030 2202 2219
Panel C Italy 1995 1996 1997 1998 1999 2000 2001 2002

Savings Banks 62 63 63 65 66 66 66 66
Total 229 269 300 464 600 692 737 754
Panel D United Kingdom1995 1996 1997 1998 1999 2000 2001 2002

Savings Banks 0 0 0 0 0 0 0 0
Total 57 65 74 86 91 97 101 100
Information content of bank performance measures

This section presents our findings that test for the relative and incremental
information content of bank performance measures for listed European
commercial banks. Table 5.3 reports the results of the relative information
content of the respective performance measures. The table reports the
results from eight regression estimates where MAR (dependent variable) is
regressed on one other performance measure and the same performance
measure with a one-year lag. Results are ordered by the adjusted R2s from
the highest (on the left) to the smallest (on the right). Information supplied
includes the coefficient estimates and their statistical significance (i.e. based
on the t-statistic), the adjusted R-square (i.e. the ability of shareholder
value drivers in explaining variation of bank shareholder value), the
Durbin-Watson test (i.e. a test for assessing if residuals for consecutive
164
Table 5.3: The relative information content performance measures of European listed banks over the period 1996–2002
Dependent variable: Market-Adjusted Returns (MAR)
Net Residual Interest Intermediation

EVAstd income income ROE ROA margin margin EVAbkg
T t–1 T t–1 t t–1 T t–1 t t–1 T t–1 t t–1 T t–1

† ‡ ‡
Estimated 0.091 –0.076 0.059* –0.063 0.076* –0.044 0.093 –0.064 0.078 –0.112 –0.040 –0.071 –0.075 –0.051 0.240 –0.145‡
Coefficients
Adj. R2 0.302 0.305 0.306 0.310 0.311 0.312 0.314 0.389
p-value* 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.00
DW 2.160 2.164 2.165 2,157 2.188 2,156 2,170 2,126
The p-value reported, based on the F-test, expresses the probability to make an error rejecting the null hypothesis (i.e. all slope coefficients are equal to zero). As such, a p-value
close to 0 signals that the performance measure under investigation is likely to have a statistically significant impact on MAR since at least one of the estimated regression
coefficients differs from zero. Broadly speaking, a p-value of 5% expresses that there is a probability of 5% that the performance measure investigated does not have a statistically
significant impact on MAR, and so on.
*/†/‡ indicate statistical significance at p<10%, p<5%, and p<1%, respectively

observations are uncorrelated) and the p-values from a two-tailed statistical

test (based on the F-statistic) expressing the probability to make an error
rejecting the hypothesis that the performance measure under investigation
does not have a statistically significant impact on MAR. As such, a p-value
close to 0 signals that it is very likely that the performance measure investi-
gated has a statistically significant impact on MAR; a p-value of 5 per cent
would express that there is a probability of 5 per cent that the performance
measure investigated does not have a statistically significant impact on
MAR, etc.
Tests reported in Table 5.4 investigate the marginal information content
provided by a performance measure to the information content of another
performance measure. By running these tests for all couples of performance
indicators, we assess the statistical significance of the F-change due to the
introduction of a second performance measure in a regression assessing
market adjusted returns (dependent variable) and another performance
measure. The null hypothesis is that the second performance measure does
not provide any incremental information content. This test enables us to
test the information content of each couple of competing sets of indepen-
dent variables to explain variations in the dependent variable.
According to the results from Table 5.3, we found that adjusted R2 range
between 0.302 and 0.389. The highest adjusted R2 is obtained by EVAbkg
while the adjusted R2s for the other performance measures are very close
(the gap between the second highest and the lowest values is 1.2 per cent),
the difference between the EVAbkg adjusted R2 and the second highest value
is 7 per cent. Since the p-value* equals zero, we reject the null hypothesis
that all performance measures have the same information content. The
analysis of p-values+ from two-tailed statistical tests also enables us to reject
the null hypothesis that EVAbkg has the same relative association of any
of the other performance measures. Regarding estimated regression co-
efficients, these are found to be positive in EVAbkg measured at the time
period (t), while coefficients of the lagged term (t–1) are negative. This
result is common to most of the other performance measures: it may be
explained by the fact that when a bank achieves a good performance (e.g. a
better EVA or ROE or Residual Income, etc.) in the current period (t),
investors judge these results positively and will probably purchase the
bank’s shares, therefore increasing MAR. In contrast, if the bank achieves a
positive performance in the previous time period (t–1), investors may judge
this negatively since they may believe that the bank will find it difficult to
improve on its performance in the current period (t): in this case, investors
will not buy the bank’s shares resulting in unaffected or lower MAR. All
EVAbkg coefficients (both at time t and t–1) obtain high statistical
significance. Regarding the other performance measures, these seem to
have a similar relative association with MAR. Estimated regression
coefficients are often (such as net income, ROA and ROE) statistically
166
Table 5.4: P-value of joint-F test among all couples of bank performance measures analysed
Second variable Net Residual Interest Intermediation

First variable EVAstd income income ROE ROA margin margin EVAbkg
EVAstd – 0.667 0.676 0.663 0.623 0.594 0.552 0.678

Net income 0.258 – 0.002 0.700 0.855 0.069 0.278 0.982
Residual income 0.204 0.001 – 0.986 0.824 0.055 0.170 0.764
ROE 0.073 0.700 0.362 – 0.295 0.024 0.179 0.600
ROA 0.052 0.032 0.230 0.224 – 0.023 0.161 0.260
Interest margin 0.031 0.009 0.010 0.011 0.014 – 0.200 0.025
Intermediation margin 0.020 0.026 0.020 0.058 0.069 0.138 – 0.106
EVAbkg 0.000 0.000 0.000 0.000 0.000 0.000 0.000 –
This table reports the p-value reported of a joint-F test among all couples of performance measures analysed. Based on the change of F-statistic tests due
to the introduction of a second set of independent variables (i.e. a performance measure for times t and t-1), the p-values reported express the
probability to make an error rejecting the null hypothesis that the second set of variables do not provide a statistically significant contribution to the
explanatory power of the first set of variables.
significant in time t, but not in the previous period. EVAstd is found to have
a poorer relative association with market adjusted returns and regression
coefficient estimates are never found to be statistically significant. In all
models estimated, we reject the null hypothesis that all performance mea-
sures have the same information content (p-value* equals zero).19
When regression coefficients are allowed to vary (as stated in equation
5.20) in response to positive and negative values of performance measures
(Table 5.5), adjusted R2s are found to range between 0.293 and 0.435. The
highest adjusted R2 is always obtained by EVAbkg: while the adjusted R2s for
the other performance measures are very close (the gap between the second
highest and the lowest value is about 1.6 per cent), the difference between
the EVAbkg adjusted R2 and the second highest value is about 9 per cent.
Regarding the estimated coefficients, these are found to be highly statisti-
cally significant for EVAbkg and statistically significant for ROE and net
income. In all models run, we reject the null hypothesis that all perfor-
mance measures have the same information content.
As such, our results suggest that EVAbkg outperforms all other perfor-
mance measures since it has the greatest ability to explain variation of
market adjusted returns (expressing shareholder value created over the
period analysed). In contrast, EVAstd is outperformed by all other perfor-
mance measures since it has generally the lowest adjusted R2: this means
that it has the lowest ability to explain variation of market adjusted
returns. These results provide evidence that it is necessary to accurately
consider the peculiar nature of capital in banking and the accounting dis-
tortions. The other performance measures analysed are found to have a
substantially equivalent relative information content: their adjusted R2
(expressing the relative association with MAR) are very close and p-values
for the comparison of couples of performance indicators enable us to reject
in a few cases the null hypothesis that two measures have the same infor-
mation content.
These results provide some useful insights into the value-relevance litera-
ture. EVAbkg is found to outperform all other measures independently from
the measure adopted to express shareholder value (added). In the previous
literature, EVA was found to have the highest relative information content
when shareholder value was measured by MVA, while net income was
usually found to have a non-inferior relative association to market adjusted
returns than that of EVA. Our results can probably be explained by the fact
that EVAbkg is obtained using a specific procedure accounting for the pecu-
liar features of commercial banks.
Regarding the incremental information content, our focus is on EVAbkg
since this measure shows the highest relative association with MAR.
Table 5.6 reports the results of the incremental information content of the
respective performance measures. We report the estimated coefficients and
their statistical significance, the adjusted R-squared of all models run, the
168
Table 5.5: The relative information content performance measures of European listed banks over the period 1996–2002: coefficient of
positive and negative values of each performance measure analysed allowed to change
Residual income Net income ROA ROE EVAbkg
+ – + – + + + – – –
† † † † † † ‡
T –0.073 0.097 –0.116 0.100 –0.028 0.087 0.132 0.082 0.190 0.261‡
Estimated
coefficients
t–1 –0.033 0.006 0.060‡ 0.003 –0.047 N/A 0.217‡ N/A 0.015 –0.151‡
Adj R2 0.308 0.310 0.312 0.326 0.435

p-value* 0.00 0.00 0.00 0.00 0.00
DW 2.191 2.191 2.185 2.201 2.104
The p-value reported, based on the F-test, expresses the probability to make an error rejecting the null hypothesis (i.e. all slope coefficients are equal to
zero). As such, a p-value close to 0 signals that the performance measure investigated is likely to have a statistically significant impact on MAR since at
least one of the estimated regression coefficients differs from zero. Broadly speaking, a p-value of 5% would express that there is a probability of 5%
that the performance measure investigated does not have a statistically significant impact on MAR, and so on.

Table 5.6: The incremental information content of EVAbkg in European banking between 1996 and 2002 (Dependent variable: Market-
Adjusted Returns – MAR)
Panel A – Estimated Regression Coefficients
Cost & income Accounting

Net commision necessary for Capital adjustment
Interest & fee income pass to NOPAT charge to calculate
margin (NetCFI) (CINOPAT) (CapChg) EVAbkg (AccAdj)
T t–1 T t–1 t t–1 t t–1 t t–1
Model 5.22 –0.040 –0.071

Model 5.23 –0.018 –0.082 –0.042 –0.051
Model 5.24 –0.255 0.175 0.413 –0.053 0.625† –0.219
Model 5.25 –0.354* 0.143 1.022‡ –0.585 1.329‡ –0.788† 0.421† –0.599‡
Model 5.26 –0.245 0.141 1.243‡ –0.260 1.895‡ –0.254 0.391† –0.539‡ 1.023‡ 0.811‡
Panel B – Regression statistics
Adjusted R2 Adjusted R2 change F-change* DW
Model 5.22 0.338 – – 2.156

Model 5.23 0.344 0.006 1.993* 2.168
Model 5.24 0.354 0.01 2.902† 2.128
Model 5.25 0.369 0.015 4.375‡ 2.138
‡
Model 5.26 0.500 0.131 50.171 2.021
169

170
Table 5.6: The incremental information content of EVAbkg in European banking between 1996 and 2002 (Dependent variable: Market-
Adjusted Returns – MAR) – continued
Panel C – Models run
INTMt INTMt–1
(5.22) Dt = b0 + b1 + b2 + et
MVEt–1 MVEt–1
(5.23) Dt = b0 + b1 + b2 + b3 + b4 + et
INTMt INTMt–1 NetCFIt NetCFIt–1 CINopatt CINopatt–1
(5.24) Dt = b0 + b1 + b2 + b3 + b4 + b5 + b6 + et
MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1
INTMt INTMt–1 NetCFIt NetCFIt–1 CINopatt CINopatt–1 CapChgt CapChgt–1
(5.25) Dt = b0 + b1 + b2 + b3 + b4 + b5 + b6 + b7 + b8 + et
MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1
INTMt INTMt–1 NetCFIt NetCFIt–1 CINopatt CINopatt–1 CapChgt CapChgt–1
(5.26) Dt = b0 + b1 + b2 + b3 + b4 + b5 + b6 + b7 + b8 +
MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1 MVEt–1
AccAdjt AccAdjt–1
b9 + b8 + et
MVEt–1 MVEt–1
R-squared changes, F change and its statistical significance, the Durbin-

Watson test of the most complete model. The analysis of the statistical sig-
nificance of the estimated regression coefficients provides evidence that
EVAbkg components are statistically significant. In detail, most EVAbkg compo-
nents (i.e. net commission and fee income (Net CFI), costs and incomes nec-
essary to calculate NOPAT (CINopat), capital charge (CAPCHG) and accounting
adjustments (ACADJbkg) in model five reported in Table 5.6) are found to be
meaningful predictors of MAR. However, the incremental contribution pro-
vided in explaining market adjusted returns by these EVAbkg components is
different.
According to our results, EVAbkg accounting adjustments are found to
provide the largest F statistic change, both its estimated regression coefficients
are statistically significant at 1 per cent and this component provides the
largest increase of adjusted R2s (13.1 per cent). Capital charge (CAPCHG) and
costs-incomes necessary to calculate NOPAT (CINopat) have a lower contribu-
tion: coefficient estimates are statistically significant, adj. R2 improved by
1.4 per cent and 1.0 per cent (respectively) and F-statistic changes are found to
be statistically significant at the 1 per cent and 5 per cent levels (respectively).
According to these results, EVAbkg seems to achieve the highest relative
information content (among all performance indicators investigated) in
being related to shareholder value created over a period and this is due to
the accounting adjustments made in calculating this measure. The other
components seem to provide a minor contribution to explaining MAR. As
such our findings show similar results for residual income, net income and
interest margin. For example, net income and residual income have similar
information content since the incremental contribution provided by ‘costs
and incomes necessary to calculate NOPAT’ is marginal and, consequently,
net income and residual income all have similar relative information
content.
We conclude from the above that in the case of listed European banks,
EVAbkg is the preferred measure of shareholder value added as it best
explains Market-Adjusted Returns (MAR) for banks in France, Germany,
Italy and the United Kingdom between 1996 and 2002. In addition,
accounting adjustments for the calculation of the appropriate EVA measure
can have a major influence in explaining market-adjusted returns.
Measuring shareholder value in Europe – listed and non-listed

banks
This section examines created shareholder value for listed and non-listed
banks in France, Germany, Italy and the United Kingdom over the period
1995–2002. As we have identified EVAbkg as our preferred measure of share-
holder value creation, the following provides estimates of EVA created in
the aforementioned banking systems.
Considering all bank categories and all four countries investigated and
shown in Table 5.7, we found that European banks generate, on average,
substantial and widespread profits since the mean level of published net
income is Euro 11.62 million and the percentage of banks generating posi-
tive annual net income is 94.15 per cent. Looking at the annual mean
levels of published net income (see Table 5.8), we found that these
increased from Euro 8.52 million in 1995 to Euro 12.82 million in 2002
and at least 92.8 per cent of banks in our sample generated annually a
positive net income.
In assessing shareholder value created, we considered bank performance
from an economic point of view and net of opportunity cost of equity
invested capital in the bank.20 We found that, on average, European banks
do not seem to have created shareholder value over the period analysed.
Considering all bank categories and all four countries investigated (Table
5.7), the mean levels of EVAbkg created over the period is Euro –4.50 million
and the percentage of banks generating positive annual EVAbkg is only
39.22 per cent. Looking at the range of mean EVAbkg annual levels between
1995 and 2002 (Table 5.8), we find constantly negative levels (except in
1999) ranging from Euro –7.17 million in 1995 to Euro 0.37 million in
1999, the ratio between created EVAbkg on invested capital is positive only
in 1997 (ranging from –1.74 per cent in 2001 and 0.40 per cent in 1999)
and the proportion of banks that created value (with a positive EVAbkg)
Table 5.7: Overall mean profits and EVAbkg created over the period 1995–2002
Country Mean St.dev %Positive

net income*
France 33.94 192.13 84.70%

Published
Germany 2.91 32.17 96.06%

Italy 14.32 55.00 94.19%
UK 83.83 221.54 91.60%
Mean 11.62 62.63 94.15%
France –16.57 221.69 50.99%

EVAbkg*
Germany –1.76 29.03 37.23%

Italy –4.63 59.94 39.55%
UK –12.91 131.74 32.68%
Mean –4.50 62.13 39.22%
France 0.23% 13.61% 50.99%

EVAbkg on
invested
capital+
Germany –0.52% 4.74% 37.23%

Italy –1.02% 6.78% 39.55%
UK –2.69% 11.69% 32.68%
Mean –0.59% 6.45% 39.22%
*+ Values in Euro million and in percentage, respectively

Table 5.8: Annual mean levels of EVAbkg between 1995 and 2002
1995 1996 1997 1998
% % % %
Mean St.dev Positive Mean St.dev Positive Mean St.dev Positive Mean St.dev Positive
France Published Net income* 17.61 76.61 80.13% 20.91 93.86 0.86 17.26 168.02 84.42% 26.86 157.32 87.72%
EVAbkg* –14.43 100.66 39.36% –13.43 91.67 43.72% –11.47 154.41 49.27% –7.44 120.66 46.95%
EVAbkg /Inv. Capital + –0.68% 11.26% 39.36% –1.40% 12.62% 43.72% 0.82% 11.76% 49.27% –0.69% 9.93% 46.95%
Germany Published Net income* 4.31 12.33 97.88% 4.13 13.05 97.36% 4.24 14.23 97.62% 5.20 21.60 97.16%
EVAbkg* –1.00 11.00 32.45% –0.58 5.60 57.12% –0.45 7.43 57.92% –0.58 20.28 38.21%
EVAbkg /Inv. Capital + –0.64% 4.42% 32.45% 0.74% 4.10% 57.12% 0.76% 3.34% 57.92% –0.17% 5.02% 38.21%
Italy Published Net income* 7.26 43.64 91.77% 4.89 74.88 95.20% 3.02 90.61 95.70% 14.45 39.14 94.84%
EVAbkg* –34.89 89.88 12.12% –26.39 71.73 13.28% –18.58 81.58 37.42% –6.03 54.21 42.58%
EVAbkg /Inv. Capital + –7.25% 6.17% 12.12% –4.87% 5.94% 13.28% –1.47% 5.90% 37.42% –0.13% 6.12% 42.58%
United Published Net income* 58.41 139.31 95.00% 63.81 176.68 97.06% 65.50 163.78 96.21% 60.70 150.73 90.00%
Kingdom EVAbkg* 5.88 112.47 35.00% 8.17 95.29 50.00% 5.81 66.54 39.88% –14.31 105.12 31.11%
Total Published Net income* 8.52 30.40 94.29% 8.65 38.40 95.42% 8.03 51.34 95.45% 11.60 46.82 95.27%
EVAbkg* –7.17 38.20 30.98% –5.57 29.82 49.14% –4.21 40.17 53.34% –2.93 42.33 39.88%
EVAbkg /Inv. Capital + –1.55% 5.87% 30.98% –0.37% 5.73% 49.14% 0.40% 5.20% 53.34% –0.39% 6.33% 39.88%
173
174
Table 5.8: Annual mean levels of EVAbkg between 1995 and 2002 – continued
1999 2000 2001 2002
% % % %
Mean St.dev Positive Mean St.dev Positive Mean St.dev Positive Mean St.dev Positive
France Published Net income* 29.25 166.47 79.61% 41.69 236.06 85.07% 53.22 249.72 86.89% 53.86 329.04 86.60%
EVAbkg * –1.23 83.04 47.94% –30.96 292.61 54.16% –26.98 351.08 57.57% –22.78 488.34 63.39%
EVAbkg /Inv. Capital + 0.12% 14.94% 47.94% 0.36% 14.94% 54.16% 0.85% 17.72% 57.57% 1.85% 14.25% 63.39%
Germany Published Net income* 4.29 25.80 95.92% 3.21 23.05 96.65% 0.13 83.84 93.74% –0.01 38.42 94.14%
EVAbkg * –0.84 22.59 33.05% –2.40 45.42 29.13% –2.43 41.06 20.47% –4.27 54.30 39.57%
Italy Published Net income* 13.52 38.38 95.51% 16.67 51.94 93.51% 19.08 57.29 94.06% 18.09 60.63 93.27%
EVAbkg * 3.21 44.35 63.06% 7.39 43.41 46.03% –2.98 63.81 31.78% –0.01 65.16 39.34%
EVAbkg /Inv. Capital + 2.05% 5.89% 63.06% 0.21% 7.13% 46.03% –1.77% 6.76% 31.78% –0.93% 8.43% 39.34%
United Published Net income* 94.10 233.35 89.60% 91.22 259.68 94.17% 108.64 295.95 91.59% 100.03 273.60 83.96%
Kingdom EVAbkg * 9.53 123.81 30.98% –33.34 139.71 25.24% –29.40 176.70 28.97% –39.22 192.68 29.25%
Total Published Net income* 12.82 53.41 93.67% 13.30 60.88 94.59% 13.57 103.45 92.97% 12.82 81.81 92.83%
EVAbkg * 0.37 38.28 40.80% –4.48 75.78 35.49% –6.16 85.17 27.36% –6.40 107.69 41.77%
*+ Values in Euro million and in percentage, respectively

ranged from 30.98 per cent of the sample in 1995 to 49.14 per cent in
1996.
However, the situation described above is not common to all countries.
UK banks appear to be the most profitable banks since these generate over
the period 1995–2002 the highest mean annual level of net income – Euro
83.33 million (from Euro 58.19 million in 1995 to Euro 100.26 million in
2002) – and the percentage of banks with a positive net income is 91.60 per
cent (Tables 5.7 and 5.8). We also measured bank profitability using tradi-
tional profitability ratios, such as ROE and ROA. On average, UK banks
achieved between 1995 and 2002 average levels of ROE around 10 per cent
(from 9.11 per cent in 2002 and 13.11 per cent in 1997) and a ROA of
around 1 per cent (from 0.85 per cent in 1998 to 1.7 per cent in 2000).
Focusing on shareholder value created, we found positive mean EVAbkg
levels from 1995 to 1998 and in 1999, and negative mean EVAbkg levels in
1998 and from 2000 to 2002. We found an even worse situation analysing
the ratio of EVAbkg on capital invested (since the average ratio is constantly
negative in every year between 1995 and 2002) and the number of UK
banks with a positive EVAbkg (around 33 per cent, ranging from 25 per cent
in 2000 and 50 per cent in 1996). Our findings show that almost all UK
banks generally generate higher profits than banks in the other countries,
but not all UK banks are able to achieve profits superior to the cost oppor-
tunity of equity capital (see Figure 5.3). For example, we found that several
banks achieved a positive net income in 2002, but lower than their capital
charge, and so they destroyed value for their shareholders. Table 5.9 reports
the list of the thirty UK banks that generated the largest EVAbkg in 2002.
We find that these banks, on average, generated an EVAbkg of 12.61 per cent
of the capital invested, with net income 21.43 per cent of equity capital
(ROE) and 2.12 per cent of total assets (ROA). In 2002, twelve of the top
thirty EVAbkg creators in the UK are in the top twenty largest banks. Among
these twenty largest UK banks, we found that seven banks destroyed share-
holder value. On average, the top twenty banks created shareholder value
of about 7.5 per cent of the capital invested and generate large profits
(namely, the average of these banks was the ROE 15.38 per cent and ROA
0.78 per cent). UK banks seem (on average) to be able to generate higher
value added for their shareholders than banks in the other three countries
analysed.
In France, we found that banks appear to generate substantially lower
mean profits than in the UK, but superior to those in Italy and Germany.
Considering all bank categories, French banks generated increasing annual
mean profit from Euro 17.61 million in 1995 to Euro 60.70 million in 2002
and the mean value over the period 1995–2002 is 33.94 per cent Euro
million (Tables 5.7 and 5.8). French banks registered constantly positive
mean levels of ROE and ROA in this period. ROE ranged between 3.3 per
cent in 1996 and 6.59 per cent in 2000 and ROA is constantly around 0.4
Panel A ROA and EVA bkg on capital invested
Commercial Banks (EVA/CAPITAL INVESTED) MEAN (CAPITAL INVESTED)

Savings Banks (ROA) Savings Banks (EVA/CAPITAL INVESTED)
Commercial Banks (ROA) MEAN (ROA)
15,00%
10,00%
ROE and EVA/Capital Invested
5,00%
0,00%
-5,00%
-10,00%
1995 1996 1997 1998 1999 2000 2001 2002
Panel B Percentage of banks with a positive ROA and EVA bkg on capital invested
Commercial Banks (EVA/CAPITAL INVESTED) MEAN (CAPITAL INVESTED)

Savings Banks (ROA) Savings Banks (EVA/CAPITAL INVESTED)
Commercial Banks (ROA) MEAN (ROA)
Percentage of bank with positive Net income and EVA
120%
100%
80%
60%
40%
20%
0%
1995 1996 1997 1998 1999 2000 2001 2002
Figure 5.3: Return on equity and EVAbkg on capital invested in UK banking between
1995 and 2002
Table 5.9: Top thirty EVAbkg creators in UK banking in 2002
EVA on
Capital
Economic Capital invested+ ROE+ ROA+
Rank Bank name Bank Specialisation NOPAT* charge* EVAbkg* (%) (%) (%)
1 Royal Bank of Scotland Plc Commercial Bank 6897 4255 2642 62.08 16.25 1.02
2 Barclays Bank Plc Commercial Bank 5861 3727 2134 57.25 15.03 0.60
3 National Westminster Bank Plc – NatWest Commercial Bank 3723 1717 2006 16.67 21.42 1.09
4 HSBC Bank Plc Commercial Bank 3151 2446 705 28.82 8.73 0.71
5 Yorkshire Bank Plc Commercial Bank 263 103 160 19.25 41.82 3.09
6 Bank of Scotland Commercial Bank 1041 891 150 1.74 11.53 0.54
7 Standard Chartered Plc Commercial Bank 1391 1258 133 1.50 18.98 14.20
8 MBNA Europe Bank Ltd Commercial Bank 183 80 103 15.94 19.11 3.44
9 HFC Bank Plc Commercial Bank 208 108 100 11.46 14.60 1.78
10 Cheltenham & Gloucester Plc Commercial Bank 355 271 84 3.83 17.02 0.62
11 Northern Bank Limited Commercial Bank 111 35 76 23.55 47.86 2.87
12 Clydesdale Bank Plc Commercial Bank 154 101 53 6.59 27.49 1.62
13 Woolwich Plc Commercial Bank 298 260 39 1.84 22.14 0.80
14 FCE Bank Plc Commercial Bank 264 237 26 1.38 10.12 0.80
15 Lloyds Bank Plc Commercial Bank 1238 1217 21 0.25 24.13 1.21
16 HBOS Treasury Services Plc Commercial Bank 73 53 20 4.75 26.01 0.12
17 TSB Bank Plc Savings Bank 239 227 12 0.66 17.76 1.30
18 Lloyds TSB Scotland Plc Savings Bank 72 61 11 2.57 16.92 1.56
19 Girobank Plc Commercial Bank 53 46 8 2.06 20.20 1.63
20 Wagon Finance Ltd Commercial Bank 14 7 7 11.91 22.46 2.20
21 Gresham Trust Plc Commercial Bank 18 13 5 4.73 17.08 13.79
22 Amalgamated Finance Ltd Commercial Bank 5 1 4 65.82 78.46 1.99
23 Coutts & Co Commercial Bank 30 26 4 2.01 16.26 0.41
24 Alliance & Leicester Group Treasury Plc Commercial Bank 32 28 4 1.42 14.20 0.13
25 British Linen Bank Ltd Commercial Bank 15 12 3 3.46 8.05 0.42
177
178
Table 5.9: Top thirty EVAbkg creators in UK banking in 2002 – continued
EVA on
Capital
26 Bank of Wales Plc Commercial Bank 9 6 3 5.94 23.42 1.59

27 Bristol & West International Commercial Bank 4 4 1 3.40 15.77 0.57
28 Granville Bank Ltd Commercial Bank 2 1 1 8.68 12.63 2.11
29 Woolwich Guernsey Commercial Bank 11 10 1 0.62 13.01 0.43
30 Reliance Bank Ltd Commercial Bank 1 1 1 8.11 24.44 0.91
* + Values in Euro million and in percentage, respectively

179
Panel A ROA and EVA bkg on capital invested
Commercial Banks (EVA/CAPITAL INVESTED) Commercial Banks (ROA)

Cooperative Banks (ROA) Savings Banks (EVA/CAPITAL INVESTED)
MEAN (CAPITAL INVESTED) MEAN (ROA)
Cooperative Banks (EVA/CAPITAL INVESTED) Savings Banks (ROA)
10,00%
8,00%
ROE and EVA/Capital invested
6,00%
4,00%
2,00%
0,00%
-2,00%
-4,00%
1995 1996 1997 1998 1999 2000 2001 2002
Panel B Percentage of banks with a positive ROA and EVA bkg on capital invested
120%
Percentage of banks with positive
100%
net income and EVA
80%
60%
40%
20%
0%
1995 1996 1997 1998 1999 2000 2001 2002
Figure 5.4: Return on equity and EVAbkg on capital invested in French banking
per cent in the period 1995–2000, while this sharply increased up to 0.65
per cent in 2002. We found that the percentage of banks with positive net
income is around 80–90 per cent that is lower than in the UK. The analysis
of profitability ratios shows that French cooperative banks are, on average,
more able to remunerate equity capital with profits, while cooperative and
commercial banks have similar ROA mean levels between 1995 and 2002.
In addition, the percentage of banks with positive net income is constantly
and substantially larger for cooperative and savings banks than for com-
mercial banks. It is interesting to note that, despite the fact that savings
banks seems to be (on average) less profitable than cooperative banks,
almost all savings banks analysed generate profits in the period analysed.
In terms of shareholder value created (Figure 5.4), we found that mean
levels of EVAbkg are usually negative in every year and for all bank categ-
ories, with few exceptions (e.g. commercial banks in 2001 and 2002). The
analysis of the ratio between EVAbkg and capital invested enables us to
quantify the size of shareholder value destroyed by French banks over the
period analysed. We found that French banks annually substantially
destroyed value for their shareholders in 1995, 1996 and 1998, while gener-
ating value in 1997 and from 1999 up to 2002. We found substantial
differences between commercial, cooperative and savings banks. French
commercial banks appear to generate higher mean ratio between EVAbkg
and capital invested than the savings and cooperative banks from 1996 up
to 2001. We found that French cooperative and savings banks achieve posi-
tive mean levels only in 2001 and 2002, while (on average) these banks
seems to have destroyed a small part of shareholder value (around 1 per
cent of capital invested). Looking at the number of banks with a positive
EVAbkg, this percentage increased over the period analysed from 39.36 per
cent in 1995 to 63.39 per cent in 2002 showing a wider ability of French
banks to create value for their shareholders. Table 5.10 reports the list of the
thirty French banks that generated the largest EVAbkg in 2002. We find that
these banks generated an EVAbkg that is 8.99 per cent of the capital invested,
while net income was 12.63 per cent of equity capital (ROE) and 0.83 per
cent of total assets (ROA). Eleven of these top thirty EVAbkg creators are
among the twenty largest French banks. Among these, we found that nine
banks destroyed shareholder value, while the other eleven generated value
for their shareholders. On average, the top twenty banks generated an
EVAbkg amount to 3.17 per cent of the capital invested, while net income
was 11.29 per cent of equity capital (ROE) and 0.50 per cent of total assets
(ROA).
In Italy, we found that banks appear to generate substantially lower
mean profits than in the UK and France. Considering all bank categories
(Tables 5.7 and 5.8), we found that Italian banks achieved annual mean
profit between of Euro 14.32 million over the period 1995–2002 (ranging
from 3.02 Euro million in 1997 to 19.19 Euro million in 2001). Italian
Table 5.10: Top thirty EVAbkg creators in French banking (in 2002)
EVA on
Capital
1 BNP Paribas Commercial Bank 6364.5 2104.3 4260.1 11.71 18.46 0.58
2 Société Générale Calédonienne Commercial Bank 1928.0 419.6 1508.4 22.19 28.29 2.02
de Banque – SGCB
3 Banque de Polynesie Commercial Bank 1437.3 312.6 1124.7 22.21 25.00 1.33
4 CCF Commercial Bank 721.3 337.5 383.8 7.02 15.51 0.85
5 Société Générale Commercial Bank 1859.4 1497.0 362.4 1.49 16.82 0.49
6 CDC Ixis Commercial Bank 762.4 417.0 345.4 5.11 12.02 1.01
7 Banque Paribas Commercial Bank 530.3 285.6 244.7 5.29 7.39 0.15
8 Groupe Banques Populaires Cooperative Bank 931.8 687.1 244.7 2.20 8.47 0.33
9 Crédit Agricole d’Ile-de-France Commercial Bank 317.0 86.1 230.9 13.07 11.63 0.98
10 Casden Banque Populaire Cooperative Bank 256.8 73.7 183.1 15.34 19.99 1.55
11 Banque Paribas Pacifique Commercial Bank 288.3 111.5 176.9 9.80 17.15 1.07
12 Crédit Industriel et Commercial – CIC Commercial Bank 435.4 264.4 171.0 3.18 12.60 0.52
13 Groupe Caisse d’Epargne Savings Bank 1321.9 1151.0 170.9 0.92 7.71 0.28
14 Compagnie Bancaire Commercial Bank 250.5 98.7 151.8 9.49 19.51 0.81
15 RCI Banque Commercial Bank 206.8 103.0 103.8 6.22 14.31 0.89
16 BRED Banque Populaire Cooperative Bank 168.7 81.1 87.6 6.67 9.01 0.32
17 Crédit Agricole du Pas-de-Calais Cooperative Bank 133.0 47.7 85.3 11.03 7.60 1.15
18 Crédit Agricole du Nord Est Cooperative Bank 155.5 79.3 76.2 5.93 6.31 0.76
19 Crédit Agricole Pyrenees Gascogne Cooperative Bank 125.3 54.4 70.9 8.05 6.24 0.61
20 Crédit Agricole du Nord Cooperative Bank 122.5 52.2 70.3 8.31 7.77 0.94
21 Crédit Industriel d’Alsace et de Commercial Bank 126.8 57.2 69.6 7.51 13.18 0.26
Lorraine – Banque CIAL
22 Crédit du Nord Commercial Bank 151.1 82.5 68.6 5.14 14.86 0.78
23 Crédit Agricole du Finistère Cooperative Bank 112.0 44.3 67.7 9.43 7.25 0.79
181
182
Table 5.10: Top thirty EVAbkg creators in French banking (in 2002) – continued
EVA on
Capital
24 Crédit Agricole Loire-Atlantique Cooperative Bank 104.9 37.5 67.5 11.11 6.48 0.73
25 Crédit Agricole du Morbihan Cooperative Bank 90.2 23.5 66.7 13.66 6.34 0.59
26 Crédit Agricole Alpes Provence Cooperative Bank 109.1 43.5 65.5 7.18 6.68 0.64
27 Crédit Agricole Loire Haute-Loire Cooperative Bank 96.0 31.8 64.3 9.89 9.65 1.11
28 Société Marseillaise de Crédit Commercial Bank 85.4 24.2 61.2 15.61 17.87 0.99
29 Banque Nationale de Paris Commercial Bank 108.1 47.2 60.9 7.46 13.89 1.39
Intercontinentale BNPI
30 Caisse régionale de Crédit Agricole Cooperative Bank 97.7 37.6 60.0 9.85 11.39 1.18
mutuel Alsace Vosges

banks registered constantly positive ROE and ROA over the period
1995–2002 that are, surprisingly higher than those of French banks. The
percentage of banks with positive net income is similar to France and the
UK (around 90–95 per cent). By distinguishing according to bank cate-
gories, we found that commercial banks achieved lower mean income than
cooperative and commercial banks between 1995 and 1997, although these
substantially increased between 1998 and 2002. However, the analysis of
profitability ratios (both ROE and ROA) enables us to quantify the profits
generation. We found that all banks in Italy achieved positive mean values
and that cooperative and savings banks are, on average, more profitable
than commercial banks. In addition, the percentage of banks with positive
net income was constantly and substantially larger for cooperative and
savings banks than for commercial banks. In terms of shareholder value
created, we found that mean levels of EVAbkg are usually negative in every
year and for all bank categories (with few exceptions). The analysis of the
ratio between EVAbkg and capital invested enables us to quantify the size of
shareholder value destroyed by Italian banks over the period analysed. We
found that Italian banks annually destroyed value for their shareholders in
1995, generated value from 1996 to 2000 and slightly destroyed value in
2001 and 2002. We found substantial differences between commercial,
cooperative and savings banks. Italian saving banks registered a lower mean
ratio between EVAbkg and capital invested than the commercial and co-
operative banks from 1995 to 1999 and a higher mean ratio from 2000 to
2002. Italian cooperative banks are usually found to have destroyed value,
except in 1999 and 2000. Commercial banks seem to have destroyed value
between 1995 and 1998 and since then created value for their shareholders.
Looking at the number of banks with a positive EVAbkg, we found that
Italian banks generate profits widely (around 90 per cent), but only a part
of these banks also create value for their shareholders. In detail, the per-
centage of banks with a positive EVAbkg is very low in 1995 and 1996
(around 13 per cent) and, since then, substantially increased as plotted in
Figure 5.5. Table 5.11 reports the list of the thirty Italian banks that gener-
ated the largest EVAbkg in 2002. We find that these banks generated an
EVAbkg of 12.66 per cent of capital invested, while net income was 15.93
per cent of equity capital (ROE) and 0.99 per cent of total assets (ROA). It is
interesting to note that, in 2002, only four of the top twenty EVAbkg cre-
ators come from the top twenty largest banks. Among the twenty largest
Italian banks, we found that eight banks destroyed shareholder value, while
the other twelve generated value for their shareholders. On average, the top
twenty banks generated an EVAbkg that was 3.25 per cent of capital
invested, with net income 12.41 per cent of equity capital (ROE) and
0.77 per cent of total assets (ROA).
German banks are found to generate lower profits and shareholder value
than in the other three countries. Considering all bank categories, we
Panel A ROA and EVA on capital invested

15,00%
10,00%
ROE and EVA/Capitial invested
5,00%
0,00%
-5,00%
-10,00%
1995 1996 1997 1998 1999 2000 2001 2002
Panel B Percentage of banks with a positive ROA and EVA on capital invested
120%
100%
net income and EVA
80%
60%
40%
20%
0%
1995 1996 1997 1998 1999 2000 2001 2002
Figure 5.5: Return on equity and EVA on capital invested in Italian banking between
1995 and 2002
Table 5.11: Top thirty EVAbkg creators in Italian banking (in 2002)
EVA on
Capital
1 Banca Commerciale Italiana SpA. (COMIT) Commercial Bank 2778.7 1166.4 1612.2 17.16 18.72 0.98
2 Rolo Banca 1473 S.P.A. Commercial Bank 818.8 334.3 484.5 17.61 22.96 1.34
3 UniCredit Banca Commercial Bank 787.5 459.0 328.4 8.69 27.93 1.42
4 Cariverona Banca SpA Commercial Bank 505.9 187.8 318.1 17.50 22.92 1.97
5 Banco Popolare di Verona e Novara Cooperative Bank 787.7 483.0 304.7 6.52 12.99 0.93
6 Banca CRT SpA – Banca Cassa di Savings Bank 431.1 167.1 264.0 16.32 26.31 1.80
Risparmio di Torino
7 Banca Nazionale dell’Agricoltura SpA Commercial Bank 317.7 90.6 227.1 30.86 –5.05 –0.17
8 Cassa di Risparmio di Parma e Savings Bank 250.8 100.7 150.1 15.41 21.97 1.14
Piacenza SpA
9 Banco Ambrosiano Veneto SpA Commercial Bank 336.7 188.2 148.5 8.15 18.78 0.69
10 Banco di Brescia San Paolo Cab Commercial Bank 190.9 79.2 111.7 14.57 27.78 0.97
SpA – Banco di Brescia SpA
11 Deutsche Bank SpA Commercial Bank 198.0 91.3 106.8 12.09 16.33 1.04
12 Banco di Sardegna SpA Commercial Bank 208.8 113.0 95.8 8.35 6.76 0.55
13 Banca Popolare di Bergamo – Cooperative Bank 443.0 350.3 92.7 2.73 12.75 0.94
Credito Varesino
14 CAB & Banca San Paolo di Commercial Bank 211.4 121.5 89.9 7.65 9.94 0.58
Brescia Combined
15 Banco di Sicilia SpA – BdS Commercial Bank 277.2 189.5 87.7 4.78 10.78 0.57
16 Banca Popolare di Brescia Scarl Cooperative Bank 145.7 62.6 83.1 13.72 21.51 0.67
17 Findomestic Banca SpA Commercial Bank 125.2 43.4 81.8 19.47 18.60 1.26
18 Banca Agricola Mantovana SpA Cooperative Bank 191.8 111.1 80.7 7.51 9.75 0.89
19 Banca Monte dei Paschi di Siena SpA Commercial Bank 1171.8 1095.5 76.3 0.74 10.37 0.52
185
186
Table 5.11: Top thirty EVAbkg creators in Italian banking (in 2002) – continued
EVA on
Capital
20 Cassa di Risparmio di Padova e Savings Bank 142.5 72.8 69.8 9.90 16.23 0.99
Rovigo SpA
21 Cassa di Risparmio della provincia Savings Bank 74.1 10.5 63.6 62.83 40.03 2.77
di Viterbo SpA
22 Banca Regionale Europea SpA Commercial Bank 148.2 88.6 59.6 6.95 10.34 1.20
23 Credito Bergamasco Commercial Bank 117.2 59.5 57.7 7.95 11.86 0.83
24 Banca Toscana SpA Commercial Bank 190.2 132.5 57.7 4.50 10.13 0.76
25 Cassamarca Cassa di Risparmio Savings Bank 80.0 26.4 53.6 20.97 24.34 1.48
della Marca Trivigiana SPA
26 Banca Opi SpA Commercial Bank 103.0 52.5 50.5 9.94 9.13 0.22
27 Cassa di Risparmio di Biella e Vercelli – Savings Bank 64.4 20.5 43.9 22.09 19.09 1.07
BIVERBANCA
28 Banca Carige SpA Commercial Bank 189.5 153.8 35.7 1.94 7.49 0.73
29 Banca Popolare dell’Emilia Romagna Cooperative Bank 201.9 168.0 34.0 1.56 9.01 0.69
30 Credito Emiliano Commercial Bank 149.64 122.15 27.49 2.53 8.20 0.56

found that German banks achieved annual mean profits between around
Euro 4 million from 1995 to 1999 and close to zero in 2000 and 2001. As a
result, German banks generated a mean annual net income of Euro 2.91
million (Table 5.7). Although net income is lower than in the other coun-
tries, we found in Germany the highest percentage of banks generating
profits (96.06 per cent). In addition, German banks registered constantly
positive ROE and ROA over the period 1995–2002, although much lower
than the UK banks. These mean levels are, however, strongly affected by
the high number of German cooperative banks (around four times the
commercial banks). By distinguishing according to bank categories, we
found that commercial German banks (similarly to Italian banks) achieved
higher mean income than cooperative and savings banks between 1995
and 2000, while recording substantial mean net losses in 2001 and 2000.
However, the analysis of profitability ratios (both ROE and ROA) enables us
to note that German cooperative and savings banks are on average more
profitable than commercial banks and that these banks achieve positive
profits also in 2001 and 2002. In addition, the percentage of banks with
positive net income is constantly and substantially larger for cooperative
and savings banks than for commercial banks.
The lower profitability of German banks seems to be also confirmed in
terms of shareholder value created. We found that the mean levels of
EVAbkg are usually negative in every year and for all bank categories (with
few exceptions). The analysis of the ratio between EVAbkg and capital
invested enables us to quantify the size of shareholder value destroyed by
German banks over the period analysed (Figure 5.6). We found that
German banks annually destroyed value for their shareholders around
1 per cent of capital invested, except in 1996 and 1997 when they created
value for shareholders. We did not find substantial differences between
commercial, cooperative and savings banks. Looking at the number of
banks with a positive EVAbkg, we found that while 90 per cent of German
banks generate profits, only a small portion of these banks also create value
for their shareholders. Only in 1996 and 1997, the percentage of banks
with a positive EVAbkg is superior to 50 per cent, while this percentage
ranges between 20 per cent and 40 per cent in the other years. Table 5.12
reports the top thirty German banks with the highest EVAbkg. It is interest-
ing to note that, in 2002, none of the large German commercial banks are
listed. Most of the largest German banks achieved positive net income
(e.g. the net income of Deutsche Bank AG was Euro 751 million), but these
generate small return on capital invested (e.g. ROA and ROE of Deutsche
Bank AG are 0.12 per cent and 3.6 per cent, respectively). In addition,
when we consider the opportunity cost of invested capital, net income
(measured on an economic basis) is lower than the capital charge required
by shareholders and, as such, most of the largest German banks seem to
have destroyed shareholder value in 2002.
Panel A ROA and EVA on capital invested

15,00%
ROE and EVA/capital invested
10,00%
5,00%
0,00%
-5,00%
-10,00%
1995 1996 1997 1998 1999 2000 2001 2002
Panel B Percentage of banks with a positive ROA and EVA on capital invested
120%
100%
net income and EVA
80%
60%
40%
20%
0%
1995 1996 1997 1998 1999 2000 2001 2002
Figure 5.6: Return on equity and EVA on capital invested in German banking
Table 5.12: Top thirty EVA creators in German banking (in 2002)
EVA on
Capital
Economic Capital Capital invested+ ROE+ ROA+
Rank Bank name Bank Specialisation NOPAT* invested charge* EVAbkg* (%) (%) (%)
1 Hamburger Sparkasse Savings Bank 604.7 2100.4 130.1 474.6 22.59 16.83 1.06
2 Berliner Volksbank eG Cooperative Bank 149.7 783.5 48.5 101.1 12.91 2.99 0.14
3 Vereins–und Westbank AG Commercial Bank 177.8 1659.0 102.8 75.1 4.52 15.98 0.66
4 BMW Bank GmbH Commercial Bank 66.8 454.9 28.2 38.6 8.49 0.00 0.00
5 Berenberg Bank–Joh. Berenberg, Commercial Bank 43.8 170.6 10.6 33.2 19.48 25.96 2.21
Gossler & Co.
6 Raiffeisen – Volksbank im Cooperative Bank 38.8 100.3 6.2 32.6 32.53 12.04 0.91
Landkreis Altoetting eG
7 Bankhaus H. Aufhäuser Commercial Bank 33.9 72.2 4.5 29.5 40.81 29.56 4.27
8 AKB Privat – und Handelsbank Commercial Bank 46.3 282.3 17.5 28.9 10.22 26.04 1.39
9 PSA Finance Deutschland Commercial Bank 35.2 179.9 11.1 24.0 13.35 19.02 1.77
10 Stadtsparkasse Düsseldorf Savings Bank 56.7 539.3 33.4 23.3 4.32 12.74 0.60
11 Kreissparkasse Biberach Savings Bank 38.6 301.6 18.7 19.9 6.61 8.14 0.37
12 Gontard & Metallbank AG Commercial Bank 25.4 144.3 8.3 17.1 11.82 19.56 1.78
13 VR Bank Suedpfalz eG Cooperative Bank 23.3 101.8 6.3 17.0 16.73 6.02 0.34
14 Volksbank Offenburg eG Cooperative Bank 22.3 88.5 5.5 16.8 18.97 6.19 0.28
15 Volksbank Freiburg eG Cooperative Bank 21.6 81.0 5.0 16.5 20.42 5.16 0.23
16 MKG Bank GmbH Commercial Bank 27.0 170.6 10.6 16.4 9.61 15.02 1.78
17 Sparkasse Krefeld Savings Bank 37.1 342.0 21.2 15.9 4.65 13.18 0.62
18 Sparkasse Aachen Savings Bank 44.9 477.7 29.6 15.3 3.20 6.81 0.33
19 Hanseatic Bank GmbH & Co. Commercial Bank 20.3 83.3 5.2 15.2 18.23 20.90 1.84
20 Volksbank in Stuttgart AG Cooperative Bank 22.1 184.0 11.4 10.7 5.80 2.08 0.09
21 Volksbank Franken eG Cooperative Bank 13.7 52.8 3.3 10.4 19.64 2.08 0.11
22 St. Wendeler Volksbank eG Cooperative Bank 11.5 26.0 1.6 9.9 37.86 2.03 0.08
23 Gross-Gerauer Volksbank EG Cooperative Bank 16.5 107.1 6.6 9.8 9.17 3.53 0.22
189
190
Table 5.12: Top thirty EVA creators in German banking (in 2002) – continued
EVA on
Capital
Economic Capital Capital invested+ ROE+ ROA+
Rank Bank name Bank Specialisation NOPAT* invested charge* EVAbkg* (%) (%) (%)
24 Volksbank Hochrhein eG Cooperative Bank 12.0 36.2 2.2 9.7 26.85 21.43 0.92
25 Westfalenbank AG Commercial Bank 26.4 271.0 16.8 9.6 3.56 15.38 1.21
26 Volksbank Speyer – Neustadt – Cooperative Bank 15.0 90.6 5.6 9.4 10.40 3.45 0.18
Hockenheim
27 Rheingauer Volksbank eG Cooperative Bank 12.0 44.6 2.8 9.2 20.69 3.73 0.24
28 ING BHF-BANK AG Commercial Bank 148.9 2254.0 139.7 9.2 0.41 6.86 0.29
29 Volksbank Raiffeisenbank Cooperative Bank 10.2 17.3 1.1 9.1 52.83 2.44 0.13
Murr-Lauter eG
30 Nissan Bank Gmbh Commercial Bank 17.1 140.5 8.7 8.4 5.97 11.28 1.09

Conclusion
This chapter has dealt with the empirical investigation of the value-rele-
vance of bank performance measures and with the measurement of share-
holder value created across European banking industries.
The first part of the chapter investigated the information content of tra-
ditional (such as interest and intermediation margins, ROE; ROA and net
income) and non-traditional (such as residual income and EVA) bank per-
formance indicators in the light of creating shareholder value (added)
within the European banking industry. While there is a general agreement
as to the concept of shareholder value, there is debate as to the best
method for assessing the value created by firms for their owners, as
researchers and practitioners grapple with different performance metrics.
There are a growing number of studies investigating which performance
measure is the most compatible with shareholder value creation, but the
evidence surrounding this issue is mixed. In addition, few papers investi-
gate this issue focusing on the banking sector. In order to identify which
performance metric is the most compatible with shareholder value cre-
ation, our analysis examines both the relative information content (which
is useful to select a single measure since performance indicators are consid-
ered mutually exclusive) and the incremental information content (which
aims to assess if a performance indicator adds information to the data pro-
vided by another measure). Here, we have focused on four European
banking systems, which comprise seventy-one publicly listed banks (i.e.
twenty French banks, thirteen German banks, twenty-eight Italian banks
and ten UK banks) over a period of seven years (from 1 January 1996 to 31
December 2002).
Following the methodology outlined earlier, we use simple regression
analysis to test the relative and incremental information content of various
performance measures in explaining shareholder value creation in banking.
Overall, our study compares two performance indicators specific to com-
mercial banks (i.e. interest margins and the intermediation margin),
various indicators used by a wide range of firms (such as ROA and ROE), as
well as EVA measures calculated using both standard (as commonly used in
studies on non-financial companies) and a bank-tailored procedure. Our
results suggest that the EVA measure that accounts for the specifics of
banking (EVAbkg) outperforms all other performance measures since it
has the greatest ability to explain variation of MAR (expressing share-
holder value created over the period analysed). In contrast, the standard
Economic Value Added measure (EVAstd) does not seem to explain share-
holder value creation better than a wide range of simple accounting and
other performance measures as it has the lowest ability to explain variation
of MAR. These results provide evidence that it is necessary to accurately
consider the peculiar nature of capital in banking as well as other account-
ing adjustments if accurate measures of bank performance (from a share-

holder wealth perspective) are to be used.
Regarding the analysis of mean levels of EVA created in France, Germany,
Italy and the United Kingdom, we found that European banks generate, on
average, substantial and widespread profits, but generally do not seem to
have created substantial shareholder value over the period analysed since
profit (measured from an economic viewpoint) is often lower than the
opportunity cost of equity invested capital. We found that the situation
described above is not common to all countries analysed. UK banks appear
to be the most profitable and the percentage of banks with a positive net
income is 91.60 per cent. We found also positive mean EVAbkg levels from
1995 to 1998 and in 1999 and negative mean EVAbkg levels in 1998 and
from 2000 to 2002. We found that almost all UK banks generally generate
higher profits than banks in the other countries, but not all UK banks are
able to achieve profits superior to the opportunity cost of equity capital.
For example, twelve of the top thirty EVAbkg creators in the UK are in the
top twenty largest banks. Among these twenty largest UK banks, we found
that eight banks destroyed shareholder value.
In France, banks appear to generate substantially lower mean profits than
in the UK, but more than in Italy and Germany. French cooperative banks
are, on average, more able to remunerate equity capital with profits, while
cooperative and commercial banks have similar ROA mean levels between
1995 and 2002. In addition, the percentage of banks with positive net
income is constantly and substantially larger for cooperative and savings
banks than for commercial banks. In terms of shareholder value created,
French banks seem to have substantially destroyed value for their share-
holders in 1995, 1996 and 1998, while generating value in 1997 and from
1999 up to 2002. We found substantial differences between commercial,
cooperative and savings banks since French commercial banks appear to
generate higher EVAbkg on capital invested than the savings and coopera-
tive banks from 1996 up to 2001. We found that eleven of the top thirty
EVAbkg creators in 2002 are among the twenty largest French bank. Among
these twenty largest French banks, we found that nine banks destroyed
shareholder value, while the other eleven generated value for their
shareholders.
In Italy, we found that banks appear to generate substantially lower
mean profits than in the UK and France. We also found that cooperative
and savings banks are, on average, more profitable than commercial banks.
Levels of EVAbkg are usually negative in every year and for all banks (with
few exceptions). Around 90 per cent of Italian banks generate profits, but
only a part of these banks also create value for their shareholders since the
percentage of banks with a positive EVAbkg is very low. Focusing on
the thirty Italian banks that generated the largest EVAbkg in 2002, we
found that only four of the top twenty EVAbkg creators are in the top
twenty largest banks. Among the twenty largest Italian banks, eight banks
destroyed shareholder value, while the remainder generated value for their
shareholders.
In Germany, banks appear to generate lower profits and shareholder
value than in the other three countries. The lower profitability of German
banks seems to be also confirmed in terms of shareholder value created. We
found that mean levels of EVAbkg are usually negative in every year and for
all bank categories (with a few exceptions). We did not find substantial
differences between commercial, cooperative and savings banks. Among
the top thirty German banks with the highest EVAbkg, we did not find any
of the large German commercial banks. Most of the largest German banks
achieved positive net income (e.g. the net income of Deutsche Bank AG is
Euro 751 million), but these supply small returns on capital invested (e.g.
ROA and ROE of Deutsche Bank AG are 0.12 per cent and 3.6 per cent,
respectively). In addition, when we consider the opportunity cost of capital
invested, net income (measured on an economic basis) is lower than the
capital charge required by shareholders and, as such, most of the largest
German banks destroyed shareholder values.
While this chapter provides a detailed account and analysis of share-
holder value creation in European banks, the next chapter examines factors
that are believed to be major drivers of shareholder value: efficiency,
productivity and customer satisfaction.
6
Measuring Shareholder Value Drivers in
Banking
Introduction
All stakeholders are dependent on each other for their success in the long
term and managing to create sustained shareholder value is not a zero-sum
game. Creating stable shareholder value requires an intense focus on deliv-
ering benefits to customers in the most efficient way, hiring and retaining a
motivated workforce, maintaining excellent supplier relationships, and
being a good corporate citizen in each of the local areas where the
company has a presence. As such, several strategies have been developed
since the 1990s to improve customer satisfaction that have included the
redesign of productive and delivery services, the development of more
flexible organisational structures, the introduction of incentive schemes
that motivate banks’ human resources to act according to shareholders’
goals, etc. In Chapter 4, we identified four primary drivers that create share-
holder value in banking. These relate to improved bank efficiency (cost
efficiency, profit efficiency and productivity), enhanced customer satisfac-
tion, optimising banks’ financial structure and developing an optimal mix
of business activities.
This chapter empirically analyses the main features of bank sector
efficiency in the systems under study (France, Germany, Italy and the UK)
and also proposes a way of empirically investigating customer satisfaction.
Banks’ efficiency is estimated focusing on cost and profit efficiency as well
as productivity changes over time. The adoption of several methods and
measures of efficiency are necessary since corporate efficiency can be mea-
sured focusing on production (expressing the company’s ability to trans-
form inputs into outputs), on costs (expressing the ability to make savings
costs for a given level of outputs) and on profits (expressing the ability to
achieve a higher level of profits for a given level of output levels).
Regarding customer satisfaction, this is usually measured by some form of
survey. However, few measurements of customer satisfaction at the bank-
ing industry level have been carried out and these are usually not publicly
194

Measuring Shareholder Value Drivers in Banking 195
available, so for our purposes we measure customer satisfaction using a

proxy from publicly available data. We analyse efficiency and customer sat-
isfaction issues in this chapter so the findings can be used in Chapter 7
which provides an empirical assessment of the determinants of shareholder
value in European banking.
Measuring cost efficiency

As outlined in Chapter 4, improving productive efficiency is one of the
main endogenous channels for creating shareholder value in a bank.
Productive efficiency is a broad concept and it can be better defined focus-
ing on different aspects: technical efficiency, allocative efficiency, overall
efficiency, and scale efficiency. Technical efficiency (TE) expresses the
ability of a firm to obtain maximal outputs from a given set of inputs or of
minimising inputs for a given target of outputs: this component focuses
only on physical quantities and technical relationships. Allocative efficien-
cies (AE) refer to the ability to use inputs in optimal proportions, given
their respective prices and production technology. Overall efficiency (OE),
also called ‘X-efficiency’ (or economic efficiency), expresses the ability of a
firm to choose its input and/or output levels and mix them to optimise its
economic goal. This section deals with the empirical investigation of
efficiency and is organised in three subsections concerning (respectively)
the methodology, the sample description and the results.
Methodology for estimating bank efficiency

We measure banking cost efficiency using two different methods:
Stochastic Frontier Analysis (SFA) and the Envelopment Analysis (DEA). We
adopt two frontier methods of efficiency measurement so as to be able to
test the robustness of our estimates. Firstly, we used Stochastic Frontier
Analysis (SFA) to estimate cost X-efficiency for a sample of European banks
over the period 1995–2002. Specifically, we use Battese and Coelli’s (1995)
stochastic frontier model, where:
ln TCit = xit β + (Vit + Uit) (6.1)
where ln TCi is the logarithm of the cost of production of the i-th bank, xi
is a kx1 vector of standardised input prices and output of the i-th bank,
β is a vector of unknown parameters, Vi are random variables which are
assumed to be i.i.d N(0,σ2v) and independent of Ui, Ui are non-negative
random variables which are assumed to account for the cost inefficiency
in production and are assumed to be i.i.d N(mit,σ2U), mit is defined as
mit = zit,δ, zit, is a px1 vector of variables which may influence the efficiency
of a bank, and δ is a px1 vector of parameters to be estimated. Our sample
is composed of different type of banks (namely, commercial, cooperative
and savings banks) and spans an eight-year period, and the Battese and
Coelli (1992) model enables us to control for whether a particular time
period influences bank efficiency.
We use the standard translog functional form and our cost function is
the following:1
3 3
ln TC = α0 + ∑ αi ln yi +∑ βj ln wj +
i=1 j=1
 3 3 3 3 
1
+ 
 ∑ ∑δ ln yi ln yj + ∑ ∑γij ln wi ln wj
ij

 + (6.2)
2  i=1 j=1 i=1 j=1 
3 3
+ ∑ ∑ ρij ln yi ln wj + ln uc + ln εc
i=1 j=1
where TC is the logarithm of the cost of production, yi is the output quanti-

ties for the i-th output, wi is the input prices for the i-th input. In order to
3 3
guarantee linear homogeneity in factor prices (i.e. ∑ βj = 1 ; ∑ γij = 0 and
3 j=1 i=1
∑ ρij = 0), it is necessary (and sufficient) to apply the following restrictions:
j=1
(1) standard symmetry: according to this restriction, it is assumed and that
δij = δij and; γij = γij; and (2) linear restriction of the cost function (model
6.2). In addition, the factor share equations (embodying restrictions
imposed by Shephard’s Lemma or Hotelling’s Lemma) are excluded since
these would impose the undesiderable assumption of no allocative
inefficiencies (see, for example, Berger and Mester (1997)).
To define bank inputs and outputs, we have adopted the value-added
approach, originally proposed by Berger and Humphrey (1992), to identify
bank outputs and we posit that labour (price of labour is measured as per-
sonnel expenses over total assets), physical capital (price of physical capital
is expressed as the total capital equipment expenses over total fixed assets)
and financial capital (price of financial capital is measured as interest costs
on borrowed funds) are inputs, whereas demand deposits, total loans and
other earning assets are outputs as illustrated in Figure 6.1. The value-added
approach identifies bank inputs and outputs depending on the contribu-
tion of bank items (on both sides of the balance sheet) to the creation of
bank value-added. In the value-added context, deposits are considered to
have some output features and various previous studies have used deposits
as outputs (i.e. the sum of demand deposits). We consider only demand
deposits as a bank output since banks usually pay low interest rates on
these funds and can charge fees on depositors for services provided with
demand deposits. One might claim that off-balance sheet items (OBS) items
may play a role in generating bank value-added and some studies have
included these items as bank outputs. However, we omit off-balance sheet
(OBS) items since our sample includes many small banks (especially,
Panel A SFA2
Total Costs (TC ) – total costs of production (comprising operating costs and interest paid on deposits);
Input 1 (w1) – average cost of labour (personnel expenses/total assets);
Input 2 (w2) – average cost of physical capital (total equipment capital expenses/total fixed-tangible assets);
Input 3 (w3) – average cost of financial capital deposits (interest costs on borrowed funds on the average
amount of borrowed funds);
Output 1 (Y1) – demand deposits;
Output 2 (Y2) – total loans;
Output 3 (Y3) – other earning assets.
Panel B DEA
Input 1 (X1) – labour (measured as average number of employees);
Input 2 (X2) – physical capital (expressed as the average value of fixed-tangible assets);
Input 3 (X3) – financial capital (measured as loanable funds);
Output 3 (Y3) – other earning assets;
Price Input 1 – average labour costs (total labour costs on number of employees);
Price Input 2 – total equipment capital expenses/total fixed-tangible assets;
Price Input 3 – interest costs on borrowed funds on the average amount of borrowed funds.
Figure 6.1: Inputs and outputs used in estimating cost efficiency in European banking
mutual banks and savings banks) that do not have OBS items or data are
not available in the Bankscope database.
In addition, we measure cost efficiency using a non-parametric approach,
i.e. Data Envelopment Analysis (DEA). We use DEA to estimate the economic
efficiency for our sample of banks, so we can distinguish between technical,
allocative and scale efficiency. We can also use these estimates to cross-check
the consistency of the parametric estimates. DEA is a linear programming
methodology which uses data on the input and output quantities of a group
of firms to construct a piece-wise linear surface over the data points.
Companies on the frontier surface are called ‘technically efficient’, while other
companies are labelled ‘technically inefficient’: efficiency scores are deter-
mined by comparing their performance to the envelopment surface. DEA was
developed by Charnes et al. (1978), who generalised the piece-wise-linear
conical hull approach to estimate the efficient frontier and radial inefficiencies
scores (proposed by Farrell, 1957) to multiple outputs and reformulated the
optimisation process as a mathematical programming problem. Successively,
Banker et al. (1984) improved the model by removing the assumption of
Constant Returns to Scale (CRS). The VRS model proposed by Banker et al.
(1984) is often solved in two stages:3 as suggested in Ali and Seiford (1993),
the first involves a proportional contraction in inputs, while the second stage
proposes a maximisation of the sum of (any remaining) slacks. This second
stage linear programming problem (which must be solved for each of the N
DMU’s involved) may be defined by:
Max λ, OS, IS -(M1’OS+k1’IS),

St. -yi + Yλ – OS ≥ 0
(6.3)
θxi – Xλ – IS ≥ 0
λ ≥ 0, OS ≥ 0, IS ≥ 0
where xi is a Kx1 vector of inputs for the i-th banks, yi is a Mx1 vector of
outputs for the i-th banks, X is an input matrix KxN, Y is an output matrix
MxN, M1 is a Mx1 vector of ones, K1 is a Kx1 vector of ones, OS is an Mx1
vector of output slacks, OS is an Kx1 vector of input slacks (note that in this
second stage, θ is not a variable, but its value is taken from the first stage
results).
However, because the second stage implies the maximisation of the sums
of slacks (rather than a minimisation) and the projected point obtained is
not invariant to the unit of measurement, the specification of the peers and
targets (necessary for the calculations of the efficiency scores) obtained in
the second stage may be unsatisfactory. To address this problem, DEA is
solved using the multi-stage DEA methodology proposed by Coelli (1998)
in our empirical analysis. This method involves a sequence of DEA models
to identify the projected efficient points and is therefore more computa-
tionally demanding than other methods:4 however, it avoids the necessity
of maximising the sum of slacks and the efficient projected points
identified are invariant to units of measurement. In detail, the multi-stage
DEA approach involves six steps:
1. Run the multiplier form of the following DEA model, originally pro-
posed by Charnes et al. (1978):
Max µ,υ (µ’yi)

St. υ’ xi = 1
(6.4)
µ’yj,–υ’ xj < 0, j=1,2, …, N
µ,υ ≥ 0.
where xi is a KX1 vector of inputs for the i-th bank, yi is a MX1 vector of
outputs for the i-th DMU, the notation change from u is an MX1 vector
of output weights and υ is an KX1 vector of input weights.
2. Identify the efficient set in a Koopmans sense (i.e. all firms without
slacks and with a technical efficiency score of θ=1). This result is
achieved by running the following linear programming (LP) model,
where any remaining slacks are maximised:5
Max λ, OS, IS -(M1’OS+k1’IS),

St. –yi + Yλ – OS ≥ 0
(6.5)
cxi – Xλ – IS ≥ 0
λ ≥ 0, OS ≥ 0, IS ≥ 0
where cxi refers to the input vector of the i-th bank which has been
contracted by being multiplied by the θ obtained in step 1, xi is a Kx1
vector of inputs for the i-th bank, yi is a Mx1 vector of outputs for the
i-th bank, X is an input matrix KxN, Y is an output matrix MxN, M1 is a
Mx1 vector of ones, K1 is a Kx1 vector of ones, OS is an Mx1 vector of
output slacks, OS is an Kx1 vector of input slacks.
3. Identify (for the i-th firm in the ‘slack’ set) all input dimensions in which
some slack may exist. This result is achieved by running a sequence of
KLP’s, where in each LP only one of the inputs6 is allowed to contract.
The LP model for the j-th input of the i-th firm is specified as follows:
Min θ, λ θ,
St. –yi + Yeλ ≥ 0
θcxi j – Xe j λ ≥ 0 (6.6)
cxi≠j – Xe≠j λ ≥ 0
λ≥0
where cxij refers to the i-th input of the i-th bank which has been con-
tracted by being multiplied by the θ obtained in step 1, Xe j is an input
matrix 1xNe vector of the j-th inputs of all efficient firms, cxi ≠j is the
(K-1)x1 vector of inputs of the i-th firm (excluding the j-th input)
which has been contracted by being multiplied by the θ obtained in
step 1, Xe j is the (K-1)xNe matrix of inputs of all efficient firms (exclud-
ing the j-th input), Ne is the number of efficient firms (identified in step
2), Ye is the number of outputs of these efficient firms, yi is a Mx1 vector
of outputs for the i-th bank (it is important to note that this LP model
breaks down when some inputs are zero).
4. Seek a radial reduction in all inputs identified as having potential slack
by running the following LP model:
Min θ, λ θ
st. –yi + Yeλ ≥ 0
θcxis – Xesλ ≥ 0 (6.7)
cxi – Xe λ ≥ 0
ns ns
λ≥0
where cxi refers to the input vector of the i-th bank which has been
contracted by being multiplied by the θ obtained in step 1, Xe j is the
(K–1)xNe matrix of inputs of all efficient firms (excluding the j-th
input), Ne is the number of efficient firms (identified in step 2), Ye is the
number of outputs of these efficient firms, yi is a Mx1 vector of outputs
for the i-th bank, θ is a scalar (note that the superscript ‘s’ refers to the
subset of inputs having potential slack and ‘ns’ identifies the remainder
of the inputs).
5. Look for the input slack remaining in any dimension by repeating
steps 3 and 4 on the projected point identified in step 4 until no slack
remains in any input.
6. Conduct a radial expansion in output slack dimensions by repeating
steps 3–5 on the project point from step 5 until no slack remains in
any output. These final projected points will be on the efficient surface.
It is interesting to note that these projected points will be unit invari-
ant to the units of measurement adopted.7
After measuring technical and scale efficiencies, cost efficiency and

allocative efficiency are estimated since price information is available and
cost minimisation is a reasonable behavioural objective for European
banks. As such, the following DEA model enables us to measure cost
efficiency under the VRS and the cost minimisation8 assumptions:
Min λ,xi* wi’ xi*

st: –yi + Yλ ≥ 0,
xi*–Xλ ≥ 0 (6.8)
N1’λ=1
λ≥ 0
where wi is a vector of input prices for the i-th bank, xi* (which is calcu-
lated by LP) is the cost minimising vector of input for the i-th DMU, given
wi and yi.
The total cost, the allocative and the scale efficiency of the i-th firm are
calculated as follows:
• Cost Efficiency is CE = wi’xi*/wi’xi, which represents the ratio of

minimum cost to be observed.
CE
• Allocative Efficiency9 is AE = .
TE
• Scale efficiency is obtained by solving for both the CRS model (pro-
posed by Charnes et al. 1978) and the multi-stage DEA VRS model and
comparing the results as follows: TECRS = TEVRS x SE. In order to analyse

whether returns to scale are increasing or decreasing, this can be deter-
mined by solving for ‘non-increasing returns to scale’ in the input-ori-
entation.
Following the value-added approach for defining bank inputs and

outputs (as in the SFA estimates), we posit10 that labour (measured as per-
sonnel expenses), physical capital (expressed as the average value of fixed-
tangible assets) and financial capital (measured as loanable funds) are
inputs, whereas demand deposits, total loans and other earning assets are
outputs.
Sample description
The sample consists of unlisted banks from France, Germany, Italy and the
UK between 1995 and 2002 with financial information obtained from
Bankscope database.11 The same sample of banks was used in the second
part of Chapter 5 to investigate shareholder value in banking. We use a
cross-section sample by year since many bank observations would have
been lost selecting a balanced panel data set.
We prefer to use a sample of domestic banks for estimating the cost
efficiency frontier since banks in the same country are more homogeneous
(and comparable) than banks working in different countries. Similarly, we
also include various specific banks according to ownership type (namely,
commercial banks and, jointly, cooperative and savings banks12) since this
seems to guarantee a greater homogeneity to the sample. As such, we have
estimated fifty-six frontiers both using DEA and SFA.13 Table 6.1 provides
the number of banks considered for estimating each frontier using SFA
(panel A) and DEA (panel B). The dimension of cross-section sample used
in SFA and DEA estimations differ since all data outside the range of ± 3
standard deviation from the median were deleted to remove extreme values
(outliers) in our sample.
Results
This section illustrates the cost efficiency estimates obtained using SFA and
DEA. Table 6.2 reports descriptive statistics of the cost X-efficiency mea-
sures derived from SFA. Overall, European banks display inefficiency scores
ranging between 7.3 per cent (UK commercial banks in 1995) and 38.9 per
cent (German cooperative banks in 2001) and the level of dispersion of
average efficiency scores varies substantially from 4.2 per cent (German
savings banks in 1996) to 27 per cent (Italian commercial banks in 2001).
On average (considering all the 20,942 bank cost X-efficiency estimates
obtained running annual cross-sectional domestic frontiers), the mean cost
X-efficiency level over the period 1995–2002 is 78.1 per cent and the mean
level of dispersion of average efficiency scores is 17.8 per cent.
202
Table 6.1: Number of banks in samples used for estimating efficiency with SFA and DEA
Panel A SFA 1995 1996 1997 1998 1999 2000 2001 2002
Commercial banks 166 179 190 198 198 201 211 201
Cooperative & 102 102 120 122 126 129 142 136
France
savings banks* (80;22) (80;22) (95;25) (97;25) (101;25) (100;29) (111;31) (98;28)
Cooperative & 1097 1263 1400 1517 1702 1829 1987 1985
savings banks* (585;512) (714;549) (832;568) (930;587) (1102;600) (1220;609) (1371;616) (1381;604)
Germany
Cooperative & 157 164 195 326 450 530 569 586
Italy
savings banks* (86;71) (102;62) (132;63) (262;64) (386;64) (464;66) (505;64) (520;66)
Cooperative & 0 0 0 0 0 0 0 0
United
savings banks*
Kingdom
* The first number in the brackets refers to the number of cooperative banks and the second to the number of savings banks
Table 6.1: Number of banks in samples used for estimating efficiency with SFA and DEA – continued
Panel B DEA 1995 1996 1997 1998 1999 2000 2001 2002
Cooperative & 103 112 120 122 126 131 147 139
France
savings banks* (81;22) (90;22) (95;25) (97;25) (101;25) (102;29) (116;31) (110;29)
Cooperative & 1099 1260 1402 1522 1710 1842 2002 2013
savings banks* (587;512) (715;549) (835;567) (937;585) (1112;598) (1234;608) (1389;613) (1403;610)
Germany
Cooperative & 150 171 198 333 458 544 584 598
Italy
savings banks* (88;62) (109;62) (135;63) (269;64) (393;65) (478;66) (518;66) (532;66)
Cooperative & 0 0 0 0 0 0 0 0
United
savings banks*
Kingdom
Source: Bankscope.
203
Table 6.2: SFA cost X-efficiency mean levels in European banking between 1995 and 2002* 204
France Germany Italy United Kingdom Mean
Mean St.dev Mean St.dev Mean St.dev Mean St.dev Mean St.dev
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Commercial banks 72.03 13.67 77.94 6.70 83.47 13.36 92.66 3.86 78.63 10.18
Cooperative banks 70.17 10.25 78.18 8.88 79.62 14.67 N/A N/A 77.49 9.69
1995
Savings banks 76.36 8.52 78.00 10.38 76.78 14.46 N/A N/A 77.80 10.79
Total 71.83 12.21 78.08 9.29 79.95 14.20 92.66 13.38 77.86 10.49
1996
Total 68.25 11.72 84.11 4.60 81.79 15.44 90.12 6.64 81.78 7.00
1997
Total 69.41 12.91 83.91 5.53 84.28 10.02 79.32 23.21 81.79 7.67
1998
Total 86.88 8.68 81.93 7.23 78.87 15.03 78.11 21.49 81.92 9.20
1999
Total 73.56 13.14 81.63 6.60 78.89 15.01 78.96 18.14 80.09 9.37
Table 6.2: SFA cost X-efficiency mean levels in European banking between 1995 and 2002* – continued
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
2000
Total 67.31 11.95 79.81 7.06 78.99 14.58 76.42 20.61 78.22 9.55
2001
Total 71.19 10.02 69.80 10.72 77.20 16.02 75.71 21.56 71.67 12.06
2002
Total 70.46 11.24 74.17 9.60 79.06 6.93 83.33 12.35 75.11 9.24
Mean
Total 72.42 11.44 78.61 7.76 79.27 13.21 80.94 16.71 78.07 9.45
1995–2002
* Efficiency estimates for commercial banks are obtained by estimating an efficient frontier in a sample comprising only this category of bank, while
efficiency estimates for savings and cooperative banks are obtained by estimating a common frontier using a sample comprising both savings and
cooperative banks.
205
Focusing on the four banking systems analysed, UK banks display the

highest efficiency levels (on average, 80.9 per cent over the period
1995–2002) and the highest dispersion of average efficiency scores (on
average, 16.7 per cent) compared to the other three countries. Over the
period analysed, mean efficiency levels in UK banking substantially
decreased between 1995 and 2001 and sharply increased between 2001 and
2002. It is worthwhile remembering that the UK sample comprises only
commercial banks. French banks show the largest opportunities for improv-
ing X-efficiency with a mean X-inefficiency of 27.6 per cent and a disper-
sion of average efficiency of 11.4 per cent. Mean efficiency levels are quite
similar in the period analysed (around 70 per cent), except in 1998 when
the mean X-efficiency sharply increases to 86.9 per cent. Italian and French
banks exhibit similar mean levels (79.3 per cent and 78.6 per cent, respec-
tively), but German banks exhibit a lower range of dispersion of average
cost efficiency scores. Over the period analysed, annual mean X-efficiency
in Italy and France follow a similar trend since, for both countries, mean
annual X-efficiency increased between 1995 and 1998, decreased between
1998 and 2001 and sharply increased in 2002.
Looking at the different bank specialisations, commercial banks display a
lower mean level of cost X-efficiency (75.0 per cent) than savings and co-
operative banks (77.7 per cent and 79.4 per cent, respectively) over
the period 1995–2002. The higher mean levels of cost X-efficiency of co-
operative and savings banks over commercial banks are found to be almost
constant in every year analysed (except in 1995 and 2001).
Focusing on the four banking systems analysed, cooperative banks show
higher mean levels of cost X-efficiency than savings banks and, especially,
compared to commercial banks in Italy and Germany in almost every year
over the period analysed. German commercial banks exhibit a mean
X-efficiency similar to the other two bank categories. In France, all cate-
gories of bank display similar cost X-efficiency levels over time. By com-
paring our findings across countries, all French banks show the largest
mean cost X-inefficiency on average over the period 1995–2002, while
UK commercial banks, Italian cooperative banks and German savings
banks achieved the highest mean cost X-efficiency in their ownership
category.
Table 6.3 reports descriptive statistics of the cost X-efficiency mea-
sures derived from DEA. Focusing on the whole period 1995–2002, UK
banks display the highest mean levels of technical, allocative, scale and
(consequently) cost efficiency among the four countries analysed.
The average cost X-inefficiency over the eight years analysed is 25.0 per
cent, which seems to be equally generated by allocative and technical
inefficiency (12.6 per cent and 13.5 per cent respectively), while scale
inefficiency displays a smaller importance (5.8 per cent). As noted earlier,
the UK sample comprises only commercial banks, while other samples
Table 6.3: DEA cost efficiency mean levels in European banking between 1995 and 200214
Panel A France* Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency
Mean St.dev Mean St.dev Mean St.dev Mean St.dev

(%) (%) (%) (%) (%) (%) (%) (%)
Commercial banks 80.84 14.46 72.48 22.89 91.30 4.85 55.26 27.48
Cooperative banks 83.39 15.36 82.79 12.98 97.84 4.83 67.83 18.27
1995
Savings banks 87.92 6.00 79.30 7.31 99.00 2.90 68.87 6.64
Total 82.14 14.05 76.02 18.77 93.81 4.69 60.00 23.15
1996
Savings banks 89.35 5.69 81.09 7.04 98.95 3.45 71.60 7.20
Total 82.57 15.48 72.65 20.39 94.75 4.66 56.83 23.80
1997
Savings banks 90.45 5.62 80.54 7.59 99.11 3.37 72.14 8.00
Total 83.56 15.46 71.70 17.73 94.81 5.45 56.96 24.26
1998
Savings banks 91.02 5.75 79.18 6.02 99.06 3.45 71.43 7.65
Total 81.22 15.73 73.65 17.92 95.44 4.56 57.89 26.35
1999
Savings banks 91.10 6.73 79.02 7.83 98.90 4.41 71.09 8.43
Total 82.02 14.24 71.58 18.26 95.45 3.86 54.98 24.20
207
Table 6.3: DEA cost efficiency mean levels in European banking between 1995 and 2002 – continued
208
Panel A France* Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency

(%) (%) (%) (%) (%) (%) (%) (%)
2000
Savings banks 89.92 22.32 84.43 11.60 95.77 11.17 73.83 22.29
Total 80.35 18.77 67.75 19.24 95.56 5.09 53.21 27.37
2001
Savings banks 89.32 9.16 72.39 10.19 98.39 6.25 63.32 9.63
Total 79.70 16.46 71.49 17.26 93.14 6.52 54.43 20.62
2002
Savings banks 90.39 9.59 70.20 9.86 98.36 6.55 62.64 12.51
Total 76.49 21.85 69.86 17.94 89.76 8.59 48.59 21.19
Savings banks 89.96 9.23 78.01 8.61 98.37 5.42 69.12 10.65
Mean
Total 80.87 16.65 71.68 18.39 94.03 5.51 55.12 23.81
1995–2002
cooperative banks.
Panel B Germany* Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency
(%) (%) (%) (%) (%) (%) (%) (%)
1995
Savings banks 94.60 4.48 53.97 26.81 99.53 0.78 50.85 25.54
Total 93.14 6.34 58.73 23.02 98.96 2.65 53.71 21.92
1996
Savings banks 94.30 4.77 62.69 25.40 99.44 0.83 59.01 24.67
Total 93.90 6.59 50.78 27.50 98.56 3.06 46.60 26.85
1997
Savings banks 93.90 4.86 73.27 21.87 99.43 0.76 68.51 21.20
Total 92.57 6.93 62.99 28.58 98.97 2.54 57.55 27.39
1998
Savings banks 93.38 5.09 74.23 19.11 98.94 1.40 68.61 18.39
Total 92.52 7.05 64.55 27.25 98.50 2.86 58.51 26.13
1999
Savings banks 99.13 2.75 75.98 19.02 97.11 2.67 73.23 18.77
Total 92.12 7.27 62.93 28.78 97.11 3.20 56.62 26.85
209
210
Panel B Germany* Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency

(%) (%) (%) (%) (%) (%) (%) (%)
2000
Savings banks 95.04 4.24 50.93 39.53 99.04 1.25 47.94 37.24
Total 90.50 7.70 50.81 34.27 97.43 2.35 44.35 31.31
2001
Savings banks 94.48 4.55 79.21 17.29 99.17 1.07 74.26 16.90
Total 92.43 7.39 66.79 27.11 98.77 2.61 60.91 25.99
2002
Savings banks 91.41 7.46 80.54 15.37 96.77 2.86 71.40 15.66
Total 90.40 9.11 68.95 22.25 97.96 3.04 60.75 21.69
Savings banks 94.53 4.78 69.18 22.97 98.65 1.48 64.51 22.22
Mean
Total 92.03 7.42 61.36 27.46 98.22 2.79 55.25 26.10
1995–2002
cooperative banks.
Panel C Italy* Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency
(%) (%) (%) (%) (%) (%) (%) (%)
1995
Savings banks 86.83 10.63 97.48 3.04 95.14 5.06 80.29 9.35
Total 86.16 10.96 83.52 9.39 95.24 6.10 67.82 12.78
1996
Savings banks 89.49 9.03 96.23 3.89 95.09 4.47 81.71 8.25
Total 80.46 15.14 88.47 7.23 94.89 6.07 68.77 13.84
1997
Savings banks 88.31 9.30 96.16 4.26 94.26 4.89 79.84 8.37
Total 85.75 10.93 84.35 9.92 96.17 4.93 69.16 13.26
1998
Savings banks 86.96 9.28 95.33 4.28 89.33 4.89 73.90 8.27
Total 80.52 12.25 84.30 8.46 93.32 6.14 63.52 11.92
1999
Savings banks 85.41 10.61 94.56 4.74 86.61 4.67 69.83 9.50
Total 78.52 11.96 88.34 6.83 93.93 6.14 65.06 11.74
211
212
Panel C Italy* Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency

(%) (%) (%) (%) (%) (%) (%) (%)
2000
Savings banks 84.16 9.80 90.17 6.70 80.47 5.35 60.94 8.82
Total 73.14 12.80 85.24 9.63 89.17 8.16 54.77 10.77
2001
Savings banks 80.80 12.38 89.46 8.25 78.85 8.60 56.83 11.22
Total 68.29 13.68 82.98 9.62 88.41 9.78 49.13 11.29
2002
Savings banks 78.53 12.59 86.36 6.91 76.92 8.88 52.34 12.78
Total 67.96 13.69 86.61 8.71 88.71 10.19 50.99 11.06
Savings banks 84.99 10.47 93.13 5.30 86.91 5.88 69.20 9.60
Mean
Total 75.04 12.85 85.48 8.76 91.35 7.83 58.24 11.69
1995–2002
* The mean levels of efficiency here reported are disaggregated according to the bank specialisation. Efficiency estimates for commercial banks are
obtained by estimating an efficient frontier in a sample comprising only this category of bank, while efficiency estimates for savings and cooperative
banks are obtained by estimating a common frontier using a sample comprising both savings and cooperative banks.
Panel D United Kingdom Technical efficiency Allocative efficiency Scale efficiency Overall cost efficiency

(%) (%) (%) (%) (%) (%) (%) (%)
1995 Commercial Bank 89.10 16.37 92.66 9.42 93.65 10.32 79.75 23.11
1996 Commercial Bank 86.71 14.90 88.11 13.16 93.17 9.27 73.64 23.10
1997 Commercial Bank 86.14 17.26 88.11 12.23 93.65 11.17 73.67 24.13
1998 Commercial Bank 86.86 18.84 85.62 18.11 94.25 9.37 73.37 26.14
1999 Commercial Bank 88.70 17.58 88.19 14.29 95.02 8.56 76.77 23.93
2000 Commercial Bank 90.06 16.56 89.19 15.14 95.72 8.20 78.34 23.41
2001 Commercial Bank 85.60 23.47 84.19 20.37 93.02 16.09 71.72 27.66
2002 Commercial Bank 86.68 21.16 85.72 18.57 94.56 12.94 73.90 26.43
Media Commercial Bank 86.47 20.80 87.40 16.11 94.20 11.20 74.97 24.98
1995–2002
213
selected include banks with different specialisations (as such, cooperative

and savings banks). The other three countries analysed display similar
mean levels of cost X-efficiency that are substantially lower than in the UK,
Italian banks show a mean cost X-efficiency level of 58.2 per cent, which is
due to a mean technical inefficiency of 25.0 per cent, to an allocative
inefficiency of 14.5 per cent and to a scale inefficiency of 8.6 per cent. In
Germany, the average cost efficiency over the eight years analysed is 55.3
per cent, which seems to be generated mainly by allocative inefficiency
(38.6 per cent), while technical and scale inefficiency are smaller (8.0 per
cent and 1.8 per cent respectively). It is worthwhile noting that mean levels
of efficiency in Germany are strongly influenced by cooperative banks that
are numerically dominant in Germany. French banks display a mean level
of cost X-efficiency of 55.1 per cent, which is a result of mean allocative
inefficiency of 28.3 per cent, technical inefficiency of 19.1 per cent and
scale inefficiency of 6.0 per cent.
By considering banks’ specialisation, commercial banks seem to be sub-
stantially less efficient (for all cost efficiency components, i.e. technical,
allocative and scale efficiency) than cooperative and savings banks in
France and Italy over all eight years considered.15 In detail, French and
Italian commercial banks achieved a mean cost efficiency level between
1995 and 2002 of 48.4 per cent and 49.9 per cent, respectively, while co-
operative and savings banks achieved substantially higher levels of cost
efficiency over the same period. Although Italian and French banking
systems present similar results, there are different reasons for the superior-
ity of cooperative and savings banks. In France, commercial banks are
found to be less technical, allocative and scale efficient than cooperative
and savings banks. Instead, Italian commercial banks seem to suffer lower
mean cost efficiency levels due to high levels of allocative inefficiency (in
some years, larger than 40 per cent) than cooperative and savings banks,
while commercial banks usually exhibit higher mean levels of technical
efficiency. Italian commercial banks also show substantial opportunity for
cost efficiency gains by reducing scale inefficiency (around 9 per cent
between 1995 and 2002). In Germany, commercial banks exhibit a higher
cost efficiency than savings and, especially, cooperative banks over the
period 1995–2002 (namely, 69.5 per cent over 64.5 per cent and 47.4 per
cent, respectively). On average, German commercial banks seem to be
slightly less technically and scale efficient than the other two categories of
banks, but display substantially larger allocative efficiency. As a possible
explanation of the larger cost inefficiency of German cooperative banks, it
is possible to note that the number of cooperative banks in Germany is at
least four times the number of commercial banks. Since DEA compare all
banks in the sample, as the sample size increases, the probability of finding
more efficient banks increases and, consequently, the average distance of
inefficient banks increases.
It is worthwhile noting that mean level of cost inefficiency estimates

(obtained by running both SFA and DEA models) express the mean dis-
tance of all inefficient banks from the efficient frontier: these are estimated
using a cross-section sample and, as such, change every year. As such, all
mean cost efficiency levels reported in Tables 6.2 and 6.3 display the
average distance from different efficient frontiers, rather than a common
frontier, and consequently these results should not been interpreted as
clear evidence of a productivity improvement in a country; for example, a
decrease in the average distance of inefficient banks from the frontier (e.g.
France from 1998 to 1999) may be also due to worsening management
practices in the current period of these banks that were best-practice in the
previous period (i.e. a negative shift of the efficient frontier) rather than an
efficiency improvement. In other words, the lower mean cost X-efficiency
levels found over the period 1995–2002 express a larger average distance
from the French cross-sectional efficient frontiers than the average distance
of Italian, German and UK banks from their cross-section efficient frontiers.
However, this does not imply that bank productivity in France is lower
than in Germany, Italy and the UK since cost X-efficiency does not enable
us to comparatively analyse the domestic efficient frontiers. Later on, we
estimate productivity change using the Malmquist Total Factor
Productivity (TFP) index and this measure enables us to distinguish
between an efficient frontier shift (i.e. technical change) and the variation
of the distance of inefficient banks from the frontier (efficiency change).
Our results appear to be consistent with previous studies. Berger and
Humphrey (1997) summarised over 120 studies in banking and found an
average efficiency of 79 per cent. Distinguishing between parametric and
non-parametric studies, the authors found an average efficiency of 84 per
cent and 72 per cent,16 respectively. Our mean efficiency estimates appear
slightly lower than Berger and Humphrey (1997) mean levels.17 Our overall
(for all four countries analysed) mean level of X-cost efficiency estimated
using SFA between 1995 and 2002 is 78.1 per cent and domestic mean
levels of the four countries ranged between 72.4 per cent (France) and 80.9
per cent (UK). Consistent with Berger and Humphrey (1997), our efficiency
levels estimated with DEA are lower than those estimated with SFA: mean
technical efficiency levels range between 75.0 per cent (Italy) and 92.0 per
cent (Germany), mean allocative efficiency levels range between 61.36 per
cent (Germany) and 87.44 per cent (UK), mean scale efficiency levels range
between 91.35 per cent and 98.2 per cent and mean overall efficiency levels
range between 55.1 per cent (France) and 75.0 per cent (UK).
In addition, Berger and Humphrey (1997) defined a sort of confidence
interval formed by the mean plus and minus one standard deviation
(between 66 per cent and 92 per cent) that captured 82 per cent of the
observations summarised in their study. Looking at our findings, all our
domestic and annual mean levels estimated using SFA (Table 6.2) are inside
the Berger and Humphrey (1997) interval. Next, we identify our own
domestic confidence interval by adding ± one standard deviation to each
country mean value over the period 1995–2002. As such, we define a
specific confidence interval per each country (France, Germany, Italy and
UK) and for each kind of efficiency estimate (cost X-efficiency, technical
efficiency, allocative efficiency, scale efficiency and overall efficiency). We
found that 90 per cent of mean levels estimated using SFA and 86.5 per
cent of mean levels estimated using DEA are captured in their respective
confidence intervals.
Among more recent studies dealing with the banking systems analysed in
this book, we note that Resti (1997c) and Casu and Girardone (2004b)
focus on Italy, Glass and McKillop (2000) and McKillop et al. (2002) inves-
tigate UK financial institutions, Dietsch and Lozano-Vivas (2000) analyse
French banks, Altunbas et al. (2000a) focus on German banking, while
Carbo et al. (2000b) analyse savings banks of several European countries,
Altunbas et al. (2001) assess cost efficiency for a large sample of European
banks and Beccalli et al. (2003) assess the efficiency of a sample of
European listed banks.
Regarding France, we estimated over the period 1995–2002 a mean cost
X-inefficiency of 27.6 per cent, technical inefficiency of 19.1 per cent,
allocative inefficiency of 28.3 per cent, scale inefficiency of 6.0 per cent and
overall inefficiency of 44.9 per cent. Our findings seem to be consistent
with previous studies. In estimating a French domestic frontier, Dietsch
and Lozano-Vives (2000) found that the mean level of cost X-inefficiency
of commercial banks between 1988 and 1992 is 22.5 per cent. Altunbas et
al. (2001) estimate that French commercial banks reduced their cost
X-inefficiency between 1989 and 1997 (from 28.8 per cent to 24.4 per
cent). Focusing on listed banks, Beccalli et al. (2003) found mean technical
inefficiency levels (using DEA) in 1999 and 2000 to be 29.4 per cent and
28.0 per cent, respectively, and mean cost X-inefficiency (using SFA) in
1999 and 2000 to be 14.9 per cent and 17.3 per cent, respectively. Focusing
on savings banks between 1989 and 1996, Carbo et al. (2000b) estimated a
mean scale inefficiency of 12.1 per cent and a mean cost X-inefficiency of
23.1 per cent. Our findings on French savings banks differ slightly from
these results: we estimate smaller scale inefficiency (1.6 per cent) and larger
cost X-inefficiency (26.8 per cent).
In Germany, we estimated over the period 1995–2002 a mean X-
inefficiency of 21.4 per cent, technical inefficiency of 8.0 per cent, alloca-
tive inefficiency of 28.6 per cent, scale inefficiency of 1.8 per cent and
overall inefficiency of 44.7 per cent. Altunbas et al. (2000a), using SFA and
DFA, found a mean cost X-efficiency level of 83.8 per cent between 1989
and 1996 and found that public and mutual banks have slight cost advan-
tages because of a possible advantage in terms of the lower cost of funds.
Altunbas et al. (2001) estimate that German commercial banks reduced
their mean X-inefficiency between 1989 and 1997 (from 21.8 per cent to
13.5 per cent). Beccalli et al. (2003) found that the mean technical
inefficiency levels (using DEA) of German listed banks in 1999 and 2000 is
6.7 per cent and 13.8 per cent, respectively, and mean cost X-inefficiency
(using SFA) in 1999 and 2000 to be 14.3 per cent and 18.1 per cent, respec-
tively. Focusing on savings banks between 1989 and 1996, Carbo et al.
(2000b) estimated a mean scale inefficiency of 7.4 per cent and a mean cost
X-inefficiency of 21.3 per cent. Our findings on Italian savings banks are
similar to these results, although we estimate substantially smaller scale
inefficiency of 1.4 per cent, and our cost X-inefficiency results of 22.0 per
cent are broadly the same.
In Italy, over the period 1995–2002 banks’ measured X-inefficiency stood
at 20.7 per cent, technical inefficiency at 25.0 per cent, allocative
inefficiency at 15.5 per cent, scale inefficiency at 8.7 per cent and overall
inefficiency at 41.8 per cent. Resti (1997a) estimates using SFA that mean
levels of productive efficiency of Italian banks between 1988 and 1992 were
around 69.5 per cent (ranging between 69.4 per cent and 69.8 per cent) and
mean levels of technical efficiency around 74.0 per cent (ranging between
73.4 per cent and 75.7 per cent). Altunbas et al. (2001) estimate that Italian
commercial banks improved their cost X-inefficiency between 1989 and
1997 reducing their mean inefficiency from 21.7 per cent to 12.6 per cent.
Beccalli et al. (2003) found that the mean technical inefficiency levels
(using DEA) of Italian listed banks in 1999 and 2000 is 20.0 per cent and
23.8 per cent, respectively, and mean cost X-inefficiency (using SFA) in
1999 and 2000 is 9.8 per cent and 11.5 per cent, respectively. Focusing on
savings banks between 1989 and 1996, Carbo et al. (2000b) estimated a
mean scale inefficiency of 12.1 per cent and a mean cost X-inefficiency of
22.5 per cent. Our findings on Italian savings banks are similar to these
results: we estimate smaller scale inefficiency of 13.1 per cent and cost
X-inefficiency of 22.4 per cent.
In the UK, over 1995–2002 and for a sample comprising only commercial
banks, mean levels of X-inefficiency, technical, allocative, scale and overall
inefficiency were 19.0 per cent, 13.5 per cent, 12.6 per cent, 5.8 per cent
and 25.0 per cent, respectively. Altunbas et al. (2001) estimate that the
mean X-inefficiency of UK commercial banks was stable around 30 per cent
between 1989 and 1997 (ranging from 28.9 per cent to 33.3 per cent).
Beccalli et al. (2003) found that the mean technical inefficiency levels
(using DEA) of UK listed banks in 1999 and 2000 is 16.7 per cent and 19.0
per cent, respectively, and mean cost X-inefficiency (using SFA) in 1999
and 2000 is 19.3 per cent and 20.8 per cent, respectively.
In conclusion, it is worthwhile noting that although it is not possible to
directly compare our findings with previous studies (since these studies
assess different time periods, use different definitions of bank inputs and
outputs, select different samples and use partly different investigation tech-
niques), our findings are broadly similar to those found in previous empir-
ical literature.
Measuring profit efficiency

As noted in Chapter 4, profit efficiency is a more comprehensive efficiency
concept than cost efficiency because it embodies both cost and revenue
components. Two profit maximisation concepts have developed in the
literature depending on which business conditions are taken as given:
standard profit efficiency expresses the bank’s ability to produce at the
maximum possible profit given a particular level of input prices and output
prices (and other variables), while ‘alternative profit efficiency’ refers to the
bank’s ability to produce at the maximum possible profit given a particular
level of output levels, rather than its output prices. Since output prices are
not completely reliable, the standard profit function will be imprecisely
estimated and, as such, we estimate the alternative profit function to drive
own profit efficiency estimates.
Methodology for estimating alternative profit efficiency

The alternative profit function is estimated using the standard translog
functional form:
n m
ln(π + θ) = α0 + ∑ αi ln yi + ∑ βj ln wj + (6.9)
i=1 j=1
 n m m m 
1 
+  ∑ ∑ δ ln yi ln yj + ∑ ∑ γij ln wi ln wj  +
ij
2  i=1 j=1 i=1 j=1 
n m
i=1 j=1
where π is published net income standardised by w3 (i.e. the average cost of

capital measured as total capital expenses over total fixed assets), θ is a con-
stant, yi is the output quantities for the i-th output, wi is the input prices for
the i-th input, n is the number of outputs, m is the number of inputs.
outputs (that we already adopted running SFA and DEA for measuring cost
efficiency), we posit18 that labour (measured as personnel expenses), physi-
cal capital (expressed as the average value of fixed-tangible assets) and
financial capital (measured as loanable funds) are inputs, whereas demand
deposits, total loans and other earning assets are outputs (see Figure 6.2).
This specification is essentially the same as that adopted to estimate cost
efficiency as outlined in the previous section. The dependent variable for
the profit function replaces the normalised lnTC with ln (π+θ), where π is
published bank’s net income (standardised by the average cost of capital)
and, following Casu and Girardone (2004b), θ is a constant, defined by
adding 1 to the absolute value of the lowest (π/w3) in the sample (i.e.
θ = |(PT/w3)min | + 1), in order to make positive the natural log of bank’s
profits. The standard Stochastic Frontier Analysis (SFA) is employed to esti-
mate X-efficiency for each bank over the years 1995–2002: the Battese and
Coelli (1992) model of a stochastic frontier function has been adopted.
π – Published bank’s net income

θ is a constant, i.e. defined as θ = | ( π / w3)min | + 1]
Input 3 (w3) – average cost of financial capital deposits (interest costs on borrowed funds on the
average amount of borrowed funds);
Figure 6.2: Inputs and outputs used for estimating alternative profit efficiency19
Sample description
The sample adopted for estimating alternative profit efficiency is the same
employed for estimating cost efficiency using SFA (Table 6.1, panel A).
As such, we estimate profit efficiency frontiers using annual cross-section
estimates for the eight years (from 1995 to 2002), for two bank categories
(firstly, commercial banks and, secondly, cooperative and savings banks)
and for three countries (France, Italy and Germany). For the UK, we estim-
ate an annual cross-section efficient frontier for eight years (1995–2002)
using a sample comprising only commercial banks.20 Table 6.1, panel A
provides the number of banks considered for estimating each frontier using
SFA.
We use a cross-section sample by year since many bank observations
would have been lost selecting a balanced panel data set. We prefer to use a
domestic sample for estimating the profit efficiency frontier since banks in
the same country are more homogeneous (and comparable) than banks
operating in different countries. As with our cost efficiency estimates, we
consider various sub-samples according to bank category (namely, commer-
cial banks and, jointly, cooperative and savings banks21) since this seems to
guarantee a higher homogeneity to the sample. As such, we estimated fifty-
six alternative profit efficiency frontiers.22 To reduce the risk of including
outliers, all data outside the range of ± 3 standard deviations from the
median are deleted.
Results
Table 6.4 reports descriptive statistics of the alternative profit efficiency
measures derived from SFA. Overall, European banks display over the
period analysed profit inefficiency scores ranging between 19.0 per cent
(UK commercial banks in 1996) to 46.8 per cent (Italian commercial banks
in 1995) and the level of dispersion of average efficiency scores varies sub-
stantially from 8.7 per cent (Italian cooperative banks in 2002) to 31.8 per
cent (Italian commercial banks in 1995). On average (considering all
20,942 bank profit efficiency estimates obtained running fifty-six annual
Table 6.4: Alternative profit efficiency scores in European banking between 1995 and 2002*
220
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
1995
Total 67.95 19.76 70.15 16.09 61.31 24.04 79.27 13.38 68.98 17.58
1996
Total 66.68 20.96 64.32 14.31 67.57 18.17 81.01 13.48 65.58 15.69
1997
Total 63.13 18.72 64.89 16.75 65.81 15.60 77.89 17.99 65.16 16.92
1998
Total 71.99 15.78 66.11 14.21 72.11 16.32 71.47 24.49 68.06 15.09
1999
Total 62.08 18.33 71.10 14.74 61.06 17.57 74.45 14.97 68.17 15.72
Table 6.4: Alternative profit efficiency scores in European banking between 1995 and 2002* – continued
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
2000
Total 63.34 16.08 61.46 15.27 65.49 14.79 72.03 20.73 62.81 15.40
2001
Total 63.47 15.92 61.02 15.61 65.60 14.94 72.90 17.99 62.56 15.56
2002
Total 64.23 16.90 62.55 17.95 67.35 12.75 68.93 22.73 63.94 16.83
Mean
Total 65.26 17.68 64.81 15.68 65.86 15.84 73.94 19.25 65.32 16.05
1995–2002
cooperative banks.
221
cross-section domestic frontiers), the mean profit X-efficiency level over the
period 1995–2002 is 65.3 per cent and the mean level of dispersion of
average efficiency scores is 16.1 per cent.
Focusing on the four banking systems analysed, UK banks display the
highest profit efficiency levels (on average, 73.9 per cent over the period
1995–2002) and the highest dispersion of average efficiency scores (on
average, 19.3 per cent) compared to the other three countries. Over the
period analysed, mean efficiency levels in UK banking substantially
decreased between 1996 and 1999 (from 81.0 per cent to 71.5 per cent) and
between 2001 and 2002 (from 72.9 per cent to 68.9 per cent). It is worth-
while noting that the UK sample comprises only commercial banks. The
other three European banking systems display similar mean levels of profit
efficiency and of dispersion of average efficiency scores. Italian banks
exhibit a mean profit efficiency of 65.9 per cent and a level of dispersion of
average efficiency scores of 15.8 per cent. In France and Germany, the
mean profit efficiency levels are 65.3 per cent and 64.8 per cent, respec-
tively, and a level of dispersion of average efficiency scores of 17.7 per cent
and 15.7 per cent, respectively.
Looking at the different bank specialisations, mean profit efficiency levels
show slight differences across bank specialisation. On average, the mean
profit efficiency between 1995 and 2002 of commercial banks is 66.4 per
cent, for cooperative banks 64.8 per cent and for savings banks 65.6 per
cent. Commercial banks show a larger dispersion of average efficiency
scores than cooperative and savings banks. Although mean profit efficiency
levels are found to be similar for the three countries analysed, there are
substantial differences among annual mean profit efficiency levels of com-
mercial, savings and cooperative banks.
As a result, we found that, on average, banks in the four European
banking systems analysed achieve only two-thirds of their potential profits.
UK banks are found to be the most profit efficient with a mean of 73.9 per
cent, while French, German and Italian are found to have similar levels of
profit efficiency (around 65 per cent).
Our profit efficiency estimates appear to be consistent with previous

studies. Berger and Humphrey (1997) found that the mean profit efficiency
of US depository institutions is 64 per cent, summarising various previous
studies. In general, results from the previous literature on profit efficiency
vary considerably. For example, Akhavein et al. (1997) estimate using DFA
that large merging US banks have a mean profit efficiency of 24 per cent
before merging and of 34 per cent after merging, while Miller and Noulas
(1996) estimate using DEA a mean profit efficiency of 97 per cent. In the
case of European banking, Lozano-Vivas (1997) estimates that Spanish
banks earn on average 72 per cent of their potential profits. More recently,
Yildirim and Philippatos (2002) estimated alternative cost and profit
efficiency levels in Eastern European countries and found that profit

efficiency is significantly lower relative to cost efficiency. The authors esti-
mate that using SFA approximately one-third of banks’ profits are lost to
inefficiency and almost one-half when the DFA approach is used. Bos and
Schmiedel (2003) estimate various sets of profit efficiency estimates for
most European countries. The authors found that profit efficiency in France
is around 47.0 per cent, in Germany it varies between 37.7 per cent and
39.7 per cent, in Italy it ranges between 46.1 per cent and 54.4 per cent, in
the UK it varies between 54.6 per cent and 66.7 per cent. Regarding Italy,
Casu and Girardone (2004b) estimate alternative profit efficiency using SFA
with a tranlog profit function and, in contrast to Bos and Schmiedel (2003),
found profit efficiency ranging between 77.3 per cent and 85.8 per cent
over the period 1995–9. In general, therefore, like our cost efficiency mea-
sures, profit efficiency estimates are broadly in accordance with the previ-
ous literature.
Measuring productivity changes

Despite the fact that efficiency and productivity do not express the same
concepts, these terms have been frequently used as synonymous in the lit-
erature. As noted by Lovell (1993), the productivity of a production unit is
expressed as the ratio of its outputs to its inputs and it is determined by the
production technology, efficiency and the environment. Corporate
efficiency is only a determinant of productivity, which can be defined as
the comparison between observed and optimal values of a firm’s inputs and
outputs. Efficiency measures (especially, cost efficiency) express the mean
distance of inefficient banks from an efficient frontier, i.e. estimated using
cross-section samples and, as such, these change every year. All com-
parisons among mean efficiency levels obtained in different countries
or by different kinds of bank or in different periods focus on the mean
distance of inefficient banks from their own efficient frontier. As such,
the expression ‘more/less cost efficient’ means that the mean distance of
inefficient banks from their own efficient frontier is smaller/larger than
the mean distance of inefficient banks of another category/country. A
lower/superior mean cost efficiency does not provide any evidence of a
lower/superior corporate productivity that considers both the mean dis-
tance from the efficiency frontier and the variation of the efficient frontier
over time.
In order to show the difference between technical efficiency and pro-
ductivity, Figure 6.3 (Panel A) shows that in a production process with a
single input and a single output not all technically efficient firms have the
same productivity. Whilst the productivity of a firm refers to the ratio of
the output that it produces to the input that it uses,23 technical efficiency
expresses the ability of a firm to obtain maximal outputs from a given set
Panel A
y
C• F’
B• D•
A•
Panel B
y
C• F’ x
B• D•
A•
Figure 6.3: A graphical example of firm productivity and efficiency
of inputs or of minimising inputs for a given target of outputs. Graphically,

the slope of the ray through the origin expresses a firm’s productivity,
while the distance from the efficient frontier expresses technical efficiency.
Figure 6.3 (Panel B) shows a situation where firms A, B and C are tech-
nically efficient, but B defines the point of maximum productivity (point of
technically optimal scale). For example, firms A and C may increase their
productivity by varying their dimensions and exploiting scale economies.
In this sense, technical efficiency estimates can be used to construct mea-
sures of scale efficiency by comparing the observed and the optimal scale of
the firm.
Methodology for estimating Total Factor Productivity

This section describes the methodological approach followed to estimate
Total Factor Productivity (TFP) growth (i.e. the change in TFP between two
consecutive periods where TFP is an index of output divided by an index of
total input usage). In detail, we estimate the Malmquist TFP index that
measures the TFP change between two data points by calculating the ratio
of the distances of each data point relative to a common technology. Let us
consider two periods (t and s) and denote the output in each period as yt
and ys, the inputs employed in each period as xt and xs. For each time
period, let the production set St model the transformation of inputs into
outputs. In an output-orientation, the Malmquist TFP change index
[M0 (yt,, Xt,, ys,, Xs)] between s (the base period) and t is given by:
dt0 (yt, xt) ds0 (yt, xt) ds0 (ys, xs)

Mo(ys, xs, yt, xt) = (6.10)
ds0 (ys, xs) dt0 (yt, xt) dt0 (ys, xs)
where the notation ds0 (yt , xt) represents the output distance between the
period t observation and s technology. A value greater than 1 will indicate
positive TFP growth from s and t, whilst a value lower than 1 indicates a
decline. This formulation of the Malmquist index allows us to distinguish
two components of TFP change. The first (i.e. represented by the ratio
outside the brackets in model 6.10) is the Efficiency Change (EC) that mea-
sures the change in the output-oriented measure of Farrell technical
efficiency between the period s and t. The second (i.e. represented by the
expression under the squared root in model 6.10) is the Technical Change
that is the geometric mean of the shift of the frontier between s and t.
We estimate the distances of each data point relative to a common tech-
nology in TFP change using DEA24 running the DEA-like linear program-
ming method. Using this method, originally proposed by Fare et al. (1994),
we calculate the measures previously stated by solving four DEA models
under the assumption of Constant Returns to Scale following Griffell-Tatjé
and Lovell (1995) who note that the TFP is not accurately estimated by
assuming Variable Returns to Scale for the technology. Following Fare et al.
(1994), we propose the ‘enhanced decomposition’ by decomposing the
efficiency change (EC) into scale efficiency (SCC) and Pure Technical
Efficiency (PTEC) components. It is important to note that all these DEA
models require data for input-output quantities, but not price information.
The decomposition becomes:
Mo (yt, xt, ys, xs) = EC × TC = PTEC × SCC × TC (6.11)
Similarly to the estimation of cost efficiency using DEA, we adopt the

value-added approach to identify bank outputs running the DEA-like linear
programming method. As such, we posit that labour (price of labour is

measured as personnel expenses over total assets), physical capital (price of
physical capital is expressed as the total capital equipment expenses over
total fixed assets) and financial capital (price of financial capital is mea-
sured as interest costs on borrowed funds) are inputs, whereas demand
Sample description
In estimating the Malmquist productivity index, it is necessary to adopt a
balanced panel data set. Looking at the available data, although the
number of banks is quite large, this number will have to be substantially
reduced if a balanced panel is to be considered over the period 1995–2002.
Because a panel data set of matched pairs of observations for each pair of
consecutive years is required to estimate TFP change, we calculate the
Malmquist indices for seven two-year balanced panel data samples
(namely, 1995–6, 1996–7, 1997–8, 1998–9, 1999–2000, 2000–1, 2001–2),
for two bank specialisations (i.e. commercial banks and, jointly, coopera-
tive and savings banks) and for three countries (i.e. France, Germany,
Italy). In the UK, we considered only commercial banks. As a result, we esti-
mated fourteen Malmquist productivity indices (i.e. seven years per two
bank categories) for France, Germany and Italy, respectively, and seven
indices for the UK (i.e. seven years per one bank category). As such, we esti-
mated forty-nine Malmquist indices. A specific DEA-like linear program-
ming model has been run for each sample.25 Table 6.5 reports the number
of banks in the sample.
Results
Table 6.6 displays productivity change estimates. In detail, TFP changes,
technical efficiency changes, technical changes, scale efficiency changes
and pure technical efficiency changes are reported for banks in different
countries and by specialisation type.
Input 1 (X1) – labour (measured as average number of employees);

Input 2 (X2) – physical capital (expressed as the average value of fixed-tangible assets);
Input 3 (X3) – financial capital (measured as loanable funds);
Output 3 (Y3) – other earning assets;
Price Input 1 – average labour costs (total labour costs on number of employees);
Price Input 2 – total equipment capital expenses/total fixed-tangible assets;
Price Input 3 – interest costs on borrowed funds on the average amount of borrowed funds.
Figure 6.4: Inputs and outputs for estimating Total Factor Productivity change in
European banking between 1995 and 2002
Table 6.5: Sample used for estimating Total Factor Productivity in European banking between 1996 and 2002
Panel B DEA 1995–6 1996–7 1997–8 1998–9 1999–2000 2000–1 2001–2
Commercial banks 208 206 207 206 206 231 233

Cooperative & 102 103 112 120 116 121 119
France
savings banks* (80;22) (81;22) (90;22) (95;25) (91;25) (96;25) (92;27)

Cooperative & 893 1273 1414 1534 1725 1853 2022
savings banks* (593;300) (721;552) (842;572) (945;589) (1121;604) (1241;612) (1404;618)
Germany
Cooperative & 172 199 335 335 459 546 586
Italy
savings banks* (110;62) (136;63) (271;64) (271;64) (394;65) (480;66) (520;66)

Cooperative & 0 0 0 0 0 0 0
United
savings banks*
Kingdom
Source: Bankscope.
227
Between 1995 and 2002, the largest TFP improvement seems to have
been achieved by French commercial banks (+41 per cent from the mean
productivity level in 1995), while the highest TFP deterioration over the
same period was obtained by French savings banks (–52 per cent from the
1995 level). Regarding French commercial banks, TFP improvement appears
mainly due to substantial technical efficiency enhancement (54 per cent)
from the 1995 level, due to a great improvement in pure technical
efficiency (43 per cent) and a slight advance in scale efficiency (9 per cent),
while technical change slightly improved (1 per cent). According to these
results, the efficient frontier did not move considerably between 1995 and
2002, while the mean distance of inefficient banks from the efficient fron-
tier reduced by almost 50 per cent and commercial banks were able to
reduce inputs by 9 per cent by exploiting scale efficiency opportunities.
Concerning French savings banks, TFP at the end of 2002 deteriorated by
52 per cent from the 1995 level and this was due both to a worsening
(namely, by 17 per cent of 1995 level) of the efficient frontier and because
inefficient savings banks moved further away from the efficient frontier.
Regarding the other three countries, it is generally possible to observe
TFP improvements over the period 1995–2002 (Figure 6.5). This improve-
ment is substantial for UK commercial banks (due to improved pure techni-
cal efficiency change and, also, to their better ability to exploit scale
efficiency), for Italian commercial banks (mainly due to technical change),
for German commercial banks (primary due to substantially improved pure
technical efficiency and a reduction in input waste) and for Italian cooper-
ative banks (mainly due to pure technical efficiency change). A substantial
regress of the efficient frontier from 1995 to 2002 generated a TFP decrease
for Italian savings banks and French cooperative banks The analysis of
annual TFP changes over the period 1995–2002 enable us to note that the
overall TFP improvement achieved by French commercial banks between
1995 and 2002 was mainly generated in 2000, when TFP improved by 27
per cent in comparison to the 1999 level, while TFP appears stable in the
other years. The TFP advance between 1999 and 2000 was generated by a
progress of 33 per cent by efficient commercial banks in comparison to
efficient banks in 1999 (i.e. technical change in 1999–2000) and inefficient
banks remained almost constant (–5 per cent) the mean distance from the
improved efficient frontier. Looking at French cooperative and savings
banks, it is possible to note that the substantial TFP reduction observed
between 1995 and 2002 is mainly generated between 1995 and 1997. At
the end of 1997, TFP reduced by 40 per cent for cooperative banks and by
50 per cent for savings banks compared to 1995 levels.
From 1997 up to 2002, TFP change is quite stable. The TFP of French
cooperative banks is 3 per cent lower than in 1997 and the TFP of French
savings banks is 1 per cent lower than in 2000. Regarding German banks, it
is possible to note that cooperative and savings banks have a stable TFP
Table 6.6: Total Factor Productivity estimates in European banking between 1995 and 2002
Panel A France* EFFCH TECHCH PECH SECH TFPCH
Mean St.Dev Mean St.Dev Mean St.Dev Mean St.Dev Mean St.Dev
Commercial Bank 1.03 0.33 0.99 0.09 1.02 0.20 1.01 0.22 1.02 0.34
Cooperative Bank 1.00 0.07 0.98 0.10 0.99 0.04 1.01 0.05 0.99 0.19
Savings Bank 0.99 0.02 0.99 0.00 0.99 0.01 1.00 0.01 0.98 0.02
1995–6
Total 1.02 0.24 0.99 0.09 1.01 0.15 1.01 0.16 1.01 0.28
Commercial Bank 1.04 0.47 0.99 0.08 1.06 0.48 0.99 0.12 1.02 0.48
Savings Bank 0.77 0.14 0.91 0.00 0.77 0.14 0.99 0.16 0.70 0.07
1996–7
Total 0.96 0.37 0.96 0.09 0.99 0.37 0.98 0.13 0.93 0.37
Commercial Bank 0.97 0.20 1.02 0.10 0.95 0.15 1.02 0.14 0.99 0.25
Savings Bank 0.80 0.16 0.92 0.00 0.80 0.16 0.98 0.13 0.73 0.11
1997–8
Total 0.93 0.19 0.99 0.10 0.93 0.16 1.01 0.14 0.92 0.22
Commercial Bank 1.00 0.19 1.01 0.13 1.00 0.25 1.01 0.09 1.01 0.22
Savings Bank 1.00 0.02 1.00 0.00 1.00 0.01 1.00 0.01 1.00 0.02
1998–9
Total 1.00 0.14 1.01 0.10 1.00 0.18 1.01 0.07 1.01 0.16
Commercial Bank 1.01 0.44 1.33 0.61 0.99 0.36 1.01 0.15 1.27 0.60
Savings Bank 1.02 0.15 1.03 0.01 0.96 0.04 1.05 0.15 1.04 0.15
Total 1.02 0.35 1.22 2.36 0.99 0.29 1.02 0.14 1.20 2.42
1999–2000
229
Table 6.6: Total Factor Productivity estimates in European banking between 1995 and 2002 – continued
230
Panel A France* EFFCH TECHCH PECH SECH TFPCH
Commercial Bank 1.05 0.22 0.97 0.54 1.07 0.22 0.99 0.14 1.02 0.57
Savings Bank 0.97 0.20 0.97 0.01 1.00 0.18 0.97 0.09 0.94 0.20
2000–1
Total 1.02 0.21 0.97 0.39 1.04 0.21 0.99 0.12 0.99 0.45
Commercial Bank 1.40 0.46 0.79 0.29 1.31 0.44 1.06 0.29 1.05 0.46
Savings Bank 0.99 0.02 1.00 0.00 0.99 0.01 1.00 0.01 0.99 0.02
2001–2
Total 1.00 0.19 1.01 0.13 1.00 0.25 1.01 0.09 1.01 0.22
Commercial Bank 1.06 – 1.00 – 1.05 – 1.01 – 1.05 –

Cooperative Bank 0.95 – 0.97 – 0.96 – 0.99 – 0.93 –
Savings Bank 0.93 – 0.97 – 0.93 – 1.00 – 0.90 –
95–02
MEAN+
Total 0.99 1.02 0.99 1.00 1.01
Commercial Bank 1.54 – 1.03 – 1.43 – 1.09 – 1.41 –

Savings Bank 0.60 – 0.83 – 0.58 – 0.99 – 0.48 –
95–02
OVERALL
Total 0.95 1.14 0.96 1.03 1.05
Where TFPCH is Total Factor Productivity change, EFFCH is (technical) efficiency change, TECHCH is technological change, PECH is pure (technical)
efficiency change and SECH is scale efficiency change
+ Geometric mean
Panel B Germany* EFFCH TECHCH PECH SECH TFPCH
Commercial Bank 0.99 0.10 1.01 0.13 1.00 0.09 0.99 0.05 1.00 0.13
Savings Bank 1.02 0.02 0.99 0.03 1.01 0.01 1.01 0.01 1.00 0.03
1995–6
Total 1.01 0.03 0.99 0.06 1.01 0.03 1.00 0.02 1.00 0.06
Commercial Bank 1.01 0.06 0.99 0.04 1.00 0.05 1.01 0.03 1.00 0.07
Savings Bank 1.01 0.02 1.00 0.06 1.00 0.02 1.01 0.02 1.01 0.07
1996–7
Total 1.01 0.03 0.99 0.04 1.00 0.02 1.01 0.02 1.00 0.05
Commercial Bank 1.01 0.21 1.00 0.03 1.09 0.31 0.99 0.09 1.00 0.21
Savings Bank 1.00 0.02 1.00 0.00 1.00 0.01 1.00 0.01 1.00 0.02
1997–8
Total 1.01 0.04 1.00 0.02 1.01 0.18 1.00 0.03 1.00 0.06
Commercial Bank 1.02 0.10 0.97 0.06 1.02 0.10 1.01 0.03 0.99 0.10
Savings Bank 1.00 0.02 1.00 0.02 1.00 0.01 1.00 0.02 1.00 0.04
1998–9
Total 1.00 0.03 0.99 0.02 1.00 0.02 1.00 0.02 1.00 0.03
Commercial Bank 0.88 0.22 1.16 0.11 1.00 0.09 0.90 0.49 1.01 0.25
Savings Bank 1.03 0.03 0.97 0.03 1.01 0.02 1.01 0.03 1.00 0.02
Total 1.00 0.07 1.00 0.05 1.01 0.05 0.99 0.08 1.00 0.07
1999–2000
231
232
Panel B Germany* EFFCH TECHCH PECH SECH TFPCH
Commercial Bank 1.01 0.08 1.00 0.03 1.01 0.08 1.00 0.04 1.01 0.07
Savings Bank 0.96 0.04 1.03 0.04 0.99 0.03 0.98 0.03 1.00 0.03
2000–1
Total 0.96 0.05 1.04 0.04 0.97 0.04 0.99 0.03 1.00 0.04
Commercial Bank 1.08 0.36 0.98 0.12 1.03 0.22 1.04 0.18 1.05 0.39
Savings Bank 1.00 0.03 1.00 0.01 1.00 0.02 1.00 0.01 1.00 0.03
2001–2
Total 1.01 0.06 1.00 0.03 1.00 0.04 1.00 0.03 1.01 0.07
Commercial Bank 1.00 – 1.01 – 1.02 – 0.99 – 1.01 –

Savings Bank 1.00 – 1.00 – 1.00 – 1.00 – 1.00 –
95–02
MEAN+
Total 1.00 – 1.00 – 1.00 – 1.00 – 1.00 –
Commercial Bank 0.99 – 1.10 – 1.16 – 0.94 – 1.06 –

Savings Bank 1.02 – 0.99 – 1.01 – 1.01 – 1.01 –
95–02
OVERALL
Total 1.00 – 1.01 – 1.00 – 0.99 – 1.01 –
+ Geometric mean
Panel C Italy* EFFCH TECHCH PECH SECH TFPCH
Commercial Bank 1.02 0.08 0.99 0.15 1.01 0.07 1.00 0.04 1.01 0.17
Savings Bank 1.01 0.04 0.99 0.03 1.01 0.04 1.00 0.03 1.00 0.04
1995–6
Total 1.01 0.05 0.99 0.05 1.01 0.05 1.00 0.03 1.00 0.07
Commercial Bank 0.96 0.06 1.03 0.06 0.98 0.06 0.98 0.03 1.00 0.08
Savings Bank 0.99 0.04 1.04 0.02 1.00 0.03 0.99 0.03 1.03 0.04
1996–7
Total 0.99 0.05 1.04 0.03 0.99 0.04 1.00 0.03 1.03 0.05
Commercial Bank 0.99 0.07 1.01 0.10 0.99 0.07 1.00 0.03 1.00 0.12
Savings Bank 1.06 0.14 0.90 0.12 1.03 0.05 1.04 0.13 0.95 0.13
1997–8
Total 1.02 0.11 0.95 0.09 1.01 0.07 1.01 0.08 0.97 0.11
Commercial Bank 0.99 0.13 0.99 0.13 1.02 0.19 0.98 0.10 0.99 0.22
Savings Bank 0.96 0.06 1.08 0.03 1.00 0.06 0.96 0.04 1.04 0.06
1998–9
Total 0.96 0.07 1.07 0.06 0.99 0.07 0.97 0.05 1.03 0.10
Commercial Bank 1.02 0.17 1.03 0.19 0.97 0.14 1.05 0.14 1.06 0.35
Savings Bank 1.01 0.08 1.01 0.06 1.01 0.07 0.99 0.05 1.01 0.06
Total 1.04 0.10 1.00 0.08 1.03 0.09 1.01 0.07 1.03 0.11
1999–2000
233
234
Panel C Italy* EFFCH TECHCH PECH SECH TFPCH
Commercial Bank 1.02 0.14 1.01 0.36 1.02 0.11 1.01 0.11 1.05 0.55
Savings Bank 1.04 0.12 0.92 0.10 1.01 0.06 1.03 0.10 0.95 0.12
2000–1
Total 1.05 0.12 0.97 0.17 1.03 0.10 1.02 0.09 1.01 0.23
Commercial Bank 0.97 0.12 1.02 0.19 0.98 0.10 0.99 0.08 0.98 0.22
Savings Bank 1.06 0.14 0.90 0.12 1.02 0.05 1.03 0.12 0.95 0.13
2001–2
Total 1.02 0.11 0.96 0.09 1.01 0.07 1.01 0.07 0.97 0.11
Commercial Bank 1.00 – 1.01 – 1.00 – 1.00 – 1.01 –

Savings Bank 1.02 – 0.97 – 1.01 – 1.01 – 0.99 –
95–02
MEAN+
Total 1.01 1.00 1.01 1.00 1.01
Commercial Bank 0.97 – 1.08 – 0.97 – 1.01 – 1.09 –

Savings Bank 1.13 – 0.84 – 1.08 – 1.04 – 0.93 –
95–02
OVERALL
Total 1.09 0.97 1.07 1.02 1.04
+ Geometric mean
Panel D United Kingdom EFFCH TECHCH PECH SECH TFPCH

1995–6 Commercial Bank 1.01 0.11 1.02 0.10 1.01 0.05 1.00 0.09 1.03 0.15
1996–7 Commercial Bank 1.00 0.06 1.00 0.06 1.00 0.05 1.00 0.03 1.00 0.08
1997–8 Commercial Bank 1.01 0.09 1.01 0.04 1.00 0.06 1.01 0.06 1.03 0.10
1998–9 Commercial Bank 1.00 0.06 0.99 0.06 1.00 0.05 1.00 0.04 0.99 0.08
1999–2000 Commercial Bank 1.03 0.04 0.96 0.14 1.01 0.07 1.02 0.04 1.05 0.04
2000–1 Commercial Bank 1.01 0.08 1.00 0.02 1.01 0.06 1.01 0.08 1.01 0.08
2001–2 Commercial Bank 1.02 0.11 1.00 0.06 1.01 0.09 1.01 0.09 1.02 0.13
+
MEAN 95–2 Commercial Bank 1.01 – 1.00 – 1.01 – 1.01 – 1.02 –
OVERALL 95–2 Commercial Bank 1.08 – 0.98 – 1.04 – 1.05 – 1.14 –
+ Geometric mean
235
Technical Efficiency Change Pure Technical Efficiency Change

Technical Change Scale Efficiency Change TFP change
1,6
1,5
1,4
1,3
1,2
1,1
1,0
0,9
0,8
0,7
0,6
0,5
0,4
.
b.
b.
b.
b.
.
b.
b.
b.
eb
sb
sb
ial
ial
ive
ial
ial
gs
ive
ing
ing
tiv
erc
rc
rc
rc
vin
rat
rat
era
me
av
me
av
me
mm
sa
pe
pe
ns
hs
op
om
om
om
oo
oo
an
co
co
lia
nc
nc
hc
hc
nc
hc
rm
Ita
Fre
an
an
lia
itis
nc
Ge
lia
nc
rm
rm
Ita
Ita
Fre
Fre
Br
Ge
Ge
Figure 6.5: Total Factor Productivity change over the period 1995–2002 in European
banking
change due to similar trends of technical change and technical efficiency

change. As illustrated earlier, German commercial banks improved their
TFP over the period 1995–2002 by 6 per cent: however, TFP is stable up to
2001 (the overall TFP improvement from 1995 to 2001 is 1 per cent) and
substantially increased in 2002, due to their better ability to exploit scale
efficiency and to improved pure technical efficiency. Similar to German
banks, Italian commercial and cooperative banks display increasing TFP
(namely, 9 per cent and 6 per cent). The TFP of commercial banks was con-
stant over the period 1995–9, sharply increased between 1999 and 2001
(due to an increase of both technical change and technical efficiency
change) and slightly decreased in 2001–2. Similarly, cooperative banks
display an increasing TFP trend between 1998 and 2001 and a TFP decrease
in 2001–2. Savings banks display similar trends relating to technical change
and technical efficiency. It is interesting to note that pure technical
efficiency change constantly improved over the period 1995–2002 provid-
ing evidence of the ability of inefficient cooperative banks to reduce the
distance from the efficient frontier.
Overall, we found that TFP increased on average in all four banking
industries analysed, but with different magnitude and for different reasons
(Figure 6.6). Our findings show the largest productivity growth in the UK
(TFP improved by 14 per cent in comparison to the 1995 level), due to pos-
itive pure efficiency and scale efficiency changes. In France, we found that
commercial banks experienced a substantial TFP growth (namely, TFP
improved by 41 per cent in comparison to the 1995 level) mainly due to
efficiency improvements, while cooperative and savings banks seem to
have decreased significantly their TFP (by 40 per cent and 52 per cent in
comparison to the 1995 level, respectively) as a result of both technological
and pure efficiency declines. In Italy and Germany, we estimate a slight
improvement in TFP since TFP improved by 4 per cent and 1 per cent in
comparison to the 1995 levels. In both countries, commercial banks experi-
ence a larger TFP improvement than cooperative and savings banks and
their improvement seems to be due to a technological positive change.
Our findings seem to be consistent with previous studies dealing with
productivity in European banking sectors. Among various studies under-
taken, we focus our comparison on Casu et al. (2004), who employ a
similar methodology for estimating the Malmquist TFP index (although
there are substantial methodological differences26), focusing on the same
period (1995–2000).
Focusing on France, Casu et al. (2004) estimate a mean annual TFP pro-
ductivity increase of 3 per cent over the period 1995–2000, while we esti-
mate a slightly lower growth of 1 per cent over the same period. Our results
(shown in Table 6.7) indicate a substantial TFP increase for commercial
banks and a reduction for savings and commercial banks. The technical
change estimates by Casu et al. (2004) are substantially equivalent to our
results for commercial banks, although our estimates differ for cooperative
and savings banks. For instance, we find mean technological change of
3 per cent which is lower than that estimated in Casu et al. (2004). Our
mean estimate of efficiency change is very similar to Casu et al.’s (2004)
results (around 0.98). Our results suggest a very slight efficiency improve-
ment over the period analysed for commercial banks and a decrease for
cooperative and savings banks and Casu et al. (2004) estimate a small
reduction of 3.1 per cent. Regarding Germany, Casu et al. (2004) estimate a
mean annual TFP productivity increase of 1.6 per cent,27 whereas we esti-
mate no substantial changes in TFP between 1995 and 2002. Our results on
TFP for German commercial, cooperative and savings banks are 0 per cent,
–0.2 per cent and 0.2 per cent, respectively. The estimates provided by Casu
et al. (2004), however, for technical, pure efficiency and scale efficiency
changes are very close to our estimates (see Table 6.7). In Italy, we find
slight TFP increase over the period 1995–2000 (i.e. 1.2 per cent) that is
similar to Casu et al. (2004). While we estimate very similar mean levels of
pure efficiency and scale change, this difference is due to the estimation of
the technological change. On average, we found that mean annual techno-
logical improvement is around 1 per cent for all bank categories, while
238
French commercial banks French cooperative banks French savings banks

Technical Efficiency Change Technical Change
Scale Efficiency Change Technical Efficiency Change Technical Change Technical Efficiency Change Technical Change
Pure Technical Efficiency Change
Pure Technical Efficiency Change Scale Efficiency Change Pure Technical Efficiency Change Scale Efficiency Change
TFP change
TFP change TFP change
1,50
1,10 1,10
1,40
1,00 1,00
1,30 0,90
0,90
1,20 0,80
0,80
1,10 0,70
0,70
1,00 0,60
0,60
0,50
0,90 0,50
95–96 95–97 95–98 95–99 95–00 95–01 95–02 95–96 95–97 95–98 95–99 95–00 95–01 95–02 0,40
95–96 95–97 95–98 95–99 95–00 95–01 95–02
German commercial banks German cooperative banks German savings banks

Technical Efficiency Change Technical Change Technical Change Technical Efficiency Change Technical Change
Technical Efficiency Change
Pure Technical Efficiency Change Scale Efficiency Change Pure Technical Efficiency Change Pure Technical Efficiency Change Scale Efficiency Change
Scale Efficiency Change
TFP change TFP change TFP change
1,20
1,05
1,08
1,15 1,04
1,10 1,03 1,06
1,02
1,05 1,04
1,01
1,00 1,00 1,02
0,99
0,95 1,00
0,98
0,90 0,97 0,98
0,96
0,85 0,96
0,95
0,80 0,94
95–96 95–97 95–98 95–99 95–00 95–01 95–02 0,94
95–96 95–97 95–98 95–99 95–00 95–01 95–02 95–96 95–97 95–98 95–99 95–00 95–01 95–02
Figure 6.6: Cumulative Total Factor Productivity estimates in European banking between 1995 and 2002
Italian commercial banks Italian cooperative banks

Technical Efficiency Change Technical Change Technical Efficiency Change Technical Change
1,14 1,10
1,12 95-96 96-97 97-98 98-99 99-00 00-01 01-02 1,08
1,10 1,06
1,08
1,04
1,06
1,04 1,02
1,02 1,00
1,00 0,98
0,98
0,96 0,96
0,94 0,94
95–96 95–97 95–98 95–99 95–00 00–01 01–02 95–96 95–97 95–98 95–99 95–00 00–01 01–02
British commercial banks Italian savings banks

Technical Efficiency Change Technical Change Technical Efficiency Change Technical Change
1,06 1,15
1,05
1,04 1,10
1,03 1,05
1,02
1,01 1,00
1 0,95
0,99
0,90
0,98
0,97 0,85
0,96 0,80
0,95
95–96 95–97 95–98 95–99 95–00 00–01 01–02
95–96 95–97 95–98 95–99 95–00 00–01 01–02
Figure 6.6: Continued.
Casu et al. (2004) estimate a growth of 13.6 per cent that is, partly, due to
their estimate between 1997 and 1998 (i.e. 32 per cent). There may be
several reasons for these differences, such as the different sample composi-
tion and the definition of bank inputs and outputs. Overall, we believe our
results are broadly consistent with the findings of Casu et al. (2004) and at
least can therefore be viewed as ‘believable’ despite one or two differences.
The remaining part of the chapter focuses on how we intend to estimate
customer satisfaction in banking.
Measuring customer satisfaction

Customer satisfaction is a primary driver for creating shareholder value, as
discussed in Chapter 4. As noted earlier, it is possible to identify a chain
of customer satisfaction indicators and distinguish two types of customer
satisfaction measures: indirect measures (which do not directly measure
customer satisfaction but their economic meaning supplies information
about customer satisfaction) and direct measures (which directly measure
customer satisfaction). Although direct measures of customer satisfaction
are more precise than indirect measures, this kind of information is gener-
ally not publicly available for banking industries as it is usually obtained
via survey/questionnaire by banks or associations and remains private. It is
also outside the scope of this book to undertake such a survey. As such, our
focus is on indirect measures of customer satisfaction of which there are
three main types as shown in Figure 6.7.
Table 6.7: Mean levels of productivity changes between 1995 and 2000 in European
banking
France EFFCH TECHCH PECH SECH TFPCH

estimates
Commercial Bank 1.010 1.061 1.003 1.008 1.057

Our
Cooperative Bank 0.940 0.969 0.948 0.994 0.919

Savings Bank 0.909 0.969 0.898 1.004 0.878
MEAN 0.985 1.030 0.984 1.006 1.009
Casu et al. (2004) 0.969 1.059 0.988 0.982 1.028
Germany EFFCH TECHCH PECH SECH TFPCH

estimates
Commercial Bank 0.981 1.024 1.021 0.979 1.000

Our

Savings Bank 1.012 0.992 1.004 1.006 1.002
MEAN 1.006 0.994 1.006 1.000 1.000
Casu et al. (2004) 1.013 1.003 1.012 1.002 1.016
Italy EFFCH TECHCH PECH SECH TFPCH

estimates
Commercial Bank 0.996 1.010 0.994 1.002 1.012

Our

Savings Bank 1.005 1.002 1.010 0.996 1.006
MEAN 1.004 1.009 1.006 0.998 1.012
Casu et al. (2004) 0.986 1.136 0.990 0.995 1.119
United Kingdom EFFCH TECHCH PECH SECH TFPCH
Our estimate 1.010 0.996 1.004 1.006 1.020

Casu et al. (2004) 1.000 1.022 1.003 0.996 1.021
The first type is given by economic measures expressing the overall

bank’s success, such as the overall economic and financial results (e.g.
profit or market value). Although profits and a sound financial structure
may be a good indicator of customer satisfaction, the link between cus-
tomer satisfaction and banks’ profit is (we believe) ‘too far’ and many other
factors can influence economic profit, such as extraordinary activities or
accounting evaluations.
The second type of indicators provide more precise information about
customer satisfaction: the number of contracts, the volume of funds inter-
mediated, etc. are often considered as a signal of the quality of product/
services supplied and of customer satisfaction. Although the link between
Figure 6.7: The chain of customer satisfaction indicators in banking
Profit
I Level
Volumes
(number of products)
II Level
No. of Average No. of Current average No. of Average increased No. of Average volume
new x volume per + customers at x CRR x volume of existing + customers at x CRR x volume of existing + customers at x CRR x of new
custo- ‘new’ the beginning products/services the beginning products/services the beginning products/services
mers customers of the period per ‘old’ customers of the period per ‘old’ customers of the period per ‘old’
customers
III Level
Customer Perceived Perceived Customer
IV Level
Satisfaction Quality Value Loyalty
Note: CRR = Customer Retention Rate

Source: modified from Di Antonio (2002: 133).
241
these measures and customer satisfaction is stricter (than at the first level)
the volume of various transactions undertaken does not necessarily imply
customer satisfaction. For example, these may be due to factors such as
banks’ market power in local areas or other commercial factors ‘forcing’
customers to acquire banks’ products rather than to customer satisfaction
per se.
The third type of measures show if commercial results derive from new
customers or existing customers, from new products/services or the sale of
existing products. In detail, it is possible to identify the following indica-
tors: (1) the number of new customers; (2) the number of existing cus-
tomers leaving the bank over a period and, complementary to this, the
number of customers that remain at the end of period. These measures are
usually expressed by a Customer Retention Rate (CRR) and Customer Loss
Rate (CLR) that are strictly related to the average length of the relationship
between the bank and its customers; (3) cross selling revenues; (4) the
number of customer defections; and (5) the number of complaints received
from customers. Although these indicators may be incomplete, they are
more precise than measures of the first and second type and these are
usually taken as preferable indirect indicators of customer satisfaction.
The definition of an appropriate indirect measure of customer satisfac-
tion (for our empirical investigation) is constrained by data availability.
The third type of measures, while preferable, cannot be used in our empiri-
cal analysis because this sort of information is not widely available. As
such, we can only use the second type of measures and this may introduce
some bias in our empirical investigation. While the link between economic
measures of customer satisfaction is theoretically straightforward, this link
is not unambiguous and other factors (such as the bank’s market power in
the local area or the market growth, etc.) can influence these economic
measures. For example, a bank may increase its sales in the long run by sat-
isfying customers, but the increase in sales may be due to other economic
factors (see Figure 6.8).
In addition, customer satisfaction and second order measures of cus-
tomer satisfaction are not time coincident. For example, the relationship
between customer satisfaction and a company’s sales is observed over a
long period: when a bank is able to satisfy its customers, customers increase
their confidence in the bank, the bank’s image and reputation improve,
customers have a more collaborative behaviour, become more loyal and the
length of the relationship between the bank and the customer increases.
Under these circumstances, banks attract new customers, sell new products
to their customers (increasing the number of customers’ needs served by
the bank), reducing the amortisation period of fixed investments and
advertising expenses. As such, the variation of bank’s sales depends on the
customer satisfaction over a previous period of time, rather than in the
same time period.
Customer
satisfaction
Market power in local

markets Bank’s sales
How long this
Macroeconomic trend process takes
within the banking market
Other factors
Independent variables
(factors affecting bank sales) Dependent variables
Figure 6.8: The relationship between customer satisfaction and bank sales
Despite the aforementioned limitations, we start off by attempting to

develop a measure of customer satisfaction based on the second type of
economic measures. Following the value-added approach for defining bank
inputs and outputs, we identify two kinds of banks’ customers: depositors
and borrowers. In order to select an appropriate measure of sales for assess-
ing depositors’, and borrowers’ satisfaction, we define two different mea-
sures of customer satisfaction in banking that relate to the total amount of
demand deposits and the overall amount of loans.
total amount of outstanding loans at time

t = f (borrowers’ satifaction) (6.12)
total amount of outstanding demand deposits
at time t = f (depositors’ satisfaction)
The second step is to standardise our indicator to better capture the cus-
tomer satisfaction concept: since customer satisfaction generates a variation
of a company’s sales, customers’ satisfaction is better investigated by con-
sidering the rate of change of deposits (D) and of loans (L), rather than the
total amount outstanding at a given moment of time.
Lt – Lt–1
= f (Borrowers’ satisfaction)
Lt–1 (6.13)
Dt – Dt–1
= f (Depositors’ satisfaction)
Dt–1
The third step aims to isolate (or at least reduce) the impact of other eco-
nomic factors on the relationship between sales and customer satisfaction.
In order to eliminate the impact due to general banking industry trends, we

consider an ‘adjusted’ rate of change of deposits and of loans by comparing
this to an appropriate benchmark. As such, customer satisfaction is mea-
sured focusing on the difference between the rate of change of deposits
(and of loans) and the rate of change of deposits (loans, respectively) mea-
sured at the industry level (i.e. France, Germany, Italy and UK): in other
words, the level of customer satisfaction is inferred looking at the differ-
ence between the bank deposit (respectively, loans) rate of change and the
industry average rate of change.
LB ’t – LB ’t–1 LSb ’t – LSb ’t–1
– = f (Borrowers’ satisfaction)
LB ’t–1 LSb ’t–1
(6.14)
DB ’t – DB ’t–1 DSb ’t – DSb ’t–1
– = f (Depositors’ satisfaction)
DB ’t–1 DSb ’t–1
where DB, t is the total amount of outstanding deposits of the bank at time
t, DI, t is the total amount of outstanding deposits at the industry level at
time t, LB, t is the total amount of outstanding loans of the bank at time t
and LI, t is the total amount of outstanding loans at the industry level at
time t.
In addition, to isolate the impact due to different banking business types,
we also adopt a larger set of specific benchmarks (of the mean rate of
change of deposits/loans) according to the various types of banks (i.e. com-
mercial banks, cooperative banks and savings banks).
LB ’t – LB ’t–1 LSb ’t – LSb ’t–1

– = f (Borrowers’ satisfaction)
LB ’t–1 LSb ’t–1
(6.15)
DB ’t – DB ’t–1 DSb ’t – DSb ’t–1
– = f (Depositors’ satisfaction)
DB ’t–1 DSb ’t–1
where DB, t is the total amount of outstanding deposit of the bank at time t,
DSb, t is the total amount of outstanding deposit for a specific type of
bank (i.e. commercial/savings/cooperative banks for each country) at
time t, LB, t is the total amount of outstanding loan of the bank at time t
and LSb, t is the total amount of outstanding loan for a specific type of bank
(i.e. commercial/savings/cooperative banks) at time t.
In order to measure the growth in deposits and loans relative to the
industry benchmark (i.e. the bank rate of change of deposits and of loans
minus the industry rate of change of deposits and of loans), we use a
sample slightly different from that used in previous sections for estimating
cost and profit efficiency since we consider a shorter time period from 1997
to 2002. We select various sub-samples according to the country (namely,
Table 6.8: Number of banks in the sample for assessing customer satisfaction in
European banking
1997 1998 1999 2000 2001 2002

France
Commercial banks 205 231 241 249 257 255

Cooperative banks 81 90 95 91 96 92
Savings banks 22 22 25 25 25 27
Germany

Savings banks 517 552 570 589 640 612

Italy

Savings banks 62 62 62 63 64 65
Kingdom
United

Savings banks 0 0 0 0 0 0
Source: Banksope.
France, Germany, Italy and the UK), to the year (namely, from 1997 to
2002) and for the type of bank. Table 6.8 reports the number of banks com-
prising the sample.
Table 6.9 reports our estimates (distinguishing for time, countries and
type of bank) for the mean rate of change of deposits and of loans, the
mean change of deposits and of loans in the industry and the net bank rate
of change of deposits and of loans. When a specific banking sector (e.g.
French cooperative banks in 2002) exhibit a positive mean ‘net’ change of
deposits (loans), this provides evidence that the number of (French cooper-
ative) banks with a deposit (loan) rate of change greater than that of the
industry (French cooperative banking) is larger than the number of (French
cooperative) banks with a rate of change smaller than that of the industry.
According to our estimates, we find that deposits and loans have con-
stantly grown between 1997 and 2002 in all four banking systems analysed
(especially in Italy and the UK). However, when we analyse the ‘net’
deposit and loan growth rates (i.e. the bank growth rate minus the mean
industry growth rate) the situation changes across countries. In France,
deposits and loans growth rates are negative between 1999 and 2002
showing that a large part of French banks achieved deposit and loan
growth rates lower than those at the industry level. French commercial
banks exhibit a high ‘net’ growth of loans in 1995 and especially in 2000,
while these achieved a deposit growth rate substantially lower than that of
Table 6.9: Deposits and loans growth rates in European banking between 1997 and 200228
246
Mean Mean Deposits annual Loans annual

industry industry
Deposits annual Loans annual ‘net’ growth ‘net’ growth
deposit loans
Panel A France growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
Mean St. Dev. Mean St. Dev. (column 3) (column 4) Mean St. Dev. Mean St. Dev.
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
Commercial Bank –0.09 47.65 3.10 57.78 4.08 8.06 –4.17 47.65 –4.96 57.78
2002
Savings Bank 7.92 4.28 5.31 4.73 7.07 2.98 0.84 4.28 2.33 4.73
MEAN 2.85 39.52 4.59 44.75 4.18 6.69 –1.32 39.52 –2.10 44.75
Commercial Bank 4.06 68.82 9.83 58.55 12.68 7.88 –8.62 68.82 1.95 58.55
2001
Savings Bank 8.03 3.53 4.85 7.31 1.83 19.90 6.20 3.53 –15.06 7.31
MEAN 5.66 48.77 8.71 41.97 11.06 8.13 –5.40 48.77 0.58 41.97
Commercial Bank 5.97 38.74 30.22 37.73 8.78 11.89 –2.82 38.74 18.33 37.73
Cooperative Bank 10.17 10.63 7.72 6.36 12.02 4.01 –1.86 10.63 3.71 6.36
2000
Savings Bank 9.78 4.21 6.78 4.36 9.28 4.66 0.50 4.21 2.12 4.36
MEAN 7.28 29.37 23.00 27.62 9.62 9.43 –2.35 29.37 13.57 27.62
Commercial Bank 2.55 47.54 8.70 43.90 6.33 6.15 –3.78 47.54 2.55 43.90
Cooperative Bank 13.18 35.93 13.50 42.27 25.56 22.21 –12.39 35.93 –8.71 42.27
1999
Savings Bank 10.15 10.54 6.97 3.83 36.87 8.92 –26.72 10.54 –1.95 3.83
MEAN 5.87 41.92 9.84 40.70 13.51 10.57 –7.63 41.92 –0.72 40.70
Table 6.9: Deposits and loans growth rates in European banking between 1997 and 2002 – continued

industry industry
deposit loans
Panel A France growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
Commercial Bank 12.13 47.27 10.34 48.44 8.41 13.25 3.72 47.27 –2.91 48.44
1998
Savings Bank 7.36 2.83 9.78 5.48 8.12 6.91 –0.76 2.83 2.87 5.48
MEAN 10.07 34.07 9.36 34.85 7.88 10.33 2.19 34.07 –0.97 34.85
Commercial Bank 12.10 59.12 14.47 81.14 1.02 5.47 11.08 59.12 9.00 81.14
1997
Savings Bank 8.42 4.89 6.89 3.33 8.74 8.23 –0.32 4.89 –1.35 3.33
MEAN 10.46 41.39 11.60 56.40 3.41 5.06 7.04 41.39 6.54 56.40
1
Mean values of loans (and of deposits) net growth rates are not weighted to the total amount of loans (and of deposits) at time t–1 (i.e. deposits at t-1
and loans at t–1).
247
248

industry industry
deposit loans
Panel B Germany growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
Commercial Bank 6.18 61.94 4.02 47.80 –11.66 5.31 17.83 174.62 –1.29 63.12
2002
Savings Bank 2.80 7.21 3.61 7.25 3.73 3.56 –0.93 7.21 0.06 7.25
MEAN 3.74 17.79 4.36 15.54 1.26 4.31 2.48 31.61 0.05 17.42
Commercial Bank 14.22 60.23 11.84 47.85 5.97 9.50 8.25 119.08 2.33 606.68
2001
Savings Bank 4.92 9.26 3.87 9.69 6.28 5.83 –1.36 9.26 –1.96 9.69
MEAN 8.06 21.94 6.09 19.91 8.73 5.34 –0.67 29.15 0.75 88.44
Commercial Bank 16.91 55.80 8.14 30.85 11.33 15.63 5.58 71.10 –7.49 30.85
2000
Savings Bank 7.82 21.13 6.99 20.71 7.77 7.71 0.04 27.46 –0.72 28.00
MEAN 10.06 27.26 7.89 23.49 10.55 10.90 –0.49 31.38 –3.01 25.92
Commercial Bank 9.96 49.97 8.77 33.65 11.82 10.67 –1.87 53.86 –1.90 33.65
1999
Savings Bank 6.40 6.73 6.15 6.14 6.06 6.59 0.34 6.73 –0.44 6.14
MEAN 8.03 24.53 8.34 18.87 9.55 8.97 –1.53 25.06 –0.63 18.87

industry industry
deposit loans
Panel B Germany growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
Commercial Bank 16.73 63.97 13.45 33.33 15.05 14.32 1.68 100.17 –0.87 33.33
Cooperative Bank 8.55 14.84 7.32 13.41 8.49 9.09 0.06 14.84 –1.78 13.41
1998
Savings Bank 6.53 8.37 6.03 8.64 5.95 6.35 0.58 8.37 –0.32 8.64
MEAN 8.89 19.03 7.66 14.30 8.42 8.77 0.47 23.91 –1.11 14.30
Commercial Bank 11.53 48.03 8.32 22.52 8.74 13.26 2.80 55.58 –4.94 22.52
Cooperative Bank 8.26 14.21 7.36 13.41 7.92 7.97 0.35 14.21 –0.62 13.41
1997
Savings Bank 8.35 15.17 7.18 10.30 7.03 7.25 1.32 15.17 –0.07 10.30
MEAN 8.72 18.98 7.41 13.33 7.67 8.36 1.06 19.95 –0.96 13.33
249
250

industry industry
deposit loans
Panel C Italy growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
Commercial Bank 14.72 49.83 26.49 61.30 3.86 0.15 10.86 49.83 26.34 61.30
2002
Savings Bank 4.55 13.20 8.86 8.44 –3.19 5.14 7.75 13.20 3.71 8.44
MEAN 11.76 24.92 17.83 56.73 7.87 8.93 3.89 24.92 8.90 56.73
Commercial Bank 27.18 79.12 50.51 291.00 7.05 7.42 16.46 79.12 60.05 291.00
2001
Savings Bank 6.38 13.67 13.01 13.57 2.85 10.28 3.53 13.67 2.73 13.57
MEAN 19.39 46.57 36.84 124.30 13.36 18.90 4.64 46.57 24.34 124.30
Commercial Bank 48.87 330.65 31.20 169.55 0.57 3.02 48.30 330.65 28.18 169.55
2000
Savings Bank 3.94 8.73 12.68 10.59 –0.14 9.23 4.08 8.73 3.45 10.59
MEAN 19.44 110.18 21.30 59.90 5.33 11.93 14.11 110.18 9.37 59.90
Commercial Bank 6.69 21.97 10.93 38.88 –1.20 –0.25 7.89 21.97 11.18 38.88
1999
Savings Bank 0.44 8.59 10.40 8.12 0.99 7.06 –0.55 8.59 3.34 8.12
MEAN 5.31 13.75 12.63 19.88 3.61 8.58 1.70 13.75 4.05 19.88

industry industry
deposit loans
Panel C Italy growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
Commercial Bank 8.13 24.82 9.01 23.74 5.23 4.61 2.90 24.82 4.40 23.74
1998
Savings Bank 7.54 8.75 10.78 7.06 3.23 7.68 4.31 8.75 3.10 7.06
MEAN 8.62 16.11 11.42 14.98 5.46 7.52 3.15 16.11 3.89 14.98
Commercial Bank 4.81 19.38 5.87 17.94 1.29 1.45 3.52 19.38 4.43 17.94
1997
Savings Bank 4.13 8.82 7.03 7.64 3.15 4.83 0.97 8.82 2.20 7.64
MEAN 5.82 14.94 8.18 13.53 3.50 4.64 5.29 14.94 5.15 13.53
251
252

industry industry
deposit loans
Panel D United Kingdom growth (column 1) growth (column 2) (columns 1–3) (columns 2–4)
growth growth
(%) (%) (%) (%) (% (%) (%) (%) (%) (%)
2002 Commercial Bank 11.54 64.03 6.72 34.36 13.65 10.39 –2.11 64.04 –3.67 34.36
2001 Commercial Bank 14.31 64.54 15.30 62.92 13.34 13.67 0.97 64.54 1.63 62.92
2000 Commercial Bank 33.59 85.57 10.60 40.71 –3.67 –3.74 37.26 85.57 14.34 40.71
1999 Commercial Bank 12.68 55.13 7.46 47.79 12.85 10.73 –0.17 55.13 –3.27 47.79
1998 Commercial Bank 15.10 68.25 16.92 59.16 9.50 13.46 5.60 68.25 3.46 59.16
1997 Commercial Bank 15.00 63.45 17.96 53.46 4.11 1.08 10.89 63.45 16.87 53.46
the industry from 1999 up to 2002. In Germany, the largest part of German
banks achieved loans growth rates lower than that of the industry between
1997 and 2000, while most banks achieved a positive net growth in the
other years. German commercial banks exhibit a high ‘net’ growth of
deposits between 2000 and 2002, while these achieved loan growth rates
substantially lower than those of the industry almost constantly in the
period analysed. In Italy and UK, the majority of Italian banks appear to
have achieved deposits and loans growth rates superior to those of the
industry level. In Italy, this is mainly due to commercial banks and savings
banks that display constantly positive mean loan and deposit growth rates.
Once we estimated the ‘net’ growth rate, we have to modify this measure
to assess customer satisfaction because the relationship between deposi-
tors’/borrowers’ satisfaction and deposits/loans ‘net’ growth rate of change
is not time coincident. As discussed earlier, banks’ deposit and loan growth
depends on customer satisfaction over a previous period of time. As such,
deposits and loans net growth rates cannot be considered as a proxy for
customer satisfaction in the same period. Since the true level of customer
satisfaction is unknown and direct measures of customer satisfaction are
not publicly available, it is impossible to undertake any statistical test for
assessing a proper lag between these two variables or, in other words, how
long the period of time is in order that the variation of customer satisfac-
tion determines a variation in bank’s sales. For this reason, we assume that
customer satisfaction is constant over a three-year window over the period
considered (i.e. 1995–2002). As a result customer satisfaction measures are
obtained for the period (1997–2002) by using six windows: e.g. depositors’
(borrowers’) satisfaction for bank j in 1997 is measured as the adjusted rate
of change of deposits (loans, respectively) for bank j over the period
1995–7, etc. Once having defined the length of the windows, it is necessary
to define a procedure for transforming annual adjusted deposits (loans) rate
of change to measures of depositors’ (borrowers’, respectively) satisfaction.
As such, the indicator of depositors’ (respectively, borrowers’) satisfaction
for bank j uses the following procedure: (1) rank (from the lowest to the
highest) the annual adjusted deposits (loans) rate of change for every year;
(2) transform the rank into a percentage rank: e.g. if bank j is ranked 48
over 200 banks, its percentage rank is 24 per cent (i.e. 48/200); (3) customer
satisfaction is simply the mean of the two percentage rank scores obtained
by bank j in every year of the window: this value varies between 0 (if the
bank achieved constantly the lowest annual adjusted deposits (loans) rate
of change for every year) and 100 per cent (if the bank achieved constantly
the highest annual adjusted deposits (loans) rate of change for every year).
In this way, we obtain a proxy for customer satisfaction based as a percent-
age (i.e. 0 lowest satisfaction and 100 per cent maximum satisfaction) that
takes account of bank type and other broad trends affecting the European
banking industry.
Conclusion
Following the analysis of drivers of shareholder value presented in Chapter
4, this chapter reports various approaches that can be used to measure
these drivers. As in the established bank efficiency literature, mean levels of
efficiency obtained using DEA are lower than those obtained using SFA.
According to our SFA estimates, European banks display over the period
1995–2002 cost inefficiency scores ranging between 7.3 per cent (UK com-
mercial banks in 1995) to 38.9 per cent (German cooperative banks in
2001). On average, the mean cost X-efficiency level over the period
1995–2002 is 78.1 per cent. Focusing on the four banking systems analysed,
UK banks display the highest efficiency levels than the other three coun-
tries. Looking at the different bank specialisations, commercial banks
display the lower mean level of cost X-efficiency (75.0 per cent) than
savings and cooperative banks (77.7 per cent and 79.4 per cent, respec-
tively).
The second empirical analysis concerns the estimation of profit
efficiency. We estimated alternative profit efficiency measures developed by
Berger and Mester (1997) and we define bank inputs and outputs according
to the value-added approach. Over the period analysed, we find that, on
average, banks in the four European banking systems analysed achieve only
two-thirds of their potential profits. UK banks are found to be the most
profit efficient with a mean of 73.9 per cent, while French, German and
Italian banks are found to have similar levels of profit efficiency (around 65
per cent). Our profit efficiency estimates appear to be consistent with previ-
ous studies.
The third part of our empirical investigation concerned the estimation of
Total Factor Productivity changes. We estimate the Malmquist productivity
index using DEA. We decompose TFP changes into technical efficiency
changes, technical changes, scale efficiency changes and pure technical
efficiency. Overall, we find that TFP improved substantially in the UK (due
to improved pure technical efficiency change and to a better ability to
exploit scale efficiency), in Italy for commercial banks (due to technical
change) and cooperative banks (due to pure technical efficiency change), in
Germany for commercial banks (due improved pure technical efficiency
and to a reduction in input waste resulting in the ability to produce at
lower optimal scale). In contrast, we observe a regress in the efficient fron-
tier from 1995 to 2002 and TFP decrease for Italian savings banks and
French cooperative banks.
The final empirical analysis regards the estimation of bank customer sat-
isfaction. The definition of an appropriate measure of customer satisfaction
(for our empirical investigation) is constrained by data availability. As such,
we define a second order measure (namely, deposits and loans growth rela-
tive to the industry average). Since our customer satisfaction estimates
range between 0 and 1 and these proxies are computed based on the per-
centage rank of net variation, we cannot discuss mean levels of customer
satisfaction in the banking industry (both over time and over different
countries). Instead, our goal has been to define an estimate of customer
satisfaction to be used in Chapter 7 for assessing the value relevance of cus-
tomer satisfaction in creating shareholder value in European banking.
7
Determinants of Shareholder Value in
European Banking and Shareholder Value
Efficiency
Introduction
In Chapter 4, we analysed some of the most common strategies developed
since the 1990s to create shareholder value, namely, increasing customer
satisfaction, enhancing bank cost and profit efficiencies and increasing pro-
ductivity. In Chapter 6 we presented various approaches to measure these
drivers. The first part of this chapter aims to use the aforementioned mea-
sures to see how important they are in describing shareholder value in
European banking. In particular, we use estimates of customer satisfaction,
cost efficiency, profit efficiency and productivity changes to see how these
factors explain shareholder value as shown in Figure 7.1. The approach
adopted is similar to that presented in Chapter 5 to test the information
content of performance measures. Focusing on the French, German, Italian
and UK banking systems over the period 1995–2002, we analyse the value
relevance of all these types of efficiency estimates for both listed banks and
non-listed banks.
The second part of the chapter develops a new measure of efficiency
labelled ‘shareholder value efficiency’. We define a bank that is able to
produce at the maximum possible shareholder value, given a particular
level of output levels, as ‘shareholder value efficient’.
Analysing the determinants of shareholder value

Methodology
This section deals with the methodology adopted in the empirical investi-
gation to examine the determinants of shareholder value in European
banking. Our aim is to assess whether efficiency, productivity and customer
satisfaction factors explain the variation of shareholder value.
256
Determinants of Shareholder Value in European Banking 257
Reducing bank costs

(cost efficiency and its
components)
Improving productivity
(TFP changes and its
components)
Customer satisfaction
(depositor and borrower
satisfaction proxies)
Profits
Shareholder value
Financial structure
Profit efficiency
Redefine bank business
mix
Merger and Acquisition
Etc.
Bank performance Bank shareholder

Bank profits
drivers value
Performance drivers empirically estimated in Chapter 6

Relationship assessed in Chapter 7
Relationship not assessed in this text
Figure 7.1: Outline of the relationship between shareholder value and its determin-
ants in banking
We apply a methodology similar to that adopted in Chapter 5. By assum-

ing equity markets are efficient in a semi-strong way and forward looking,
we apply the following model:
Dt = b′0 + b′1 Xt + b′2 Xt–1 + b′3 Xt–2 + et (7.1)
where Dt is the variable representing shareholder value created over the

period t, Xt is the variable that we are analysing to assess its information
content. We use two different dependent variables (Dt). For publicly listed
banks, we measure shareholder value created over the period t using Market
Adjusted Returns (MAR), i.e. the increment of equity market value (calcu-
lated considering a twelve-month non-overlapping period ending four
months after the firm’s fiscal year) and dividend per share paid in this
period, both standardised by market value of equity at the beginning of the

period and net of expected rate of return. For non-publicly listed banks, we
use the ratio between Economic Value Added, estimated using a procedure
accounting for banking peculiarities (EVAbkg), and the invested capital at
time t–1.
The independent variables analysed (Xt) are cost efficiency and its com-
ponents (namely, technical, allocative and scale efficiency), alternative
profit efficiency, and Total Factor Productivity (TFP) changes and its
components (namely, technological change, technical efficiency change,
pure technical efficiency change and scale efficiency changes). We run
model 7.1. for each of the variables listed in Figure 7.2.
We decided to use all these variables since each of these measures has a
different economic meaning expressing a different ability of a company. In
detail, technical efficiency expresses the ability of a firm to obtain maximal
outputs from a given set of inputs or of minimising inputs for a given
target of outputs; allocative efficiency refers to the ability to use the input
in optimal proportions, given their respective prices and production tech-
nology; cost and profit efficiency expresses the ability of a firm to choose
inputs and/or output levels and to mix these to minimise cost or maximise
profit, respectively; and productivity changes (TFP-CH) measures how the
ratio of a bank’s outputs to its inputs changes over two consecutive periods
(t and t+1). By assessing the value relevance of these variables, we are able
to analyse whether the creation of shareholder value is generated mainly
by a technical ability of the bank, or by allocative ability or both.
One criticism of this approach is that investors look at efficiency changes
over time rather than efficiency levels at a given moment: in other words,
Technical Efficiency estimated using DEA under the assumption of Variable Return to Scale (TE)
Allocative Efficiency estimated using DEA under the assumption of Variable Return to Scale (AE)
Scale Efficiency estimated using DEA (SE)
Cost (or Overall) Efficiency estimated using DEA under the assumption of Constant Return to Scale (CE)
Cost X-efficiency estimated using SFA (X-EFF)
Alternative Profit Efficiency (PROF-EFF)
Total Factor Productivity Change (TFP-CH)
Technological Change (TECH-CH)
Technical Efficiency Change (TE-CH)
Pure Technical Efficiency Change (PUTE-CH)
Scale Efficiency Change (SE-CH)
Depositor satisfaction (DS)
Borrower satisfaction (BS)
Figure 7.2: Determinants of shareholder value analysed (i.e. independent variables in

model 7.1)
investors may purchase securities of banks that improve their efficiency levels
over time (e.g. by 10 per cent in t–2, 40 per cent in t–2 and 70 per cent in t–3)
rather than those that are constantly more efficient (e.g. 90 per cent in t, in
t–1 and t–2). As such, market adjusted return may be affected by efficiency
changes. Some studies (such as Casu et al. 2004) recognise this fact and assess
the value-relevance of efficiency estimates running the following model:
Rjt = β0 + β1 Ejt + εjt (7.2)
where Rjt is the raw market return of bank j at the period ending at t and Ejt
is the annual change of efficiency estimates for bank j at the period ending
at t for two consecutive periods. Although abnormal return may certainly
be affected by efficiency changes, it seems to be imprecise to estimate
efficiency changes by comparing efficiency estimates obtained with DEA or
SFA in two different periods since efficiency estimates are obtained measur-
ing the distance from two different efficiency frontiers. As such, a bank
may improve its efficiency level over time (e.g. by 10 per cent in t–2, 40 per
cent in t–2 and 70 per cent in t–3) because other banks become less
efficient (i.e. a negative shift of the efficient frontier). In our opinion, the
influence of efficiency changes on shareholder value created over a period
can be better analysed by focusing on a decomposition of the productivity
changes over time. Since productivity is determined by the production
technology, efficiency and the operating environment, Total Factor
Productivity (TFP) can be decomposed into technical change (TECH-CH),
i.e. the shift of the efficient frontier between t and t+1, and efficiency
change (TE-CH), i.e. the change in the output-oriented measure of Farrell
technical efficiency between the period t and t+1 by assuming a constant
technology (namely, that the efficient frontier did not change over the two
periods). Efficiency change can been further decomposed by analysing the
extent to which efficiency changes between t and t+1 are due to scale
efficiency (SE-CH) or to pure efficiency (PUTE-CH) change. In other words,
the value relevance of these measures (TFP-CH TECH-CH, TEFF-CH, PURE
TEFF-CH and SE-CH) enables us to test the impact of the shift of the tech-
nology, of the change of technical efficiency and of scale efficiency on
shareholder value created over a period.
An interesting issue concerns the number of lags to be considered in the
model. Although it would be possible to consider any order of lags, we use
a model limited to two lags. The economic rationale for using a model with
two lags is that shareholder value created over the current period (t) is
assumed to be influenced by information (such as customer satisfaction
levels, efficiency levels and productivity changes) over the last two periods
(t, t–1 and t–2), while older information is already fully captured in market
prices since the equity markets are assumed to be efficient (at least in the
weak form).
In order to investigate the incremental information content, we focus on:

(1) cost efficiency components obtained using DEA; (2) TFP change com-
ponents; (3) cost and profit efficiency, obtained from stochastic frontier
estimations; and (4) depositors’ and borrowers’ satisfaction.
By testing DEA cost efficiency components, we are able to assess the
information content provided by allocative efficiency and scale efficiency
to the information content of technical efficiency. In addition, we can
assess if it is worthwhile to decompose cost efficiency estimates into its
components by comparing the explanatory power of the regression model
7.1 using cost efficiency estimates with the explanatory power of the regres-
sion model 7.5 considering all cost efficiency components. This test is
undertaken in three steps: firstly, we assess the value relevance of technical
efficiency estimated under Variable Returns to Scale (TE), then we intro-
duce in the model the allocative efficiency estimates (AE) and finally we
run again the model introducing the scale efficiency estimates (SE). We
examine whether the explanatory power (adjusted R2) of each model differs
and the contribution made by additional variables. The models run are:
Dt = b′0 + b′1 TEt + b′’2 TEt–1 + b′3 TEt–2 + et (7.3)

Dt = b′0 + b′1 TEt + b′2 TEt–1 + b′3 TEt–2 + b′4 AEt + b′5 AEt–1
(7.4)
+ b’6 AEt–2 + et
Dt = b′0 + b′1 TEt + b′2 TEt–1 + b′3 TEt–2 + b′4 AEt + b′5 AEt–1
(7.5)
+ b′6 AEt–2 + b′7 SEt + b′8 SEt–1 + b′9 SEt–2 + et
By testing TFP changes components, we are able to assess the additional

information content provided by pure technical efficiency changes and scale
efficiency changes to the information content of technological changes. In
addition, we can assess if it is worthwhile to decompose TFP changes into its
components by comparing the information content of TFP change estimates
(adjusted R2) and the overall information content of all TFP change com-
ponents (i.e. the adjusted R2 of model 7.8). This test is undertaken in three
steps: firstly, we assess the value relevance of technological changes estimates
(TECH-CH), then we introduce in the model the pure technical efficiency
changes estimates (TE-CH) and finally we run again the model introducing
the scale efficiency change estimates (SE-CH). The models run are:
Dt = b′0 + b′1 TECH-CHt,t–1 + b′2 YECH-CHt–1,t–2

(7.6)
+ b′3 TECH-CHt–2,t–3 + et
Dt = b′0 + b’1 TECH-CHt,t–1 + b′2 TECH-CHt–1,t–2 + b′3 TECH-CHt–2,t–2
(7.7)
+ b′4 PUTE-CHt,t–1 + b′5 PUTE-CHt–1,t–2 + b′6 PUTE-CHt–2,t–3 + et
Dt = b′0 + b′1 TECH-CHt,t–1 + b′2 TECH-CHt–1,t–2 + b′3 TECH-CHt–2,t–2
+ b′4 PUTE-CHt,t–1 + b′5 PUTE-CHt–1,t–2 + b’6 PUTE-CHt–2,t–3 (7.8)
+ b′7 SE-CHt,t–1 + b′8 SE-CHt–1,t–2 + b′9 SE-CHt–1,t–2 + et
We test the incremental information content provided by our alternative

profit efficiency estimates (PROF-EFF) to the information content of cost X-
efficiency (X-EFF). This analysis enables us to assess if it is worthwhile to
estimate both cost and profit efficiency in assessing the creation of share-
holder value of a bank. We decide to analyse the incremental information
content of profit efficiency in comparison to cost X-efficiency (rather than
cost efficiency measured using DEA) since both variables are estimated
using a parametric methodology and, consequently, are more homoge-
neous. The models run are:
Dt = b′0 + b’1 X-EFFt + b′2 X-EFFt–1 + b′3 X-EFFt–2, + et (7.9)

Dt = b′0 + b′1 X-EFFt + b′2 X-EFFt–1 + b′3 X-EFFt–2,
(7.10)
+ b′4 PROF-EFFt + b′5 PROF-EFFt–1 + b′6 PROF-EFFt–2, + et
Finally, we test the incremental information content provided by bor-

rower satisfaction (BS) to the information content of depositor satisfaction
(DS). The explanatory power of the regression model 7.12 provides us with
an overall assessment of bank customer satisfaction in creating shareholder
value. The models run are:
Dt = + b′0 + b′1 DSt + b′2 DSt–1 + b′3 DSt–2, + et (7.11)

Dt = b′0 + b′1 DSt + b′2 DSt–1 + b′3 DSt–2 + b′4 BSt + b′5 BSt–1
(7.12)
+ b′6 BSt–2, + et
Sample description
In analysing the value relevance of determinants of shareholder value, we
use two different samples.1 The first comprises only publicly listed banks in
France, Germany, Italy and the UK over the period 1995–2002. The sample
of listed banks is the same as used in Chapter 5 to assess the value-rele-
vance of innovative and traditional performance measures. The second
sample comprises all listed and non-listed banks in France, Germany, Italy
and the UK over the period 1995–2002. We also select four domestic sub-
samples (i.e. one to each of the four banking systems considered). This
sample is the same as used in Chapter 6 to estimate the determinants of
shareholder value (namely, cost and profit efficiency, Total Factor
Productivity (TFP) changes and customer satisfaction).
Regarding the time period analysed, our data set considers firm observa-
tions between 1995 and 2002 and, as such, we have to select samples over
the following periods:
• to test the information content of technical, allocative, scale, cost and

X-efficiencies, we use a model with two lags and, consequently, we con-
sider the period 1997–2002;
• to test the information content of TFP change, we use a model with two
lags and the variables are constructed considering the change between
two consecutive years and, consequently, we consider the period
1998–2002. As such, in 1998, we consider the term in 1998 (t), 1997
(t–1) and 1996 (t–2) and this latter term covers the change between two
consecutive periods (i.e. 1995–6)
• to test the information content of depositor and borrower satisfaction,
we use a model with two lags and the variables are constructed consider-
ing the deposit and loan rate of change between two consecutive years
and, consequently, we view the period 1999–2002. As such, in 1999, we
consider the term in 1999 (t), 1998 (t–1) and 1997 (t–2) and this refers to
two rates of change between two consecutive periods (namely, 1995–6
and 1996–7).
In brief, we observe that listed banks have, on average, a larger size than
non listed-banks. German banks are, on average, smaller than in the other
three countries. This is due to the high number of cooperative banks (about
four times the number of commercial banks) in Germany. Regarding
profitability, UK banks exhibit higher mean values of net income, ROE and
ROA than banks in the other three countries. Regarding the correlation
among shareholder value drivers, we find a positive (and in many cases
statistically significant) correlation among these variables in all samples.
Results for European publicly listed banks

In this section, we outline the main results of the analyses carried out to
test the relative and incremental information content of shareholder
drivers obtained analysing our sample of French, German, Italian and UK
listed banks. As examined in Chapter 3, the relative information content
refers to the association between stock market values and information of
firm specific features (e.g. accounting information, efficiency measures,
etc.) and the incremental information content refers to the contribution
provided by information of firm specific features in explaining market
returns given other specified variables.
We first assess the relative information content by looking at differences
in the explanatory power (adjusted R2) of regressions where market-
adjusted returns (MAR) is the dependent variable and determinants of
shareholder value (namely, cost efficiency, profit efficiency, productivity
changes and customer satisfaction) are the independent variables.
Table 7.1 reports the results of the relative information content by
reporting the coefficient estimates and their statistical significance (i.e.
based on the t-statistic), the adjusted R-squared, the Durbin-Watson test
(i.e. a test for assessing if residuals for consecutive observations are uncor-
related) and the p-values from a two-tailed statistical test (based on the
F-statistic) expressing the probability of rejecting the hypothesis that the
Table 7.1: Relative information content of shareholder value drivers analysing European publicly listed banks (dependent variable:
Market Adjusted Return)
Technical efficiency Allocative efficiency Scale efficiency Cost efficiency Cost X-efficiency Alternative profit
(DEA) (DEA) (DEA) (DEA) (SFA) efficiency (SFA)
t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2
† † ‡ ‡
Estimated 0.215 –0.13 0.076 0.117 0.238 –0.161* –0.026 0.175* –0.052 0.177* 0.176* –0.143* 0.053 0.308 0.006 0.112 0.261 0.006
coefficients
Adj. R2 0.395 0.393 0.337 0.371 0.433 0.431
P-value & F-stat 0 (F=19.670) 0 (F=17.924) 0 (F=15.568) 0 (F=17.889) 0 (F=21.754) 0 (F=21.549)
DW 1.929 1.986 1.977 1.972 2.073 1.998

Period analysed 1997–2002 1997–2002 1997–2002 1997–2002 1997–2002 1997–2002
The p-value reported, based on the F-test, expresses the probability of making an error rejecting the null hypothesis (i.e. all slope coefficients are equal to zero). As such, a
p-value close to 0 signals that the performance measure investigated is likely to have a statistically significant impact on MAR since at least one of the estimated regression
coefficients differs from zero. Broadly speaking, a p-value of 5% express that there is a probability of 5% that the performance measure investigated does not have a statistically
significant impact on MAR, and so on.

263
264
Table 7.1: Relative information content of shareholder value drivers analysing European publicly listed banks (dependent variable:
Market Adjusted Return) – continued
Technical efficiency Technological Pure technical Scale efficiency Total factor Depositor Borrower
change change efficiency change change productivity change satisfaction satisfaction
t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2
‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡
Estimated 0.300 0.253 –0.006 0.064 0.294 0.248 0.325 0.312 0.170 –0.138 –0.066 –0.237 0.232 0.411 0.147 0.017 0.351 0.012 –0.021 0.281 0.031
coefficients
Adj. R2 0.287 0.312 0.417 0.199 0.460 0.380 0.327
P-value & F-stat 0 (F=9.683) 0 (F=11.226) 0 (F=17.170) 0 (F=6.607) 0 (F=20.264) 0 (F=15.271) 0 (F=12.326)
DW 1.949 1.968 1.977 1.918 1.916 1.904 2.013
Period analysed 1998–2002 1998–2002 1998–2002 1998–2002 1998–2002 1999–2002 1999–2002
The p-value reported, based on the F-test, expresses the probability of making an error rejecting the null hypothesis (i.e. all slope coefficients are equal to zero). As such, a p-value close to 0 signals that the
performance measure investigated is likely to have a statistically significant impact on MAR since at least one of the estimated regression coefficients differs from zero. Broadly speaking, a p-value of 5% express
that there is a probability of 5% that the performance measure investigated does not have a statistically significant impact on MAR, and so on.

determinant of shareholder value investigated (e.g. cost X-efficiency) does

not have a statistically significant impact on MAR. As such, a p-value close
to 0 signals that it is very likely the factor investigated has a statistically
significant impact on MAR; a p-value of 5 per cent would express that there
is a probability of 5 per cent that the determinant of shareholder value
analysed does not have a statistically significant impact on MAR, etc.
According to Table 7.1, TFP changes have the highest relative informa-
tion content among the shareholder drivers analysed since this explains
about 46.0 per cent of variation of MAR. All estimated regression co-
efficients are found to be positive and statistically significant at the 1 per
cent level showing that TFP improvements lead to increasing bank share-
holder value. Among TFP components, we find that technological change
has a higher value-relevance (31.2 per cent) than technical efficiency
change (28.7 per cent). By decomposing this latter variable, pure technical
efficiency change is found to have the highest information content (41.7
per cent) while scale efficiency change has substantially lower information
content. All estimated regression coefficients for TFP change components
(except for scale efficiency) are found to be positive and most of these are
statistically significant at the 1 per cent level showing that positive tech-
nological changes and/or technical efficiency changes lead to greater
shareholder value.
Cost X-efficiency is the determinant of shareholder value driver with the
second largest relative information content with an ability to explain
43.3 per cent of the variation in MAR (Table 7.1). Estimated regression
coefficients are positive and the estimated coefficient at time t–1 is statis-
tically significant at the 1 per cent level. These results provide evidence that
there is a positive relationship between cost efficiency and bank share-
holder value. Alternative profit efficiency is found to have slightly lower
relative information content since its adjusted R-squared is 43.1 per cent.
Similarly to cost X-efficiency, estimated regression coefficients are positive
and the estimated coefficient at time t–1 is statistically significant at 1 per
cent. As such, the relationship between profit efficiency and shareholder
value seems therefore to be positive.
DEA cost efficiency estimates have a relative information content slightly
lower than that of cost X-efficiency estimated using SFA. DEA cost ef-
ficiency seems to be able to explain 39.3 per cent of variation of MAR (see
Table 7.1). Among its components, we find that technical and allocative
efficiency have an equivalent value-relevance (39.5 per cent and 39.3 per
cent, respectively), while scale efficiency has a substantially lower relative
information content (namely, 33.7 per cent). Most of the estimated regres-
sion coefficients for technical and allocative efficiency are positive. Since
we do not find that negative estimated regression coefficients are statisti-
cally significant at least at the 10 per cent level, while several positive
coefficients are highly statistically significant, there seems to be a positive
relationship between technical and allocative efficiency and shareholder

value. Regarding cost efficiency, we find that estimated regression coefficients
at time t and t–1 are positive, while that at time t–2 is negative. All coefficients
are found to be statistically significant at the 10 per cent level. There is a posi-
tive relationship between cost efficiency and shareholder value since the most
recent terms are positive and both positive estimated regression coefficients
have a larger magnitude than that of the negative value.
Customer satisfaction (analysed looking both at depositors and bor-
rowers) is found to be the shareholder value driver with the lowest relative
information content since depositor and borrower satisfaction allows us to
explaining 38.0 per cent and 32.7 per cent, respectively, of variation in
MAR. Estimated regression coefficients are usually positive and the
coefficient at time t–1 is also statistically significant at the 1 per cent level
showing a positive relationship between bank customer satisfaction and
bank shareholder value.
Table 7.2: Incremental information content of cost and profit efficiency estimates
(using SFA) analysing European listed banks between 1997 and 2002. Dependent
variable (Dt): Market Adjusted Return
Panel A – Estimated Regression Coeffficients
Cost X-efficiency (X-EFF) Profit-efficiency (PROF-EFF)
t t–1 t–2 t t–1 t–2

‡
Model 7.9 0.106* 0.229 0.020
Model 7.10 0.089 0.206‡ –0.013 0.086 0.152‡ 0.009
*/ †/ ‡ indicate statistical significance at p<10%, p<5%, and p<1%, respectively
Model 7.9 0.431 – –

Model 7.10 0.466 0.035 6.828‡ 2.040
Panel C – Summary of models run
(7.9) Dt = b0 + b1 X-EFFt + b2 X-EFFt–1 + b3 X-EFFt–2, + e1

(7.10) Dt = b0 + b1 X-EFFt + b2 X-EFFt–1 + b3 X-EFFt–2, + b4 PROF-EFFt–1 +
b5 PROF-EFFt–1 + b6 PROF-EFFt–2 + e1
We also analysed the incremental information content looking at the

adjusted R2 changes and the statistical significance of F-changes running
models 7.3 to 7.10. Our analysis focuses on: (1) cost efficiency components
obtained using DEA; (2) TFP changes components; (3) X-efficiency
and profit efficiency; and (4) depositors’ and borrowers’ satisfaction.
Tables 7.2–7.5 report the results of the incremental information content of
Table 7.3: Incremental information content of Total Factor Productivity change

components analysing European listed banks between 1997 and 2002. Dependent
Pure technical
Technological change efficiency change Scale efficiency
(TECH-sCH) (PUTE-CH) change (SE-CH)
t t–1 t–2 t t–1 t–2 t t–1 t–2

‡ ‡
Model 7.6 0.064 0.294 0.248
Model 7.7 0.092 0.093 0.123† 0.299‡ 0.231‡ 0.126†
Model 7.8 0.073 0.123* 0.097 0.279‡ 0.233‡ 0.087 –0.053 0.002 –0.123
Model 7.6 0.312 – –

Model 7.7 0.436 0.124 16.840‡
Model 7.8 0.440 0.004 1.566 1.917
(7.6) Dt = b0 + b1 TECH-CHt,t–1 + b2 TECH-CHt–1,t–2 + b3 TECH-CHt–2,t–3 + et

(7.7) Dt = b0 + b1 TECH-CHt,t–1 + b2 TECH-CHt–1,t–2 + b3 TECH-CHt–2,t–3 +
b4 PUTE-CHt,t–1 + b5 PUTE-CHt–1,t–2 +b6 PUTE-CHt–2,t–3 + et
b4 PUTE-CHt,t–1 + b5 PUTE-CHt–1,t–2 + b6 PUTE-CHt–2,t–3 +
b7 SE-CHt,t–1 + b8 SE-CHt–1,t–2 + b9 SE-CHt–2,t–3 + et
Table 7.4: Incremental information content of cost efficiency (using DEA)

components analysing European listed banks between 1997 and 2002. Dependent
Technical efficiency Scale efficiency Allocative efficiency

(TE) (SE) (AE)
t t–1 t–2 t t–1 t–2 t t–1 t–2

†
Model 7.3 0.215 –0.013 0.076
Model 7.4 0.184* 0.042 0.066 0.009 0.197* –0.177*
Model 7.5 0.152 0.053 0.064 0.022 0.207* –0.208* –0.045 0.234* –0.142
Model 7.3 0.395 – –

Model 7.4 0.398 0.003 1.535
Model 7.5 0.405 0.007 2.243* 1.915
(7.3) Dt = b0 + b1 TEt + b2 TEt–1 + b3 TEt–2 + et

(7.4) Dt = b0 + b1 TEt + b2 TEt–1 + b3 TEt–2 + b4 AEt + b5 AEt–1 + b6 AEt–2 + et
(7.5) Dt = b0 + b1 TEt + b2 TEt–1 + b3 TEt–2 + b4 AEt + b5 AEt–1 + b6 AEt–2 + b6 SEt +
b6 SEt–1 + b6 SEt–2 + et
the shareholder value drivers. For each of these analyses, we report all
coefficient estimates and their statistical significance in all steps of the
analysis, the adjusted R-squared of all models run in every step of the
analysis, the R-square change,2 F change and its statistical significance and
the Durbin-Watson test of the most complete model.
The analysis of the incremental information content provided by profit
efficiency (see Table 7.2) enables us to assess that the explanatory power of
the cost efficiency model improves by 3.5 per cent (that is statistically
significant at the level 1 per cent level) when we add profit efficiency.
Table 7.5: Incremental information content of depositor and borrower satisfaction

estimates analysing European listed banks between 1997 and 2002. Dependent
Panel A – Estimated Regression Coeffficients
Depositor satisfaction (DS) Borrower satisfaction (BS)
t t–1 t–2 t t–1 t–2

‡
Model 7.11 0.017 0.351 0.012
Model 7.12 0.119 0.321‡ 0.064 –0.190 0.142 –0.091
Model 7.11 0.380 – –

Model 7.12 0.378 –0.002 0.874 1.876
(7.11) Dt = b0 + b1 DSt + b2 DSt–1 + b3 DSt–2 + et

(7.12) Dt = b0 + b1 DSt + b2 DSt–1 + b3 DSt–2 + b4 BSt + b5 BSt–1 + b6 BSt–2 + et
Similarly, the analysis of TFP change components shown in Table 7.3 illus-
trates that only pure technical efficiency changes provide additional infor-
mation content to the explanatory power of technological changes since
the adjusted R-squared increased by 12.4 per cent (i.e. statistically sig-
nificant at the 1 per cent level). In contrast, scale efficiency changes
increased R-squared by 0.4 per cent. The poor additional information
content of scale efficiency is also confirmed when scale efficiency is mea-
sured in terms of levels. The analysis of the DEA cost efficiency com-
ponents (i.e. technical, allocative and scale efficiency) show that only
allocative efficiency provides statistically significant (at the 1 per cent level)
information content to that of technical efficiency (Table 7.4). Finally,
we find that borrower satisfaction does not provide additional informa-
tion content (R-squared falls by 0.2 per cent) to the explanatory power of
depositor satisfaction (see Table 7.5).
In analysing shareholder value creation of European listed banks, it

seems possible to conclude that it is worthwhile to consider both cost
X-efficiency and profit efficiency since the explanatory power of variation
in MAR substantially increases. Similarly, the decomposition of TFP changes
and DEA cost efficiency estimates appear useful for the additional contribu-
tion provided by some of their components, while the analysis of customer
satisfaction seems to show that it is sufficient to focus on depositor satisfac-
tion since borrower satisfaction does not seem to provide any improvement
in explaining bank shareholder value.
Results for European publicly listed and non-listed banks

This section summarises the main results of the analyses carried out to test
the relative and incremental information content of shareholder value
drivers obtained focusing on a larger sample of listed and non-listed banks
in France, Germany, Italy and the UK.
In this section, we assess the relative information content by looking at
differences in the explanatory power (expressed by the adjusted R2) of
regressions where EVAbkg on invested capital at time t–1 is regressed against
various determinants of shareholder value (namely, cost efficiency, profit
efficiency, productivity changes and customer satisfaction). Table 7.6
reports the results and shows: the coefficient estimates3 and their statistical
significance; the adjusted R-squared; the Durbin-Watson test and the
p-values from a two-tailed statistical test assessing the null hypothesis that
all regression coefficients are equal to zero (meaning that the determinants
of shareholder value investigated do not have a statistically significant
impact on EVAbkg on invested capital at time t–1.
Again TFP changes have the highest relative information content among
the shareholder drivers analysed since this shareholder value driver allows
us to explain about 29.3 per cent of the variation of the ratio between
EVAbkg and invested capital at time t–1. All estimated regression coefficients
are found to be positive and statistically significant at the 1 per cent level
showing that TFP improvements lead to increased bank shareholder value.
Among TFP components, technological changes display greater explanatory
power than technical efficiency change. By decomposing this latter vari-
able, we observe that pure technical efficiency change has higher informa-
tion content than scale efficiency change. All estimated regression
coefficients for TFP change components (except for scale efficiency) are
found to be positive and statistically significant showing that positive
technological changes and/or technical efficiency changes lead banks to
generate shareholder value.
Cost X-efficiency is the determinant of shareholder value with the
second largest relative information content with an ability to explain
29.1 per cent (i.e. very close to the TFP change) of variation of the ratio
EVAbkg on invested capital at time t–1. Estimated regression coefficients are
Table 7.6: Relative information content of shareholder value drivers analysing the overall European sample of listed and non-listed
banks (EVAbkg on invested capital = dependent variable)
OVERALL Technical efficiency Allocative efficiency Scale efficiency Cost efficiency Cost X-efficiency Alternative profit
(DEA) (DEA) (DEA) (DEA) (SFA) efficiency (SFA)
t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2
Estimated 0.117‡ 0.271‡ 0.137‡ 0.103‡ 0.130‡ –0.036‡ 0.043‡ 0.009 0.224‡ 0.145‡ 0.222‡ –0.020† 0.196‡ 0.369‡ 0.031‡ 0.029‡ 0.373‡ –0.008
coefficients
Adj. R2 0.207 0.047 0.077 0.104 0.291 0.168
P–value & F-stat 0 (F=304.583) 0 (F=57.273) 0 (F=96.369) 0 (F=135.307) 0 (F=462.308) 0 (F=225.628)
DW 1.948 2.084 2.049 2.103 1.891 2.046
Period analysed 1997–2002 1997–2002 1997–2002 1997–2002 1997–2002 1997–2002
The p-value reported is the marginal significance level of the F-test (reported in brackets): it expresses the probability of incurring an error once it is rejected the null hypothesis that all slope
coefficients are equal to zero.
OVERALL Technical Technological Pure technical Scale efficiency Total Factor Depositor Borrower
efficiency change change efficiency change change Productivity change satisfaction satisfaction
t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2
‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡ ‡
Estimated 0.230 0.073 0.091 0.013* 0.512 0.126 0.293 0.228 0.237 –0.076 –0.142 –0.088 0.055 0.483 0.154 0.017 0.575 –0.156 0.090 0.464 –0.073‡
coefficients
Adj. R2 0.094 0.305 0.235 0.060 0.293 0.264 0.265
P-value & F-stat 0 (F=93.739) 0 (F=395.137) 0 (F=276.633) 0 (F=58.570) 0 (F=370.856)0 (F=267.925) 0 (F=269.188)
DW 1.925 1.906 1.943 1.850 1.972 2.054 2.049
Period analysed 1998–2002 1998–2002 1998–2002 1998–2002 1998–2002 1999–2002 1999–2002
The p-value reported is the marginal significance level of the F-test (reported in brackets): it expresses the probability of incurring an error once it is rejected the null hypothesis that all slope
271
coefficients are equal to zero.

positive and statistically significant at the 1 per cent level providing evi-
dence that there is a positive relationship between cost X-efficiency and
bank shareholder value. Alternative profit efficiency is found to have a sub-
stantially lower explanatory power since its adjusted R-squared is 16.8 per
cent. Estimated regression coefficients at times t and t–1 are positive and
statistically significant at the 1 per cent level so it seems possible to con-
clude that profit efficiency and shareholder value are positively related.
DEA cost efficiency estimates have a substantially lower explanatory
power than that of cost X-efficiency estimates using SFA. DEA cost
efficiency seems to be able to explain 10.4 per cent of variation of the ratio
EVAbkg on invested capital at time t–1. Among its components, we find that
technical efficiency has the highest adjusted R2, while the other compo-
nents have a substantially lower explanatory power (namely, 4.7 per cent
allocative efficiency and 7.7 per cent scale efficiency). All estimated regres-
sion coefficients are positive and statistically significant at the 1 per cent
level showing that technical, allocative and scale efficiency have a positive
influence on shareholder value.
Customer satisfaction is measured looking at both depositors and bor-
rowers. We find that both depositors’ and customers’ satisfaction have a
relative information content slightly lower than cost X-efficiency allowing
us to explain around 26.5 per cent of variation of the ratio EVAbkg on
invested capital at time t–1 (see Table 7.6). Estimated regression coefficients
at times t and t–1 are positive. Since most of the coefficients are statistically
significant at the 1 per cent level and the magnitude of positive coefficients
is larger than that of the negative coefficient, it seems possible to posit a
positive relationship between customer satisfaction and shareholder value.
The incremental information content is assessed looking at the adjusted
R2 changes and the statistical significance of F-changes running models 7.3
to 7.10 and, similarly to our analysis for the sample of listed banks, we focus
on: (1) cost efficiency components obtained using DEA; (2) TFP changes
components; (3) cost X-efficiency and alternative profit efficiency; and (4)
customer satisfaction. Tables 7.7 to 7.10 report our findings about the incre-
mental information content of these determinants of shareholder value.
In all analyses carried out, we found that it is worthwhile to consider the
various components in analysing the creation of shareholder value since
these provide the other components with additional information that is
statistically significant at the 1 per cent level, although improvements in
the explanatory power are often small. For example, we find that pure tech-
nical efficiency changes and scale efficiency changes increased the R2 by 7.6
per cent and 0.8 per cent, respectively, and these variables (plus techno-
logical change) enable us to explain 38.9 per cent of shareholder value vari-
ation. Profit efficiency increases the explanatory power of cost efficiency by
1.8 per cent and, globally, these two variables enable us to explain 32.5 per
cent of shareholder value variation. Scale and allocative efficiency improve
Table 7.7: Incremental information content of cost efficiency (using DEA)

components analysing the overall European sample of listed and non-listed banks
between 1997 and 2002 (EVAbkg on invested capital (Dt) = dependent variable)
Technical efficiency Scale efficiency Allocative efficiency

(TE) (SE) (AE)
t t–1 t–2 t t–1 t–2 t t–1 t–2
Model 7.3 0.117‡ 0.271‡ 0.137‡

Model 7.4 0.098‡ 0.232‡ 0.153‡ 0.022* 0.007 0.123‡
Model 7.5 0.097‡ 0.224‡ 0.144‡ 0.027† 0.003 0.117‡ 0.061‡ 0.062‡ –0.029‡
Model 7.3 0.207 – –

Model 7.4 0.223 0.016 95.933‡
Model 7.5 0.230 0.007 44.218‡ 1.870
(7.3) Dt = b0 + b1 TEt + b2 TEt–1 + b3 Tt–2 + et

(7.4) Dt = b0 + b1 TEt + b2 TEt–1 + b3 Tt–2 + b4 AEt + b5 AEt–1 + b6 AEt–2 + et
(7.5) Dt = b0 + b1 TEt + b2 TEt–1 + b3 Tt–2 + b4 AEt + b5 AEt–1 + b6 AEt–2 +
b7 SEt + b8 SEt–1 + b9 SEt–2 + et
the adjusted R2 of technical efficiency by 1.6 per cent and 0.7 per cent,
respectively, and these three variables enable us to explain 23.0 per cent of
shareholder value variation. Finally, borrower satisfaction increases the
depositor satisfaction explanatory power by 4.3 per cent and globally these
two variables enable us to explain 30.8 per cent of shareholder value vari-
ation. As such, one can conclude that, although all variables analysed
provide statistically significant incremental information content, TFP com-
ponents and customer satisfaction appear to be more important than the
Table 7.8: Incremental information content of Total Factor Productivity change

components analysing overall European sample of listed and non-listed banks
Technological change Pure technical efficiency Scale efficiency

(TECH-CH) change (PUTE-CH) change (SE-CH)
t t–1 t–2 t t–1 t–2 t t–1 t–2
Model 7.6 0.013* 0.512‡ 0.126‡

Model 7.7 0.018‡ 0.404‡ 0.096‡ 0.199‡ 0.160‡ 0.137‡
Model 7.8 0.017† 0.392‡ 0.098‡ 0.196‡ 0.159‡ 0.134‡ –0.054‡ –0.062‡ –0.046‡
Model 7.6 0.305 – –

Model 7.7 0.381 0.076 443.654‡
Model 7.8 0.389 0.008 46.875‡ 1.936
(7.6) Dt = b0 + b1 TECH-CHt,t–1 + b2 TECH-CHt–1,t–2 + b3 TECH-CHt–2,t–3 + et

b4 PUTE-CHt,t–1 + b5 PUTE-CHt–1,t–2 + b6 PUTE-CHt–2,t–3 + et
b4 PUTE-CHt,t–1 + b5 PUTE-CHt–1,t–2 + b6 PUTE-CHt–2,t–3 +
b7 SE-CHt,t–1 + b8 SE-CHt–1,t–2 + b9 SE-CHt–2,t–3 + et
other determinants in terms of creating shareholder value in European

banking.
We have discussed above the results for the pooled sample of European
listed and non-listed banks. We also repeat the estimation exercise for each
banking system under study. A summary of the relative information
findings is given in Table 7.11. Across the four banking systems analysed
we note that TFP changes have the highest explanatory power of variation
Table 7.9: Incremental information content of cost and profit efficiency estimates
(using SFA) analysing the overall European sample of listed and non-listed banks
Cost X-efficiency (X-EFF) Profit efficiency (PROF-EFF)
t t–1 t–2 t t–1 t–2
Model 7.9 0.192‡ 0.386‡ 0.032‡

‡ ‡
Model 7.10 0.174 0.330 0.028‡ –0.062‡ 0.185‡ 0.007
Model 7.9 0.307 – –

Model 7.10 0.325 0.018 121.897‡ 1.859
(7.9) Dt = b0 + b1 X-EFFt + b2 X-EFFt–1 + b3 X-EFFt–2, + e1

(7.10) Dt = b0 + b1 X-EFFt + b2 X-EFFt–1 + b3 X-EFFt–2, + b4 PROF-EFFt–1 +
b5 PROF-EFFt–1 + b6 PROF-EFFt–2 + e1
of the ratio between EVAbkg and invested capital at time t–1 in Germany,
Italy and the UK. All regression coefficients in all countries are found to be
positive and highly statistically significant providing clear evidence that
productivity improvements create shareholder value. By analysing the com-
ponents of TFP changes, we find that technological change has a larger
information content than technical efficiency change in France, Germany
and the UK, while in Italy technical efficiency change seems to have a
superior value relevance than technological change. The estimated regres-
sion coefficients of all TFP components (except scale efficiency) are gener-
ally positive and statistically significant in all countries. As such, it is
possible to conclude for all four banking systems analysed that improve-
ment in these TFP components seem to lead banks to generate shareholder
value. One might claim that pure technical efficiency changes are found to
have a value-relevance higher than technical efficiency change. The reason
Table 7.10: Incremental information content of depositor and borrower satisfaction

estimates analysing the overall European sample of listed and non-listed banks
Depositor satisfaction (DS) Borrower satisfaction (BS)
t t–1 t–2 t t–1 t–2
Model 7.11 0.017 0.575‡ –0.156‡

† ‡
Model 7.12 –0.039 0.390 –0.126‡ 0.072‡ 0.264‡ –0.042‡
Model 7.11 0.265 – –

Model 7.12 0.308 0.043 167.346‡ 2.007
(7.11) Dt = b0 + b1 DSt + b2 DSt–1 + b3 DSt–2 + et

(7.12) Dt = b0 + b1 DSt + b2 DSt–1 + b3 DSt–2 + b4 BSt + b5 BSt–1 + b6 BSt–2 + et
is probably due to limited poor relative information content of efficiency

changes (common to all countries analysed): in other words, when we
decompose technical efficiency into two components, we find that the
influence of pure technical efficiency change is high whereas the influence
of scale efficiency changes is low.
The explanatory power of cost efficiency seems to substantially change
across countries and methodology applied. In three of the four banking
system analysed, cost efficiency estimates tend to have a greater adjusted R2
when estimated using SFA rather than DEA. In both sets of cost efficiency
estimates, estimated regression coefficients are generally positive and statis-
tically significant providing us with strong evidence that cost efficiency
improvement leads to greater shareholder value creation. Among DEA cost
efficiency components, scale efficiency is found to have always the lowest
explanatory power of variation of the ratio between EVAbkg and invested
Table 7.11: Relative information content of shareholder drivers using the sample of listed and non-listed banks: a comparison among
countries (EVAbkg on invested capital (Dt) = dependent variable)
France Germany Italy United Kingdom
Estimated regression Estimated regression Estimated regression Estimated regression

coefficients and coefficients and coefficients and coefficients and
statistical significance statistical significance statistical significance statistical significance
t t–1 t–2 t t–1 t–2 t t–1 t–2 t t–1 t–2
Value relevance
Value relevance
Value relevance
Value relevance
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Technical efficiency (DEA) (++)1 (++)1 (–)5 15.6 (++)1 (++)1 (–)1 32.7 (++)1 (++)1 (– –)1 21.3 (– –)1 (– –)10 (++)1 12.7
Allocative efficiency (DEA) (–)5 (++)1 (–)1 8.6 (+)1 (++)1 (+)1 10.7 (+)>10 (++)1 (++)1 22.7 (– – –)1 (+++)1 (–)>10 20.9
Scale efficiency (DEA) (+)>10 (+)5 (+)5 7.0 (++)1 (+)1 (++)1 16.5 (–)>10 (–)1 (++)1 14.6 (++)5 (– – –)1 (++)1 12.4
Cost efficiency (DEA) (+)5 (+++)1 (– –)1 20.9 (++)1 (++)1 (+)1 19.6 (+)1 (++)1 (+)1 28.3 (– –)1 (+++)1 (– –)5 25.4
Cost X–efficiency (SFA) (++) 1 (++) 1 (–) >10 34.4 (++) 1 (+++) 1 (–) 1 39.7 (++)1 (++) 1 (+)1 24.1 (++)1 (++) 1 (++)1 30.5
Alternative profit efficiency (+) >10 (++) 1 (–) 10 25.4 (– –) 1 (+++) 1 (–) 1 15.9 (++)1 (++) 1 (+)1 28.3 (++)1 (++) 1 (–)>10 20.9
Technical efficiency change (++) 1 (+) 1 (+) 1 8.6 (++) 1 (++) 1 (+) 1 19.0 (++)1 (++) 1 (++)1 21.3 (++)1 (+) >10 (+)>10 22.6
Technological change (–) 10 (+++) 1 (+) 1 22.7 (+) 1 (++) 1 (+) 1 22.9 (+)1 (++) 1 (+)1 17.2 (++)1 (++) 1 (++)1 30.2
Pure technical efficiency change (++) 1 (++) 1 (++) 1 19.9 (++) 1 (+) 1 (+) >10 22.2 (++)1 (++) 1 (+)1 44.2 (+)>10 (++) 1 (++)1 41.0
Scale efficiency change (–) >10 (–) 1 (+) 1 5.0 (–) 1 (– –) 1 (–) >10 5.8 (– –)1 (– –) 1 (+)1 9.3 (–)5 (–) 5 (–)1 15.7
TFP change (+) >10 (++) 1 (++) 1 26.9 (++) 1 (+++) 1 (+) 1 42.5 (++)1 (+++) 1 (++)1 38.0 (++)1 (++) 1 (++)1 37.4
Depositor satisfaction (++) 5 (+++) 1 (–) 1 36.1 (+) 10 (+) 1 (++) 1 12.6 (+)>10 (++) 1 (++)1 34.5 (–)5 (+++) 1 (– –)1 29.3
Borrower satisfaction (++) 1 (++) 1 (–) >10 35.9 (+) 1 (++) 1 (–) 1 18.1 (–)1 (+++) 1 (+)5 34.0 (–)>10 (+++) 1 (– –) 29.0
(+) / (++) / (+++) Estimated regression coefficients range, respectively, between 0 and 0.15, between 0.151 and 0.5, between 0.51 and 1
(–) / (– –) / (– – –) Estimated regression coefficients range, respectively, between 0 and –0.15, between –0.15 and –0.5, between –0.5 and –1
277
capital at time t–1. In France and Germany, technical efficiency seems to

have a higher value-relevance than allocative efficiency, while we note the
opposite in Italy and the UK.
By comparing pure technical and scale efficiency changes with, respec-
tively, technical and scale efficiency, we observe in France, Italy and the UK
that pure efficiency changes have a substantially higher ability in explain-
ing variation of the ratio between EVAbkg and invested capital at time t–1
than that of technical efficiency. These differences seem to provide evid-
ence that it is preferable to measure efficiency in terms of changes (rather
than levels as in DEA estimates) in assessing the value-relevance of the
determinants of shareholder value.
An interesting finding is that profit efficiency is found to have lower
value relevance than cost X-efficiency when we look at both listed and
unlisted banks. This result appears unusual since one would expect the rela-
tionship between profit efficiency and shareholder value to be stronger
than that between cost efficiency and shareholder value. As a possible
explanation, we note that banks may increase their profits by engaging in
more risky activities. As such, the opportunity cost of capital increases since
shareholders would require higher returns. Instead, by improving cost
efficiency, a bank will probably increase its profits without increasing risks
and, consequently, creating value for shareholders. As such, it is possible
that cost X-efficiency has a closer statistical association with shareholder
value than profit efficiency. This explanation seems to be partially con-
firmed by the analysis of the incremental information of profit efficiency
since we find (in all countries analysed) that profit efficiency provides sub-
stantial (and statistically significant) additional information to the value-
relevance of cost X-efficiency. Finally, we observe that our findings for the
sample of listed and non-listed banks appear to be broadly consistent with
those for the sample of only listed banks.
The concept of shareholder value efficiency

The analysis presented so far in this chapter provides the first empirical evi-
dence (as far as we are aware) that banks can create shareholder value
through cost and profit efficiency improvements, enhancing customer sat-
isfaction and especially increasing Total Factor Productivity. These factors
were found to explain about a third of variation of shareholder value in
European banking. In the final part of this chapter, we advance a new
measure of efficiency labelled ‘shareholder value efficiency’. We define a
bank that is able to produce at the maximum possible shareholder value,
given a particular level of output levels, as ‘shareholder value efficient’.
Following Berger and Mester (1997), who developed the alternative profit
efficiency measure, we define two value maximisation concepts depending
on which business conditions are taken as given. The two concepts are:
• standard shareholder value maximisation; and

• alternative shareholder value maximisation.
The standard shareholder value efficiency measures how close a bank is to
producing the maximum possible shareholder value given a particular level
of input prices and output prices (and other variables). The standard share-
holder value maximisation is implemented using a shareholder value func-
tion that specifies output prices in the business conditions vector in place
of the output quantities specified in the cost function, but all other busi-
ness conditions remain the same. Thus, output prices are taken as exoge-
nous and firms are assumed to choose their output in response to relative
output prices and other factors in the maximisation process. The standard
shareholder value function is given by:
ln (τ + φ ) = ƒ π (X π ) + ln uπ + ln επ (7.13)
where τ is the variable shareholder value, φ is the scalar added to every

firm’s dependent variable in a given time period before logging, so that the
log is taken of a positive number (φ varies over time), Xn = (lnw, lnp, lnz, lnv)
is the set of logged exogenous ‘business conditions’ that affect costs,
specifically, variable input prices (lnw), logged output prices (lnp,), fixed
netput quanties (lnz) and environmental variables (lnv), fτ(.) is a best prac-
tice shareholder value function, ln uτ is an inefficiency factor that is zero
for the best-practice firms and negative for other terms, reducing their
shareholder value below the best practice level, ln ετ is a random error with
mean of zero each period.
The standard shareholder value efficiency is defined as the ratio of the
predicted actual shareholder value to the predicted maximum shareholder
value that could be earned if the bank was as efficient as the best bank in
the sample, net of random error, or the proportion of maximum share-
holder value that is actually generated. For example, the standard share-
holder value efficiency (std τ-EFF) of bank h at period t is given as:
τh {exp[fτt ( Xτth )] • exp[ln uτth ] – φt}

Std τ – EFFh = = (7.14)
τ max
{exp [fτt ( Xth )] • exp[ln uτtmax ] – φt}
A standard shareholder value efficiency ratio of 0.80 would indicate that

bank h has lost in period t about 20 per cent of shareholder value that
could be earned due to excessive costs, deficient revenues, or both or to
excessive opportunity cost of invested capital. If the standard shareholder
value efficiency is equal to 1, bank h would be a best-practice firm, max-
imising shareholder value given the operating conditions. It is worthwhile
to note that, compared to cost X-efficiency estimates, shareholder value
efficiency (as profit efficiency) ratios can be negative since a firm can throw
away more than 100 per cent of the potential shareholder value.
We develop a second value maximisation concept labelled ‘alternative

shareholder value efficiency’ to distinguish from the previous concept. The
alternative shareholder value efficiency measures how close a bank comes
to earning maximum shareholder value given its output levels rather than
its output prices. In other words, alternative shareholder value maxi-
misation has the same objective as the standard shareholder value maxi-
misation concept, but the logged output quantities (lny) are specified in the
X vector rather than the logged output prices (lnp). The alternative share-
holder value function is given by:
ln (τ + φ) = ƒaτ (Xc) + ln uaτ + ln εaτ (7.15)
where Xc = (lnw, lny, lnz, lnv) is the set of logged exogenous ‘business con-
ditions’ that affect costs, specifically, variable input prices (lnw), logged
output quantities (lny,), fixed netput quantities (lnz) and environmental
variables (lnv). Therefore, the alternative shareholder value function
employs the same dependent variable as the standard shareholder value
function and the same exogenous variables as the cost function. This
efficiency measure expresses the proportion of maximum shareholder value
that is earned, like the standard shareholder value efficiency approach.
These efficiency ratios are based on a comparison with the best-practice
point of shareholder value maximisation within the data set. The alter-
native shareholder value efficiency (alτ-EFF) of the bank h at the period t is
given as:
aτh {exp [faτ t ( Xch )] • exp [ln uaτ th ] – φt}
alτ-EFFh = = (7.16)
aτ max
{exp [faτ t ( Xch )] • exp [ln uaτ tmax ] – φt}
Comparing these two shareholder value efficiency measures, the standard

shareholder value efficiency approach seems to be preferable to the alterna-
tive shareholder value efficiency since it is unrealistic that firms take their
outputs as given and maximise shareholder value, as the alternative
approach implies. Similarly to profit efficiency estimates, the alternative
shareholder value function should be employed when the assumptions
behind the standard shareholder value maximisation do not hold precisely:
standard efficiency appropriately measures how well the firm is producing
output and how it is employing inputs relative to the best practice firms,
given the underlying assumptions.
Berger and Mester (1997: 902) identified four conditions under which
their alternative profit concept may provide useful information in efficiency
measurement. In our opinion, these conditions can be applied as well to
shareholder value efficiency and so alternative shareholder value efficiency
may provide useful information when: (1) there are substantial unmeasured
differences in the quality of banking services; (2) outputs are not completely
variable, so that a bank cannot achieve every output scale and product mix;
(3) output markets are not perfectly competitive, so that banks have some
market power over the process they charge; and (4) output prices are not
accurately measured, so they do not provide accurate guides to opportun-
ities to earn revenues and profits in the standard profit function.
Regarding banks, Berger and Mester (1997: 906) observe that the fourth
condition is very relevant in banking, since output prices are usually not
accurately measured and, regarding the estimation of profit efficiency, note
that ‘if prices are inaccurately measured – as is likely, given the available
banking data – the predicted part of the standard profit function would
explain less of the variance of profits and yield more error in the estimation
of the efficiency terms ln uπ. In this event, it may be appropriate to try spec-
ifying other variables in the profit function that might yield a better fit,
such as the output quantity vector, y, as in the alternative profit function.’
Since the construction of shareholder value concepts follows that of profit
efficiency, it seems reasonable to extend this point to shareholder value
efficiency. As such, when output prices are not completely reliable, we con-
sider that the standard shareholder value function would be imprecisely
estimated and the alternative profit function is very useful in providing
profit efficiency estimates which do not depend on output prices.
Measuring shareholder value efficiency

Since output prices are usually not accurately measured in banking, we
prefer to estimate alternative shareholder value efficiency rather than stan-
dard shareholder value. The alternative shareholder value function is esti-
mated using the following standard translog functional form:4
n m
ln (τ + φ) = α0 + ∑ αi ln yi + ∑ βj ln wj
i=1 j=1
 n m m m 
1 
+ 
 ∑ ∑ δij ln yi ln yj + ∑ ∑ γij ln wi ln wj  + (8.5)
2  i=1 j=1 i=1 j=1 
n m
i=1 j=1
where τ is Economic Value Added, estimated using a procedure accounting

for banking peculiarities (EVAbkg), standardised by w3 (i.e. the average cost
of capital measured as total capital expenses over total fixed assets), φ is a
constant, yi is the output quantities for the i-th output, wi is the input prices
for the i-th input, n is the number of outputs, m is the number of inputs.
outputs (that we already adopted running SFA and DEA for measuring
cost efficiency), we posit5 that labour (measured as personnel expenses),
physical capital (expressed as the average value of fixed-tangible assets) and
τ – Economic Value Added, estimated using a procedure accounting for banking peculiarities (EVAbkg)
φ is a constant, i.e. defined as φ = | (τ / w3)min | + 1]

Input 3 (w3) – average cost of financial capital deposits (interest costs on borrowed funds on the average
amount of borrowed funds);
Figure 7.3: Inputs and outputs used for estimating alternative shareholder value
efficiency
financial capital (measured as loanable funds) are inputs, whereas demand

This specification is essentially the same as that adopted to estimate cost
X-efficiency and profit efficiency with a few changes. The dependent vari-
able for the alternative shareholder value function replaces the normalised
ln (p + θ) (used in estimating alternative profit efficiency, respectively) with
ln (t + φ), where τ is EVAbkg (standardised by the average cost of capital) and φ
is a constant defined by adding 1 to the absolute value of the lowest (t/w3) in
the sample [i.e. φ = | (τ/w3)min | + 1] in order to make positive the natural log
of bank’s EVAbkg. The standard Stochastic Frontier Analysis (SFA) is employed
to estimate alternative shareholder value efficiency for each bank over the
years 1995–2002: Battese and Coelli’s (1995) Stochastic Frontier Model has
been adopted since this model enables us to control for the effects of differ-
ent bank ownership on our shareholder value efficiency estimates.
Sample description
The sample adopted for estimating alternative shareholder value efficiency
is the same as employed in Chapter 6 for estimating cost and alternative
profit efficiency using the Stochastic Frontier Model. As such, we estimate
shareholder value efficiency frontiers using annual cross-sectional samples
for eight years (from 1995 to 2002), for two broad types of bank (firstly,
commercial banks and, secondly, cooperative and savings banks) and for
three countries (France, Italy and Germany). For the UK, we estimate an
annual cross-sectional efficiency frontier for eight years (1995–2002) using
a sample comprising only commercial banks. Table 7.12 provides the
number of banks considered for estimating each frontier using SFA.
We use a cross-section sample by year since many bank observations
would have been lost selecting a balanced panel data set. We prefer to use
an individual country sample (domestic sample) for estimating the profit
efficiency frontier since banks in the same country are more homogeneous
Table 7.12: Number of banks in samples used for estimating efficiency with SFA and DEA
1995 1996 1997 1998 1999 2000 2001 2002
Cooperative &
savings banks* 100 109 117 119 123 123 121 122
France
(79;21) (88;21) (92;25) (94;25) (98;25) (94;29) (96;25) (94;28)
Cooperative &
savings banks* 1093 1260 1397 1514 1699 1825 1980 1977
Germany
(584;589) (714;546) (832;565) (930;584) (1102;597) (1220;605) (1368;612) (1378;529)

Cooperative &
Italy
savings banks* 146 168 195 329 450 531 570 585
(87;59) (108;60) (133;62) (266;63) (387;63) (465;66) (506;64) (519;66)
Cooperative &
United
savings banks* N/A N/A N/A N/A N/A N/A N/A N/A
Kingdom
Source: Bankscope.
283
(and comparable) than banks working in different countries. Similarly, we

consider various specific sub-samples according to bank type (namely, com-
mercial banks and, jointly, cooperative and savings banks) since this seems
to guarantee a higher homogeneity to the sample. As such, we estimate
fifty-six alternative shareholder value efficiency frontiers.
Shareholder value efficiency: results

Table 7.13 reports descriptive statistics of the alternative shareholder value
efficiency measures derived from SFA. Over the period analysed
(1995–2002), European banks display shareholder value inefficiency scores
ranging between 22.17 per cent (German commercial banks in 1999) and
54.63 per cent (French cooperative banks in 1998). On average (considering
all 19,887 shareholder value efficiency estimates obtained running fifty-six
annual cross-sectional domestic frontiers), the mean alternative share-
holder value efficiency is 64.06 per cent and the mean level of dispersion of
average efficiency scores is 11.00 per cent.
Focusing on the four banking systems analysed, UK banks display a
slightly higher shareholder value efficiency level (on average, 65.67 per
cent between 1995 and 2002) than France and Italy. Mean efficiency levels
in UK banking decreased between 1995 and 1996 (from 60.71 per cent to
57.45 per cent) and between 2001 and 2002 (from 69.71 per cent to 65.52
per cent) while they increased between 1996 and 2001 (from 57.45 per cent
to 69.71 per cent). These data enable us to observe that UK banks (which
we found in Chapter 5 to generate higher levels of shareholder value than
the other three banking systems) do not exploit about one-third of the
potential shareholder value. The increasing mean efficiency levels observed
between 1996 and 2001 provide evidence that the mean distance between
best-practice banks and shareholder value inefficient banks reduced over
the time period analysed. However, it is not possible to affirm that share-
holder value creation improved since efficiency levels express a relative val-
uation of cross-sectional data and do not enable us to consider efficiency
change over time.
We found that Italian and German banks have slightly lower mean levels
of shareholder value efficiency and of dispersion of average efficiency
scores than UK banks. German and Italian banks exhibit mean shareholder
value efficiency of 65.02 per cent and 63.80 per cent, respectively, and a
level of dispersion of average efficiency scores of 8.49 per cent and 13.61
per cent, respectively. In France, the mean shareholder value efficiency
between 1995 and 2002 is 58.39 per cent and a level of dispersion of
average efficiency scores of 19.01 per cent.
Looking at the different bank specialisations mean shareholder value
efficiency levels exhibit slight differences across different type of banks. On
average, the mean shareholder value efficiency between 1995 and 2002 is
62.46 per cent for commercial banks, 65.19 per cent for cooperative banks
Table 7.13: Shareholder value efficiency scores in European banking between 1995 and 2002*
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
1995
Total 68.50 9.79 65.52 8.51 66.29 16.13 60.71 21.79 65.90 10.08
1996
Total 63.49 13.71 60.12 7.98 65.81 10.56 57.45 33.22 61.21 9.92
1997
Total 57.91 14.58 65.20 9.94 64.78 19.30 64.47 17.87 64.14 12.03
1998
Total 49.38 28.73 57.64 11.99 62.60 15.90 67.35 21.53 57.82 14.99
1999
Total 58.72 32.96 75.73 5.53 62.86 13.09 64.37 24.18 70.90 10.65
285
286
Table 7.13: Shareholder value efficiency scores in European banking between 1995 and 2002* – continued
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
2000
Total 51.93 18.49 66.35 8.80 67.16 16.10 69.95 20.01 65.17 11.69
2001
Total 63.62 15.43 66.25 6.81 62.00 11.55 69.71 23.28 65.19 9.07
2002
Total 55.76 15.99 61.75 9.03 62.16 10.47 65.52 12.72 61.38 10.10
Mean
Total 58.39 19.01 65.02 8.49 63.80 13.61 65.67 21.18 64.06 11.00
1995–2002
* Efficiency estimates for commercial banks are obtained by estimating an efficient frontier in a sample comprising only this category of bank, while efficiency
estimates for savings and cooperative banks are obtained by estimating a common frontier using a sample comprising both savings and cooperative banks.
and 62.79 per cent for savings banks. These data provide evidence that
commercial banks achieved a percentage of their potential shareholder
value that is almost 3 per cent lower than that of cooperative banks. These
results should not be interpreted as a signal that cooperative banks created
larger shareholder value than commercial banks since these efficiency esti-
mates express the mean distance between best-practice and inefficient
banks using cross-sectional data. Furthermore, we find similar mean levels
of shareholder value efficiency for all types of banks in Italy, while savings
banks seem to be, on average, slightly less efficient than cooperative and
commercial banks in France and Germany.
In conclusion, we found that, on average, banks in the four European
banking systems analysed achieved only two-thirds of their potential share-
holder value between 1995 and 2002. German, Italian and UK banks are
found to have similar mean levels of shareholder value efficiency (around
65 per cent) while French banks achieve a substantially lower mean level.
One might question that this result contrasts with our previous findings
(discussed in Chapter 5) about the shareholder value generated by Euro-
pean banks (we found that UK banks outperform French banks in terms
of profits and EVAbkg, French banks outperform Italian banks and Italian
banks outperform German banks).
However, these results are consistent since shareholder value efficiency
levels express the existing gap between the best-practice banks in trans-
forming bank inputs into shareholder value and non-efficient banks and
larger mean inefficiency levels express only a larger gap between efficient
and inefficient banks in the market. For example, we estimate that mean
shareholder value inefficiency levels in the UK are equivalent to France and
Italy and this mean level is substantially higher than the mean profit
inefficiency. These results are strongly consistent with our findings (dis-
cussed in Chapter 5) when we estimated that almost all UK banks generally
generate higher profits than banks in the other countries (this explain why
mean profit efficiency is larger for UK banks), but not all UK banks are able
to achieve profits superior to the opportunity cost of equity capital.
In addition, we also investigated the value relevance of our new share-
holder efficiency measure. Our results (not reported here) provide strong
evidence that there is a positive relationship between shareholder value
efficiency and bank shareholder value. In other words, a bank can generate
value for its shareholders when this bank is able to reduce the distance
between best-practice banks in creating value for their shareholders. We
found that shareholder value efficiency has higher relative information
content than that estimated for cost X-efficiency and alternative profit
efficiency. This superiority is very small for listed banks, while it is substan-
tial for non-listed banks. As such, shareholder value efficiency seems to be
particularly useful for non-listed banks for investigating the ability of these
banks to create value for their shareholders/stakeholders.
Conclusion
The first part of this chapter investigated the drivers of shareholder value
for both listed and non-listed European banks. Here we found that Total
Factor Productivity (and its components) best explain variation in Euro-
pean bank shareholder value creation between 1995 and 2002. Cost
inefficiency is also found to be more important in explaining shareholder
value than profit efficiency. The second part of the chapter uses the alter-
native profit efficiency estimation framework advanced by Berger and
Mester (1997) to develop a new efficiency measure which we call ‘alterna-
tive shareholder value efficiency’. According to our estimates, European
banks display shareholder value inefficiency scores that suggest, on
average, banks are 36 per cent shareholder value inefficient, although the
situation varies across years, countries and types of banks. Among the four
banking systems analysed, we found that Italian, German and UK banks
have similar levels of shareholder value efficiency levels, while French
banks are found to generate only 58.39 per cent of their potential share-
holder value. These results appear consistent with our previous findings
about the shareholder value generated by European banks. We also find
that shareholder value efficiency seems to be particularly useful for non-
listed banks for investigating the ability of these banks to create value for
their shareholders/stakeholders.
8
Conclusions
The main aim of this book has been to examine the drivers of shareholder
value in banking and to illustrate empirical relationships between factors
that are believed to add to the wealth of bank owners. The first part of the
book provides a framework for analysing shareholder value theory by dis-
cussing how shareholder value can be defined, if it can be considered a
valid strategic objective for banks, how shareholder value can be measured
and how it can be created. The second part presents various empirical
investigations in order to measure shareholder value (using the Economic
Value Added approach) and some of its drivers (such as cost and profit
efficiency, productivity changes and customer satisfaction). The final part
analyses the importance of these drivers in creating shareholder value and
also briefly develops a new measure of bank efficiency (shareholder value
efficiency).
In our early discussions we examine the economic objectives of a bank
showing that managing to create sustained and sustainable shareholder
value is an important goal for European and other banks. In Chapter 2 we
outline the economic foundations of the theory of the firm by discussing the
marginalists and behaviouralists and the debate between shareholder and
stakeholder models. Ultimately we show that although it is undeniable that
stakeholders (such as customers, shareholders and managers) have their
own interests, these objectives are not incompatible and managing to
create shareholder value is not a zero-sum game where the shareholders
prosper at the expense of other stakeholders. On the contrary, we note that
generating stable shareholder value growth requires an intense focus on
delivering benefits to customers in the most efficient way, hiring and
retaining a motivated workforce, maintaining excellent supplier relation-
ships, and being a good corporate citizen in each of the company’s local
communities (the so-called ‘stakeholder symbiosis’). The service value chain
shows the long-term relationship between customer satisfaction, customer
loyalty, productive efficiency, employee motivation and satisfaction and
shareholder value. In a competitive market, shareholder value is based on
289
customer satisfaction and productive efficiency, which requires employees’

and managers’ satisfaction.
Our discussion on the foundations of shareholder value theory focuses
on four main issues: the concept of shareholder value and wealth, the evo-
lution of shareholder value theory, measurement performance methodo-
logies and a review of the empirical literature. Regarding the first issue, we
adopt a classic definition of value: a company creates value for the share-
holders over a given time period when the return on invested capital is
greater than its opportunity cost, or than the rate that investors could earn
by investing in other securities with the same risk. In order to highlight
that this concept of ‘value’ implies a comparison of the shareholder return
with opportunity cost, the word ‘added’ often joins the term ‘value’ in
some new performance measures (such as Economic Value Added).
Regarding the second issue, we focus on two seminal works: Miller and
Modigliani (1961) and Porter (1980). Miller and Modigliani (1961), com-
monly considered as the start of Shareholder Value Theory, show the irrele-
vance of dividend (distribution) policies in an ‘ideal’ environment, where
the value of a business is linked to the present value of expected dividends
and the company’s residual value. This division is the starting point for
most of the subsequent research on financial evaluation methods. The
value-chain concept developed in Porter (1980) is the key to understanding
competitive advantage and, as such, to creating sustainable shareholder
value. With regards to the third issue, Chapter 3 reviews the main tech-
niques proposed by consultants to measure company performance and the
creation of shareholder value, such as Discounted Cash Flow; Shareholder
Value Added; Cash Flow Return on Investment; Risk Adjusted Performance
Measures; the Market Value Added and Economic Value Added (EVA).
Finally, the chapter presents a detailed review of methodologies and
findings of empirical studies dealing with shareholder value.
Chapter 4 analyses how commercial banks can create shareholder value
by outlining a set of strategies that appear able to improve, ceteris paribus, at
least one of the determinants of shareholder value. The chapter outlines
these strategies by distinguishing between endogenous and exogenous
channels (i.e. strategies implemented within the bank and strategies involv-
ing external parties). Regarding the endogenous channels, these relate to
improvements in customer satisfaction, greater cost or profit efficiency, the
redefinition of the business mix and achievement of optimal capital struc-
tures. The external channel for potentially creating shareholder value
relates to M&A, joint-ventures and other forms of financial consolidation.
We then present various empirical investigations into shareholder value
creation in European banking. In Chapter 5 we undertake a comparative
analysis of traditional (such as interest and intermediation margins, ROE;
ROA and net income) and non-traditional (such as residual income and
EVA) bank performance indicators in the light of creating shareholder
Conclusions 291
value within the European banking industry. In order to assess which per-
formance metric is the most compatible with shareholder value creation,
our analyses examine both the relative information content (which is
useful if one needs to select a single performance measure that best
explains shareholder value creation) and the incremental information
content (which aims to assess if a performance indicator adds information
to the data provided by other measures). Here, we focus on publicly listed
banks in four European countries, which comprise seventy-one publicly
listed banks (i.e. twenty French banks, thirteen German Banks, twenty-
eight Italian banks and ten UK banks) over 1996 to the end of 2002. Our
results suggest that EVA measures that are adjusted to account for the
specific features of banking operations have the greatest ability to explain
variation of market-adjusted returns (MAR). In contrast, standard Economic
Value Added measures used to examine shareholder value issues for non-
financial firms, do worse than a wide range of simple accounting perfor-
mance measures in explaining variations in MAR. We also find that the
treatment of banking capital and various accounting adjustments is central
when one calculates shareholder value created in banking.
The second part of Chapter 5 discusses the levels of shareholder value
created by European banks looking at a larger sample comprising both
listed and non-listed banks in France, Germany, Italy and UK. We found
that European banks generate, on average, substantial and widespread profits,
but these do not seem to have created shareholder value in many cases since
profits (measured from an economic viewpoint) are often lower than the
opportunity cost of equity invested capital. This situation is not common to all
countries analysed. UK banks appear to be the most profitable banks and to
create higher shareholder value than those in the other three countries.
German banks exhibit the lowest profits and created shareholder value.
After examining the levels of value creation in European banking we
present a detailed analysis of the determinants of this shareholder value.
Estimates on cost and profit efficiency, total factor productivity (TFP) and a
proxy for customer satisfaction are developed in Chapter 6 and these are
then linked to shareholder value creation in Chapter 7. Overall we find
that TFP changes best explain variations in shareholder value (measured by
market adjusted returns for listed banks and by the ratio of EVA (adjusted
for the specifics of banking) to invested capital for non-listed banks. Both
cost and profit efficiency are found to have a lower explanatory power in
explaining shareholder value creation in European banking although the
former is more influential than the latter. While our measures of depositor
and borrower satisfaction explain some variation in shareholder value, the
influence is slightly lower than for TFP or cost efficiency. The final part of
Chapter 7 presents a new measure of efficiency labelled ‘shareholder value
efficiency’. A bank is defined as shareholder value efficient if it is able to
produce at the maximum possible shareholder value given a particular level
of output. Using the alternative profit efficiency estimation framework

advanced by Berger and Mester (1997), we develop a new efficiency
measure which we call ‘alternative shareholder value efficiency’. According
to our estimates, European banks display shareholder value inefficiency
scores that suggest, on average, banks are 36 per cent shareholder value
inefficient. The situation changes across years, countries and types of
banks. Among the four banking systems analysed, we found that Italian,
German and UK banks have similar levels of shareholder value efficiency,
while French banks are found to generate only 58.39 per cent of their
potential shareholder value. These results are consistent with earlier
findings (discussed in Chapter 5) about the shareholder value generated by
European banks. We also discover that shareholder value efficiency better
explains the creation of shareholder value in European banking than cost
or profit efficiency. This superiority, however, is marginal for listed banks,
while it is substantial for non-listed banks. As such, shareholder value
efficiency seems to be particularly useful for non-listed banks for inves-
tigating the ability of these banks to create value for their shareholders/
stakeholders.
Limitations
As in all studies of this kind there are some limitations. First, the choice of
the appropriate measure of shareholder value for non-listed banks is partic-
ularly difficult. As noted in Chapter 5, while there is a general agreement as
to the concept of shareholder value, there is debate as to the best method
for assessing the value created by firms for their owners, as researchers and
practitioners grapple with different performance metrics. There are a grow-
ing number of studies investigating which performance measure is the
most compatible with shareholder value creation, but the evidence sur-
rounding this issue is mixed. In addition, few papers investigate this issue
focusing on the banking sector. For this reason, we examine the informa-
tion content of many performance measures. However, our sample is quite
small since we could only focus on listed banks in Europe. In addition, we
select some of the most well-known performance measures, but several
others perhaps could have been considered.
A second limitation of the text concerns our investigation of bank
efficiency. We estimate cost efficiency by using two frontier methodologies,
i.e. Stochastic Frontier Analysis and Data Envelopment Analysis. We
applied two established methodologies in the empirical literature to make
more robust our results and both sets of efficiency estimates are generally
consistent with each other. However, some other methodologies (such as
the Free Disposal Hull, Distribution-Free Approach and Thick Frontier
Approach) may have been applied. We also estimated alternative profit
efficiency measures developed by Berger and Mester (1997). This measure is
Conclusions 293
theoretically less accurate than the standard profit efficiency. However,

output prices are not accurately measured in banking (as stated in most of the
empirical studies) and, as such, we preferred to focus on alternative profit
efficiency. In all frontier estimations, we define bank inputs and outputs
according to the value-added approach (since bank inputs and outputs are
chosen in this approach depending on the contribution of bank items to the
creation of bank value-added), but several other definitions may have been
applied and this could lead us to different sets of estimates. Moreover, we use
the translog functional form for the cost and profit functions in our stochastic
estimation, but various alternative functional forms could have been chosen.
The translog form was adopted following some recent studies that have
identified various limitations associated with using the Fourier functional
form, especially when dealing with heterogenous data sets. However, the
Fourier flexible form has been used in a number of recent studies.
In addition, the estimates of cost and efficiency are derived from individ-
ual country frontiers. However, various recent studies analysing banks of
different countries tend to pool these banks in the same sample and, con-
sequently, estimate a ‘meta-frontier’. This probably would have allowed us
to develop a broader discussion enabling us to analyse the impact of a
single European banking market. However, we prefer to focus on individual
country frontiers since these estimates seem to be closer to the scope of this
book. The estimation of a meta-frontier would require us to assume that all
banks in the sample really compete in the same market and in our opinion
this seems currently to happen only for a limited number of banks. Since
our aim is to test the impact of efficiency on shareholder value, we prefer to
assume that banks compete in their own domestic banking market.
A third limitation of the analysis concerns the estimation of customer
satisfaction. The definition of an appropriate measure of customer satis-
faction is constrained by data availability. As such, we define a measure
based on deposits and loans growth relative to the industry average that,
however, may be biased by various other factors (such as market power
considerations).
Future research
Despite these limitations, however, this text presents (as far as we are
aware) a novel insight into shareholder value creation in banking. Future
research could focus on comparing different shareholder value measures in
banking across different countries – perhaps investigating differences in
shareholder value creation in the US, Japanese and other banking systems.
The methodology could also be applied to investigate the impact of regula-
tory changes, such as Basel 2 (to be implemented at the end of 2006) or
the adoption of International Accounting Standards (IAS39) (to be
adopted from January 2005) on bank performance.1 Our study links mea-
sures of efficiency, productivity and customer satisfaction to market-based

indicators of bank performance. More work is required in this area to
further inform the ongoing debate about measuring bank performance.
There is also substantial scope for the application of this type of analysis to
investigate shareholder value creation and performance of non-banking
firms.
Notes
Chapter 1 Why Study Shareholder Value Creation in European

Banking?
1. For example, see Biddle et al. (1997, 1999), Charreaux and Desbrières (2001),
Fernández (2002), Weaver and Weston (2003).
2. For example, see Stewart (1991), Schuster (2000), Copeland et al. (2000), Black
et al. (1998), Rappaport (1986, 1998), Weissenrieder (1997), Schroeck (2002) and
Belmont (2004)
Chapter 2 Economic Objectives of Banks

1. The conflict of interest among stakeholders has been analysed by some authors
in terms of contracting costs. Companies have been in fact considered as a
‘nexus of contracts’: contracts between shareholders and managers, contracts
between managers and employees, contracts between firms and supplier, con-
tracts between different business unit within the firms, etc. For further details,
see Jensen and Meckling (1976) and Glassman (1997).
2. The literature dealing with corporate governance problems in financial insti-
tutions is extensive. For further details, see the fundamental contributions of
Fama (1980), Fama and Jensen (1983) and Jensen and Meckling (1976). For more
recent research, see Prowse (1997), Carretta (1998), Carretta et al. (1998),
Forestieri (1998) and Schwizer (1998).
3. For further details on this debate, see Hall and Hitch (1939), Lester (1946, 1947),
Machlup (1946, 1947), Stigler (1947) and Gordon (1948).
4. For example, March and Simon (1958), Cyert and March (1963) and Jensen and
Meckling (1976).
5. See Dacrema and De Sury (1986).
6. The valuation ratio is the ratio of share price to the total asset per share.
7. For further details on the relationship between rationality and maximising
behaviour, see Becker (1962, 1976).
8. For further details on the limits of the perfect rationality assumption, see
Koopmans (1957), Ansoff (1965) and Salvati (1967).
9. See Dalborg (1999) and Rappaport (1998).
10. This is not a model in the sense of a theoretically defined approach to manage-
ment.
11. The word Prosumer, which is composed of two words (pro-ducer and con-sumer),
expresses a characteristic of many services: consumers do not simply receive
services: they play an active role in the production process as well. For example,
in a bank branch office, the presence of ‘too many’ customers (or an inexpert
consumer) makes the queue longer and decreases the satisfaction of the other
consumers.
12. For further details, see Dalborg (1999).
13. See Copeland et al. (2000), Hörter (2000), Venanzi and Fidanza (2005).
14. For further details, see Rappaport (1998).
295
296 Notes
15. For further details, see Stewart (1996), Al Ehrbar (1998), Regalli (2003).
16. These measures are analysed in detail in Chapter 3.
17. For further details, see Lehen and Makhija (1996), Uyemura et al. (1996) and
Black et al. (1998).
18. Rappaport (1998).
19. The concepts of ‘economic management’ (economicità), ‘efficiency’ (efficienza)
and ‘effectiveness’ (efficacia) are taken from Cavalieri (2000).
20. For further details, see Anderson et al. (1997), AISM (1990) and AISM-Galgano
(1994).
21. See Dalborg (1999), who reports the results of a questionnaire submitted to the
group of Institut International d’Etudes Bancaire (IIEB) banks in February 1999.
22. The idea is taken from the service value chain proposed by Heskett et al. (1994).
23. For further details, see Dalborg (1999).
24. Source of data: Tylecote and Tarhan (2000: 11) reporting estimates from consul-
tants Booz, Allen and Hamilton.
25. One may reach a different conclusion for web-based services. However, it is
worthwhile to note that web-based services are complementary (rather than sub-
stitutes) to traditional banking services. In constrast, web-based services appear
to be an alternative to traditional brokerage services.
26. Mike O’Neill, Chief Financial Officer of BankAmerica noted ‘we are also working
on pushing these incentives into lower levels of the organisation’. For further
details, see Stewart (1996).
27. Jim Hatch is senior vice president of First Union Corporation. For further details,
see Stewart (1996).
28. See Dalborg (1999), who reports the results of a questionnaire submitted to the
group of Institut International d’Etudes Bancaire (IIEB) banks in February 1999.
29. Remarks by Alan Greenspan at the annual convention of the American Bankers
Association, Honolulu, Hawaii, 5 October 1996.
Chapter 3 Shareholder Value: a Literature Review

1. See Hamilton (1777) and Marshall (1890).
2. For example, this is the case for measures such as ‘Economic Value Added’, i.e.
the surplus value created by a company in a given period measured as the firm’s
profit net of the cost of all capital, and ‘Market Value Added’, i.e. the current
market value of all capital elements minus the historical amount of capital
invested in the company. The word ‘added’ joins ‘value’ to emphasise that
company’s return (for EVA) or valuation (for MVA) implies a comparison with
an appropriate measure of the opportunity cost of capital.
3. To avoid any confusion, it is worthwhile to note that the definition of ‘share-
holder return’ proposed above differs slightly from the ‘Total Shareholder Return
(TSR)’ proposed in the FT’s European Company performance survey, which is
the percentage gain (or loss) received by a shareholder over the period, assuming
that all dividends distributed by the company are immediately reinvested in its
shares. For further details on TSR, see http://specials.ft.com/europerformance/.
4. This section summarises the Modigliani and Miller (M&M) dividend proposi-
tions. In a previous seminal work in 1958 (see Miller and Modigliani 1958),
M&M proposed three propositions dealing with corporate capital structure. The
first proposition (labelled the invariance proposition) states that the value of a
Notes 297
firm (i.e. the total value of its securities) is unrelated to its capital structure (i.e.
the debt/equity composition of the liabilities), but it depends only on the
earning power and risk of operating assets. The second proposition states that, if
proposition 1 is met, the expected return on leveraged shares will be a linear
increasing function of leverage. The third proposition states that the cost of
capital (i.e. the minimum expected rate of return required for undertaking a
project) is unrelated to the securities used to finance the project, but it depends
on the project and its risk.
5. Without aiming to be exhaustive, we report some of these measures and its pro-
ponents: Economic Value Added (Stern Stewart), Cash Flow Return on
Investment (Holt), Total Business Return (Boston Consulting Group), Economic
Profit (McKinsey), Shareholder Value Added (LEK/Alcar).
6. MVA is considered by many consultants (such as Al Ehrbar 1998, Uyemura et
al.,1996) as the most accurate measure of shareholder value.
7. For further details, see Lehen and Makhija (1997: 90).
8. For further details, see Santorum (2002).
9. Rappaport was the first to develop this metric. For further details on this
approach, see Rappaport (1986, 1998).
10. CFROI was originally supported by Boston Consulting Group (BCG) and HOLT
Value Associates, a Chicago firm that advises asset managers on questions of val-
uation (see Economist 1997).
11. The examination of statistical models developed in risk management would
require a detailed assessment, which is outside of the scope of this text. For a
detailed review of these methods, see Credit Suisse Financial Products (1997),
Crosbie (1999), Saita (2000) and De Laurentis (2001).
12. Research and Development (R&D) expenses, amortisation of goodwill and
inflation are some examples of economic distortions derived by the application
of GAAP.
13. See, for example, Al Ehrbar (1998), McTaggart and Gillis (1998) and (in banking)
Uyemura et al. (1996).
14. Stock market prices are expectational because investors value stock depending
on future expected company’s profits.
15. For unlisted companies, MVA can be estimated only using a proxy of the market
value of capital.
16. This investigation is based on the Stern Stewart Performance 1000 database,
which contains EVA and MVA estimates for the top 1000 US value-creator firms.
Results presented above are taken from Al Ehrbar (1998: 78).
17. Peterson and Peterson (1996) and Biddle et al. (1997).
18. ‘Goodwill refers to the capacity of the company to generate abnormal returns in
the future, because it represents the expected earnings power of a collection of
assets, combined with a given set of managerial skills’ (Barker 2001: 135).
19. For a complete review of these studies, see Holthausen and Watts (2001) and
Kothari (2001).
20. For example, Uyemura et al. (1996) propose a rank for the first 100 US banks (in
the years 1995, 1994 and 1990) in terms of EVA and MVA.
21. The latest publicly available Stern Stewart MVA Performance rankings are the
2001 and 2000 ranks (concerning 2000 and 1999, respectively). Stern & Stewart
MVA performance ranking changed the currency between 1999 (GBP) and 2000
(USD).
22. It is important to note that the cost of capital invested for all banks in the UK
sample is the same (i.e. 11.2 per cent in 1999 and 10.0 per cent in 2000). It
298 Notes
seems therefore that Stern & Stewart applied a common estimated cost of capital
invested to all British banks, rather than calculating the cost of capital for each
individual bank (as Stern & Stewart does in the ranking for Italian banks).
23. For further details, see Kothari (2001) and Lee (2001).
24. The coefficient r-square allows one to evaluate the proportion of the variability of
the dependent variable that is explained by the selected explanatory variables. This
coefficient ranges between 0 and 1 and the closer the model to 1, the better the
model. The adjusted coefficient of determination (or adjusted R2) is expressed as:
(n – 1)R2 – p
adj R2 =
n–p–1
where n is the number of observations and p the number of explanatory variables.
25. Raw rate of return expresses the increase of equity market value in terms of rate
of return. This can be expressed as:
Pt
Raw Return = – 1.
Pt–1
26. Several studies (such as Easton, 1998, Easton and Sommers 2003) empirically
demonstrated the distortion generated in value relevance studies by ‘scale
effects’.
27. E.g. in Hirschey (1985).
28. E.g. in Rees (1997), Hand and Landsman (1998).
29. E.g. in Lo and Lys (2000) and Brown et al. (1999).
30. E.g. in Easton (1998).
31. E.g. in Easton and Sommers (2003).
32. Easton and Sommers (2003) proposed a model on US data where: (1) market cap-
italisation (i.e total equity market value) is the dependent variable; (2) closing
book value of common equity and net income are the dependent variables.
Chapter 4 How Banks Create Shareholder Value

1. Beta is calculated as the co-variance (Cov) between the market price of the share
(i) and the market portfolio (m) standardised for the Variance (Var) of the
market portfolio.
2. For further details, see Di Antonio (2002). Olson (1977) was the first to postulate
a trade-off between price and service quality: for example, a low price contributes
positively to the product selection, but it affects negatively the quality of services
expectations. Tse (2001) investigates the relationship between ‘quality of the
service’ and ‘price’ in the service sector.
3. The average length of the relationship between a bank and its customers is posi-
tively related to customer retention rate.
4. Data reported in Hesselink and Van der Wiele (2003).
5. See Fornel and Wernerfelt (1987), Peters (1988) and Reichheld and Sasser (1990).
6. This example follows the procedure described in Munari (2000).
7. For an interesting statistical test of the relationship between TQM and customer
satisfaction, see Forza and Filippini (1998).
8. Allocative Efficiency is the terminology currently adopted by the most recent lit-
erature, whilst Farrell (1957) originally labelled this measure as ‘Price Efficiency’.
9. For further details, see Schroeck (2002) and Cohen (2004a).
10. See also Cohen (2004b) and Belmont (2004).
11. Source: Davis and Lee (1997).
Notes 299
12. Source: Resti (1999).

13. For example see Carretta (2002).
14. ABI (2004).
15. BIS (2004).
16. The Glass-Steagall Act, passed by the US Congress in 1933, prohibited commer-
cial banks from collaborating with full-service brokerage firms or participating in
investment banking activities.
17. For further details, see Group of Ten (2001: 65).
18. See, for example, Berger and Hannan (1997, 1998).
19. See also Berger et al. (2004) for a discussion of this point.
20. See, for example, Veld and Veld-Merkoulova (2004).
21. Such as Peristiani (1997), Akhavein et al. (1997), Resti (1998) and Focarelli and
Pozzolo (2001).
Chapter 5 Bank Performance Measures and Shareholder Value

1. Raw rate of return expresses the increase of equity market value in term of rate
of return. This can be expressed as:
Pt
Raw Return = – 1.
Pt–1
2. A period of four months is selected since financial reports are usually published
within four months after the firm’s fiscal year.
3. See, for example, Stewart (1991), Uyemura et al. (1996), Rappaport (1998),
Al Ehrbar (1998).
4. Al Ehrbar (1998) recognises that there may be several EVA values according to
the number of accounting adjustments. As a result, it is possible to identify a
spectrum of EVA values: the ‘basic EVA’ is obtained using unadjusted GAAP
operating profit and GAAP balance sheet capital; the ‘disclosed EVA’ is obtained
making some standard adjustments to publicly available accounting data. This
measure, which improves the basic EVA by solving the main GAAP problems, is
usually adopted by external analysts; the ‘true EVA’ can be calculated using ‘all’
internal data that reflect the true economic condition of the company; the ‘tai-
lored EVA’ is obtained using specific internal information (e.g. organisation
structure, business mix, strategies, accounting mix) to adjust accounting figures.
Internal analysts use a part of all internal data that balances the trade-off
between simplicity and precision.
5. For further details on standard adjustments in EVA calculation, see Santorum
(2002).
6. Stern Stewart publishes annually a performance report for the 1000 largest com-
panies (see Stewart, 1991).
7. Since data availability limitation does not allow us to evaluate the present value
of expected lease commitments over time, the present value of expected future
lease commitments capitalised (in every year between 1995 and 1999) is
assumed to be equal to the overall amount of operating leases expenses over
the period 1995–9. The amount amortised every year (i.e. from 1995 to 1999) is
near to the overall amount of R&D expenses over the period 1995–9.
8. Otherwise, it would be necessary to distinguish between borrowed funds
assigned to finance banking operations and those representing a productive
input. Since our data set does not enable us to make this differentiation, we
prefer to focus only on equity capital.
300 Notes
9. This point is also supported by Uyemura et al. (1996: 102) and Di Antonio
(2002: 103).
10. See, for example, Damodaran (1999d).
11. Fama and French (2002) propose a different model using dividend and earnings
growth rates to measure the expected rate of capital gain and estimate the equity
risk premium: this model is the earning growth model.
12. Regarding the US risk premium, the estimation provided by Damodaran (1999c)
has been adopted: 6.10 per cent.
13. Since one would expect the country equity risk premium to be larger than the
country default risk spread, the country equity spread is obtained by adjusting
the country bond spreads as follows:
σ
Country Equity Risk Premium = Country Bond Spread × equity
σ bond
Where: (1) the volatility of the equity market (σ equity) has been estimated focus-
ing on CAC 40, DAX 30, MIBTEL storico, FTSE 100: 2, the volatility of the bond
market (σ bond) has been estimated focusing on the French-, Italian-, German-,
UK- J. P. Morgan Government Bond return indices.
14. Namely: Beta = Regression Beta (0.67) + 1.00 (0.33).
15. These models with one-year lag are equivalent to the levels of changes
specification proposed by Easton and Harris (1991). Since the equity markets are
assumed efficient (at least in the weak form), one year lag seems to be reason-
able: it is unlikely that two-year-old accounting information has explanatory
power of variation in market values.
16. Hayn (1995), Burgstanhler and Dichev (1997) and Collins et al. (1997) found
that firms that make losses have smaller earning response coefficients than do
profitable firms.
17. This model is, of course, inappropriate for those performance measures having
only positive values.
18. A period of four months is selected since financial reports are usually published
within four months after the firm’s fiscal year.
19. The values for DW tests are very close and generally slightly higher than 2, show-
ing a slight negative autocorrelation (a common problem in time-series data).
20. On average, the estimated opportunity cost of capital is 6.87 per cent for France,
6.40 per cent for Germany, 12.7 per cent for Italy and 14.31 per cent for the UK.
Italian and UK banks have higher levels since we found that these countries
have higher risk-free rates, country bond spreads and average betas over the
period analysed.
Chapter 6 Measuring Shareholder Value Drivers in Banking

1. The choice of the use of translog is motivated by two reasons: first, Altunbas and
Chakravraty (2001) identified some problems associated with using the Fourier
functional form, especially when dealing with heterogeneous data sets.
Secondly, Berger and Mester (1997) observe that the translog functional form and
Fourier flexible form are substantially equivalent from an economic viewpoint and
both rank individual bank efficiency in almost the same order.
2. Since linear homogeneity is assumed, TC, w1 and w2 are standardised by the
price of capital w3.
3. For further details, see Ali and Seiford (1993).
Notes 301
4. Such as, for example, the two-stages DEA suggested in Ali and Seiford (1993).
5. Although steps 1 and 2 are carried out in the standard two-stage method (pro-
posed by Ali and Seiford 1993), the projected points obtained in this second step
are not used in the method. The second step aims simply to identify the
‘Koopmans’ efficient set’ and to ‘have slack set’.
6. Potential slacks exist in that input when the contraction is achieved.
7. For a proof of the Units Invariance Theorem, see Cooper et al. (2000).
8. It is also possible to consider revenue maximisation and allocative inefficiency
in output mix selection in a similar manner. For further details, see Lovell (1993:
33).
9. For the DEA model used to measure allocative efficiency, see Cooper et al.
(2000).
10. This selection of inputs and outputs follows the studies by Sathye (2001) and
Dietsch and Lozano-Vives (2000), Aly et al. (1990) and Hancock (1986), where a
methodology based on user costs is used to determine the outputs and inputs of
a banking firm.
11. We use unconsolidated balance-sheet information.
12. In Italy and France, the number of savings banks is quite small. In order to have
enough degrees of freedom in estimating the SFA models, we prefer to pool
savings and cooperative banks in a unique sample and estimate a common fron-
tier for both categories of banks. In the UK, there are only three savings banks
that have been included in the sample of commercial banks.
13. For both DEA and SFA, we estimate an annual cross-section frontier for eight
years (1995–2002), for two bank categories (firstly, commercial banks and, sec-
ondly, cooperative and savings banks) and for three countries (France, Italy and
Germany). For the UK, we estimate an annual cross-section efficient frontier for
eight years (1995–2002) for only commercial banks (although the sample com-
prises three savings banks).
14. Technical efficiency is estimated by assuming Variable Returns to Scale, allocative
efficiency is estimated by assuming Variable Return of Scale, cost efficiency is
obtained by multiplying technical efficiency, allocative efficiency and scale
efficiency estimates obtained assuming variable returns to scale (i.e. equivalent to
estimates of cost efficiency under constant return of scale).
15. It is worthwhile remembering that mean cost efficiency measures express the
mean distance of inefficient banks from an efficient frontier, i.e. estimated using
a cross-sectional sample and, as such, it changes every year. As such, all compar-
ison among mean efficiency levels obtained for different countries or by dif-
ferent kinds of bank (for example) focus on the mean distance of inefficient
banks from their own efficient frontier. As such, the expression ‘more/less cost
efficient’ means that the mean distance of inefficient banks from their own
efficient frontier is smaller/larger than the mean distance of inefficient banks of
another category/country. A lower/superior mean cost efficiency does not supply
any evidence of lower/superior Total Factor Productivity.
16. This mean value mostly accounts for technical efficiency levels.
17. It is worthwhile noting that most of the studies summarised by Berger and
Humphrey (1997) focus on the US banking system, on different time periods
and use various bank input and output definition than used in our analysis.
18. This selection of inputs and outputs follows the studies by Sathye (2001) and
Dietsch and Lozano-Vives (2000), Aly et al. (1990) and Hancock (1986), where
they develop a methodology based on user costs to determine the outputs and
inputs of a banking firm.
302 Notes
19. Since we assume linear homogeneity, π , w1 and w2 are standardised by the price
of capital w3.
20. The UK sample also comprises three savings banks. Firstly, we choose to include
these three banks in the UK sample since the number of banks in the UK sample
is smaller than for the other three countries. Secondly, we decide to label our UK
sample as ‘commercial banks’ since three savings banks represent less than 5 per
cent of the sample.
21. In Italy and France, the number of savings banks is quite small. In order to have
enough degrees of freedom for estimating the SFA models, we prefer to pool
savings and cooperative banks and estimate a common frontier for both cate-
gories of banks. In the UK, there are only three savings banks that have been
included in the sample for commercial banks.
22. We estimated 16 profit efficient frontiers (i.e. 8 years × 2 bank categories) for
France, Germany and Italy. For the UK, we estimated 8 frontiers since we have
only one type. As such, we estimated 56 frontiers (i.e. 16 × 3 + 8).
23. When a firm uses more than one input, it is necessary to adopt a method for
aggregating these inputs into a single index of inputs. The same problem occurs
for multiple outputs.
24. A recent study of Casu et al. (2004: 2538) measured productivity change in
European banking between 1994 and 2002 using both parametric and non-para-
metric methodologies (namely, DEA) and found that ‘overall, we find that the
competing methodologies do not yield markedly different results in terms of
identifying the main components of productivity growth’.
25. For further details, see Brown (1996).
26. Casu et al. (2004) define bank inputs and outputs using the intermediation
approach (while we use the value-added approach) and their sample comprises
only large banks (over Euro 450 million) from France, Germany, Italy, Spain and
the UK (while we use a larger sample comprising commercial banks, cooperative
banks and savings banks and estimate different Malmquist indices according to
bank specialisation).
27. We prefer to use the geometric mean as an average measure of the change that
occurred in the period analysed.
28. Mean values of loans (and of deposits) net growth rates are not weighted to the
total amount of loans (and of deposits) at time t–1 (i.e. deposits at t–1 and loans
at t–1).
Chapter 7 Determinants of Shareholder Value in European Banking

and Shareholder Efficiency
1. In both samples, we use dummy variables to capture the impact due to: (1) bank
specialisation (namely, commercial, cooperative and savings banks); (2) time
periods; and (3) different domestic economic conditions.
2. R-squared change is the change in the R-squared statistic that is produced by adding
or deleting an independent variable. If the R2 change associated with a variable is
large, that means that the variable is a good predictor of the dependent variable.
3. We do not report the estimated coefficients on all the dummy variables.
4. Following the Berger and Mester (1997) findings and considering our research
aims, the translog functional form is preferred to the Fourier-flexible since it is sub-
stantially equivalent from an economic viewpoint and both rank individual bank
efficiency in a similar order.
Notes 303
5. This selection of inputs and outputs follows Sathye (2001), Dietsch and Lozano-
Vives (2000), Aly et al. (1990) and Hancock (1986).
Chapter 8 Conclusions
1. See Basel Committee (2004) and International Accounting Standards Board
(2004) for technical details relating to Basel 2 and the implementation of IAS 39
on ‘fair value’ accounting, respectively.
References
Aaker, D. A. and R. Jacobson (1994) ‘The Strategic Role of Product Quality’, Journal of
Marketing 51: 31–44.
Abarbanell, J. and V. Bernard (1994) ‘Is the U.S. Stock Market Myopic?’ Working paper,
University of Michigan.
ABI (1997) Osservatorio ABI sulla customer satisfaction, Rome: Bancaria Editrice.
ABI (2004) ‘Rapporto ABI sui mercati finanziari e creditizi (gennaio 2004)’, Bancaria,
no. 2.
Acheampong, Y. J. and M. E. Wetzstein (2001) ‘A Comparative Analysis of Value-added
and Traditional Measures of Performance: an Efficiency Score Approach’, working
paper FS 01-04, Department of Agricultural and Applied Economics, College of
Agricultural and Environmental Sciences, University of Georgia. Available at
http://ssrn.com/abstract=259200.
Affinito, M., R. De Bonis and F. Farabullini (2003) ‘Concorrenza e convergenza tra i
sistemi bancari dell’area dell’euro’, Enti per gli studi Monetari, Bancari e Finanziari Luigi
Einaudi, Quaderni di Ricerche, no. 40. Available at http://www.enteluigieinaudi.it/pdf/
Pubblicazioni/Quaderni/Q_40.pdf.
Afriat, S. N. (1972) ‘Efficiency Estimation of Production Functions’, International
Economic Review 13: 568–98.
Ahmed, A. S., R. M. Morton and T. F. Schaefer (2000) ‘Accounting Conservatism and
the Valuation of Accounting Numbers: Evidence on the Felthman-Ohlson (1996)
Model’, Journal of Accounting, Auditing and Finance 15, 3: 271–92.
Aigner, D. J. and S. F. Chu (1968) ‘On Estimating the Industry Production Functions’,
American Economic Review 58: 826–39.
Aigner, D. J., C. A. K. Lovell and P. Schmidt (1977) ‘Formulation and Estimation of
Stochastic Frontier Production Function Models’, Journal of Econometrics 6: 21–37.
Airoldi, G. and G. Forestieri (eds) (1998) Corporate Governance, Milan: ETAS Libri.
AISM (1990) La qualità in banca, vols I–III, Milan: Ricerca AISM.
AISM-Galgano (1994) La qualità in banca: indagine presso il sistema bancario italiano,
Milan: Ricerca AISM.
Akbar, S. and A. W. Stark (2003) ‘Discussion of Scale and the Scale Effect in
Market-based Accounting Research’, Journal of Business, Finance and Accounting 30, 1:
57–72.
Akhavein, J. D., A. N. Berger and D. B. Humphrey (1997) ‘The Effect of Megamergers
on Efficiency and Prices: Evidence from a Bank Profit Function, Review of Industrial
Organisation 12: 95–139.
Al Ehrbar (1998) EVA – The Real Key to Creating Wealth, London: John Wiley & Sons.
Al Ehrbar (1999) ‘Using EVA to Measure Performance and Assess Strategy’, Strategy and
Leadership, May–June: 20–4.
Alarm, I. M. S. (2001) ‘A Non-Parametric Approach for Assessing Productivity
Dynamics of Large Banks’, Journal of Money, Credit and Banking 33: 121–39.
Allen, L. and A. Rai (1997) ‘Operational Efficiency in Banking: an International
Comparison’, Journal of Banking and Finance 20: 655–85.
Ali, A. I., C. S. Lerme and L. M. Seiford (1995) ‘The Components of Efficiency
Evaluation in Data Envelopment Analysis’, European Journal of Operational Research
80: 462–73.
304
References 305
Ali, A. I. and L. M. Seiford (1993) ‘The Mathematical Programming Approach

to Efficiency Analysis’, in H. O. Fried, C. A. K. Lovell and S. S. Schmidt (eds),
The Measurement of Productive Efficiency, New York: Oxford University Press,
pp. 120–59.
Allen, F. and A. M. Santomero (1997) ‘The Theory of Financial Intermediation’, Journal
of Banking and Finance 21: 1461–85.
Altunbas, Y., S. Carbo, E. P. M. Gardner and P. Molyneux (2004) ‘Examining the
Relationship between Capital, Risk and Efficiency in European Banking’, paper pre-
sented at the annual conference of the European Association of University Teachers
in Banking and Finance, Krakow University, September.
Altunbas, Y. and S. P. Chakvararty (1998) ‘Efficiency Measures and the Banking
Structure in Europe’, research paper no. 1, University of Wales Bangor, Institute of
European Finance.
Altunbas, Y. and S. P. Chakravarty (2001) ‘Frontier Cost Functions and Bank
Efficiency’, Economics Letters 72: 233–40.
Altunbas, Y., L. Evans and P. Molyneux (2000a) ‘Bank Ownership and Efficiency’,
Journal of Money, Credit and Banking 33: 926–54.
Altunbas, Y., E. P. M. Gardener, P. Molyneux and B. Moore (2001) ‘Efficiency in
European Banking’, European Economic Review 45, 10: 1931–55.
Altunbas, Y., J. Goddard, and P. Molyneux (1999) ‘Technical Change in Banking’,
Economics Letters 64: 215–221.
Altunmas, Y., M. H. Liu, P. Molyneux and R. Seth (2000b) ‘Efficiency and Risk in
Japanese Banking’, Journal of Banking and Finance 24: 1605–28.
Altunbas, Y. and P. Molyneux (1996) ‘Economies of Scale and Scope in European
Banking’, Applied Financial Economics 6: 367–75.
Altunbas, Y., P. Molyneux and E. P. M. Gardener (1996) Efficiency in Banking, London:
John Wiley & Sons.
Aly, H. Y., R. Grabowski, C. Pasurka and N. Rangan (1990) ‘Technical, Scale and
Allocative Efficiencies in US Banking: an Empirical Investigation’, Review of Economics
and Statistics 72: 211–18.
Amel, D., C. Barnes, F. Panetta and C. Salleo (2003) ‘Consolidation and Efficiency
in the Financial Sector: a Review of the International Evidence’, CEIS Tor Vergata,
research paper series, vol. 7, no. 20.
Anderson, E. W. (1995) ‘Word of Mouth as a Consequence of Customer Satisfaction’,
working paper, National Quality Research Center, Business School, University of
Michigan.
Anderson, E. W., C. Fornell, D. R. Lehmann (1994) ‘Customer Satisfaction, Market
Share, and Profitability: Findings from Sweden’, Journal of Marketing 58, 3: 53–66.
Anderson, E. W., C. Fornell and R. T. Rust (1997) ‘Customer Satisfaction, Productivity,
and Profitability: Differences between Goods and Services’, Marketing Science 16:
129–45.
Anderson, E. W. and M. W. Sullivan (1993) ‘The Antecedents and Consequences of
Customer Satisfaction for Firms’, Marketing Science 12: 125–43.
Ando, A. and A. J. Auerbach (1988) ‘The Cost of Capital in the U.S. and Japan:
a Comparison’, Journal of Japanese and International Economies 2: 134–58.
Ando, A. and A. J. Auerbach (1990) ‘The Cost of Capital in Japan: Recent Evidence and
Further Results’, Journal of Japanese and International Economics 4: 323–50.
Ansoff, H. I. (1965) Corporate Strategy, New York: McGraw-Hill.
Arthur Andersen & Co (1994) Customer Satisfaction Strategies, and Tactics, Chicago:
Arthur Anderson & Co.
Aubrey, II C. A. (1991) La qualità globale nei servizi finanziari, Milan: Editoriale Itaca.
306 References
Babkus, E. and G. W. Boller (1992) ‘An Empirical Assessment of the SERVQUAL Scale,
Journal of Business Research 24: 253–68.
Bacidore, J. M., J. A. Boquist, T. T. Milbourn and A. V. Thakor (1997) ‘The Search for
the Best Financial Performance Measure’, Financial Analysts Journal, May–June,
11–20.
Bagella, M., L. Cavallo and S. P. S. Rossi (1998) ‘Efficiency Determinants in the
European Banking Systems: a Stochastic Frontier Approach’, CEIS working paper,
no. 68, University of Rome Tor Vergata.
Bajetta, L. (1994) ‘La customer satisfaction – possono essere soddisfatte le banche della
soddisfazione della clientela?’ Bancaria, 10: 74–83.
Baltagi, B. H. and J. M. Griffin (1988) ‘A General Index of Technical Change’, Journal of
Political Economy 96: 20–41.
Banerjee, A. and E. Cooperman (2000) ‘Returns to Targets and Acquirers: Evidence
for Bank Mergers in the 1990s’, working paper, University of Colorado,
Denver.
Banker, R. D. (1996) ‘Hypothesis Tests Using Data Envelopment Analysis’, Journal of
Productivity Analysis 7: 139–60.
Banker, R. D., A. Charnes and W. W. Cooper (1984) ‘Some Models for Estimating
Technical and Scale Inefficiencies in Data Envelopment Analysis’, Management
Science 30: 1078–92.
Banker, R. D., A. Charnes, W. W. Cooper, J. Swarts and D. A. Thomas (1989)
‘An Introduction to Data Envelopment Analysis with Some of its Models and their
Uses’, in J. L. Chan and J. M. Patton (eds), Research in Governmental and Non-profit
Accounting, Greenwich, CN: JAI Press, pp. 125–63.
Banker, R., G. Potter and D. Srinivasan (1998) ‘An Empirical Investigation of
an Incentive Plan Based on Non-financial Performance Measures’, working
paper University of Texas at Dallas, Cornell University and University of
Pittsburgh.
Barker, R. (2001) Determining Value, Valuation Models and Financial Statements, Financial
Times-Prentice Hall, Pearson Higher Education.
Bar-Yosef, S., J. L. Callen and J. Livnat (1996) ‘Modelling Dividends, Earnings and Book
Value Equity: an Empirical Investigation of the Ohlson Valuation Dynamics’, Review
of Accounting Studies 1: 207–24.
Basel Committee (2004) Basel 2: International Convergence of Capital Measurement and
Capital Standards: a Revised Framework, Basel Committee Publications, June, Bank For
International Settlements, Basel.
Battese, G. E. and T. J. Coelli (1988) ‘Prediction of Firm-level Technical Efficiency with
a Generalised Frontier Production Function and Panel Data’, Journal of Econometrics
38: 387–99.
Battese G. E. and T. J. Coelli (1992) ‘Frontier Production Functions, Technical
Efficiency and Panel Data: With Application to Paddy Farmers in India’, Journal of
Battese, G. E. and T. J. Coelli (1995) ‘A Model for Technical Inefficiency Effects
in a Stochastic Frontier Production Function for Panel Data’, Empirical Economics
20: 325–32.
Battese, G. E. and G. S. Corra (1977) ‘Estimation of a Production Frontier Model: With
Application to the Pastoral Zone of Eastern Australia’, Australian Journal of Agricultural
Economics 38: 387–99.
Battese, G. E., A. Heshmati and L. Hjalmarsson (2000) ‘Efficiency of Labour Use in the
Swedish Banking Industry: a Stochastic Frontier Approach’, Empirical Economics
25: 623–40.
References 307
Bauer, H. H., M. Hammerschmidt and M. Braehler (2001) ‘The Customer Lifetime

Value Concept and its Contribution to Corporate Valuation’, Yearbook of Marketing
and Consumer Research 1: 47–67.
Bauer, P. W. (1990) ‘Recent Development in the Econometric Estimation of Frontier’,
Journal of Econometrics 46: 39–56.
Bauer, P. W., A. N. Berger, G. D. Ferrier and D. B. Humphrey (1997) ‘Consistency
Conditions for Regulatory Analysis of Financial Institutions: a Comparison of
Frontier Efficiency Methods’, Federal Reserve Board, Finance and Economics
Discussion Series, no. 1997–50.
Bauer, P. W. and Hancock D. (1993) ‘The Efficiency of the Federal Reserve in Providing
Check Processing Services’, Journal of Banking and Finance 17: 287–311.
Baumol, W. J. (1959) ‘On the Theory of the Expansion of the Firm’, American Economic
Review 52: 1078–87.
Bebko, C. P. (2000) ‘Service Intangibility and its Impact on Consumer Expectations of
Service Quality’, Journal of Service Marketing 14, 1: 9–26.
Beccalli, E. (2004) ‘Cross-country Comparisons of Efficiency: Evidence from the UK
and Italian Investment Firms’, Journal of Banking and Finance 28: 1363–83.
Beccalli, E., B. Casu and C. Girardone (2003) ‘Efficiency and Stock Performance in
European Banking’, Working Papers Series, EFMA Helsinki Meetings.
Becher, D. A. (1999) ‘The Valuation Effects of Bank Mergers’, Journal of Corporate
Finance 6: 189–214.
Becker, G. S. (1962) ‘Irrational Behaviour and Economic Theory’, Journal of Political
Economy 70: 1–13.
Becker, G. S. (1976) The Economic Approach to Human Behaviour, Chicago: University of
Chicago Press.
Beitel, P. and D. Schiereck (2001) ‘Value Creation at the Ongoing Consolidation of the
European Banking Markets’, IMA working paper, no. 05/01.
Beitel, P., D. Schiereck and M. Wahrenburg (2004) ‘Explaining M&A Success in
European Banks’, European Financial Management 10: 109–39.
Belmont, (2004) Value Added Risk Management in Financial Institutions: Leveraging Basel
II and Risk Adjusted Performance Measurement, London: John Wiley & Sons.
Benston, G. J, G. A. Hanweck and D. B. Humphrey (1982) ‘Scale Economies in
Banking: a Restructuring and Reassessment’, Journal of Money, Credit and Banking 14:
435–56.
Berenson, M. L. and D. M. Levine (1992) Basic Statistics: Concept and Application, New
Jersey: Prentice-Hall.
Berger, A. N. (1993) ‘Distribution-free Estimates of Efficiency in the US Banking System
and Tests of the Standard Distributional Assumptions’, Journal of Productivity Analysis
4: 261–83.
Berger, A. N. (1994) ‘The Profit Structure Relationship in Banking: Tests of Market
Power and Efficient Structure Hypotheses’, Journal of Money, Credit and Banking 17:
317–47.
Berger, A. N. (1995) ‘The Profit Structure Relationship in Banking: Tests of Market
Power and Efficient Structure Hypotheses’, Journal of Money, Credit and Banking 27:
404–31.
Berger, A. N. (2000) ‘The Integration of the Financial Services Industry: Where are the
Efficiencies?’ Federal Reserve Board, Finance and Economics Discussion Series,
no. 36 (June).
Berger, A. N., S. M. Davies and M. J. Flannery (1998) ‘Comparing Market and
Supervisory Assessment of Bank Performance: Who Knows What When’, Federal
Reserve Board, Finance and Economics Discussion Series, no. 32.
308 References
Berger, A. N., Asli Demirgiif-Kunt, R. Levine and J. G. Haubrich (2004) ‘Introduction:

Bank Concentration and Competition: an Evolution in the Making’, Journal of
Money, Credit and Banking 36: 433–53.
Berger, A. N., R. S. Demsetz, and P. E. Strahan (1999) ‘The Consolidation of the
Financial Services Industry: Causes, Consequences, and Implications for the Future’,
Journal of Banking and Finance 23: 135–94.
Berger, A. N. and R. De Young (1996) ‘Problem Loans and Cost Efficiency in
Commercial Banks’, Board of Governors of the Federal Reserve System, working
paper (August).
Berger, A. N. and R. De Young (2001) ‘The Effect of Geographic Expansion on Bank
Efficiency’, Journal of Financial Service Research 19, 2/3: 163–84.
Berger, A. N., D. Hancock, and D. B. Humphrey (1993) ‘Bank Efficiency Derived from
the Profit Function’, Journal of Banking and Finance 17: 317–47.
Berger, A. N. and T. H. Hannan (1989) ‘The Price Concentration Relationship in
Banking’, Review of Economics and Statistics 71: 291–9.
Berger, A. N. and T. H. Hannan (1997) ‘Using Efficiency Measures to Distinguish
among Alternative Explanations of the Structure Performance Relationship in
Banking’, Managerial Finance 23, 1: 6–31.
Berger, A. N. and T. H. Hannan (1998) ‘The Efficiency Cost of Market Power in the
Banking Industry: a Test of the “Quit Life” and Related Hypotheses’, Review of
Economics and Statistics 80, 3: 454–65.
Berger, A. N. and D. B. Humphrey (1992) ‘Measurement and Efficiency Issues in
Commercial Banking’, in Zvi Griliches (ed.), Output Measurement in the Service Sectors,
Chicago: University of Chicago Press, pp. 245–79.
Berger, A. N. and D. B. Humphrey (1997) ‘Efficiency of Financial Institutions:
International Survey and Direction for Future Research’, Federal Reserve Board,
Finance and Economics Discussion Paper, no. 11.
Berger, A. N., W. C. Hunter and S. G. Timme (1993) ‘The Efficiency of Financial
Institutions: a Review and Preview of Research Past, Present and Future’, Journal of
Banking and Finance 17: 221–49.
Berger, A. N., J. H. Leusener and J. J. Mingo (1997) ‘The Efficiency of Bank Branches’,
Journal of Monetary Economics 40: 141–62. Also published as a working paper of the
Wharton Financial Institutions Center, no. 92–47.
Berger, A. N. and L. J. Mester (1997) ‘Inside the Black Box: What Explains Differences
in the Efficiency of Financial Institutions’, Journal of Banking and Finance 21, 7:
895–947.
Berger, A. N. and L. J. Mester (1999) ‘What Explain Dramatic Changes in Cost and
Profit Performance of the US Banking Industry?’, Wharton Financial Institutions
Center, working paper, 99–10.
Berger, A. N. and L. J. Mester (2001) ‘Explaining the Dramatic Changes in Performance
of US Technological Change, Deregulation and Dynamic Changes in Competition’,
Wharton Financial Institutions Center, working paper, no. 01–22.
Berle, A. A. and J. C. Means (1932) The Modern Corporation and the Private Property,
New York: Macmillan.
Bernard, V. L. (1995) ‘The Felthman-Ohlson Framework: Implication for Empiricists’,
Contemporary Accounting Research 11: 733–47.
Berry, L. L. and A. Parasuraman (1991) Marketing Services: Competing through Quality,
New York: Free Press.
Biddle, G. C., R. M. Bowen and J. S. Wallace (1997) ‘Does EVA Beat Earnings? Evidence
on Association With Stock Returns and Firm Values’, Journal of Accounting and
Economics 24, 3: 301–36.
References 309
Biddle, G. C., R. M. Bowen and J. S. Wallace (1999) ‘Evidence on EVA’, Journal of

Applied Corporate Finance 12, 2: 67–79.
Biddle, G. C., P. Chen and G. Zhang (2001) ‘When Capital Follows Profitability:
Non-linear Residual Income Dynamics’, Review of Accounting Studies 6:
229–65.
Biddle G. C. and Seow G. (1991) ‘The Estimation and Determinants of Association
between Returns and Earnings: Evidence from Cross-Industry Comparisons’, Journal
of Accounting, Auditing and Finance 6: 183–232.
Biddle, G. C., G. Seow and A. Siegel (1995) ‘Relative versus Incremental Information
Content’, Contemporary Accounting Research 12: 1–23.
BIS–Bank for International Settlements (2004) 74th Annual Report, 1 April 2003–31
March 2004.
Black, A., P. Wright and J. E. Bachman (1998) In Search of Shareholder Value, London:
Financial Times/Pitman Publishing.
Blake, F. (2000) Financial Market Analysis, 2nd edn, London: John Wiley & Sons.
Bloemer, J., K. D. Ruyter and P. Peeters (1998) ‘Investigating Drivers of Bank Loyalty:
the Complex Relationship between Image, Service Quality and Satisfaction’,
International Journal of Bank Marketing 17, 7: 276–86.
Boles, J. N. (1966) ‘Efficiency Squared: Efficiency Computation and Efficiency Indexes’,
Proceedings of the Thirty-Ninth Meeting of the Western Farm Economic Association,
pp. 137–42.
Bolton, R. and J. H. Drew (1991a) ‘A Multi-stage Model of Customers’ Assessments of
Service Quality and Value’, Journal of Consumer Research 17: 375–84.
Bolton, R. N. and J. H. Drew (1991b) ‘A Longitudinal Analysis of the Impact of Services
Changes on Customer Attitudes’, Journal of Marketing 55 (January): 1–9.
Bos, J. W. B. and H. Schmiedel (2003) ‘Comparing Efficiency in European Banking:
a Meta-Frontier Approach’, De Nederlandsche Bank, research paper, no. 57.
Botosan, C. A. and M. A. Plumlee (2001) ‘Estimating Expected Cost of Equity Capital:
a Theory-based Approach’, working paper, University of Utah. Available at
http://ssrn.com/abstract=279309).
Botosan, C. A. and M. A. Plumlee (2005) ‘Assessing Alternative Proxies for the Expected
Risk Premium’, forthcoming in Accounting Review.
Bradley, M., A. Desai and E. Kim (1998) ‘Synergistic Gains from Corporate Acquisitions
and their Division Between the Stockholders of Target and Acquiring Firms’, Journal
of Financial Economics 21: 3–40.
Brealey, R. A. and S. C. Myers (2000) Principles of Corporate Finance, New York: McGraw-
Hill/Irwin.
Bressler, R. G. (1966) ‘The Measurement of Productivity Efficiency’, Proceedings
of the Thirty-Ninth Meeting of the Western Farm Economic Association,
pp. 129–36.
Brewer, E. I., W. E. I. Jackson and J. A. Jagtiani (2000) ‘The Price of Bank Mergers in
the 1990s’, Federal Reserve Bank of Chicago, Economic Perspectives, First Quarter,
pp. 2–23.
Brockett, P. L., A. Charnes, W. W. Cooper, Z. M. Huang and D. B. Sun (1997) ‘Data
Transformation in DEA Cone Ratio Envelopment Approaches for Monitoring Bank
Performances’, European Journal of Operational Research 98: 250–68.
Brown, M. Z. (1995) ‘The Measurement of Relative Efficiency in Banking’, in J. Revell
(ed.), Deregulation and Bank Efficiency, University of Wales, Bangor, Institute of
European Finance, Research Monograph in Banking and Finance, M95/1.
Brown, M. Z. (1996) ‘Relative Efficiencies and Productivity Indices: an Empirical
Investigation of the Banking Sectors in Eight European Countries’, paper presented
310 References
at the annual conference of the European Association of University Teachers in

Banking and Finance, University of Malta, September.
Brown, S., K. Lo and T. Lys (1999) ‘Use Of R2 in Accounting Research: Measuring
Changes in Value Relevance over the Last Four Decades’, Journal of Accounting and
Brown, T. J., G. A. Churchill Jr. and J. P. Peter (1993) ‘Improving the Measurement of
Service Quality’, Journal of Retailing 69, 1: 127–39.
Burgstanhler, D. and I. Dichev (1997) ‘Earnings, Adaptation and Equity Value’,
Accounting Review 73: 187–215.
Buzzell, R. D. and B. T. Gale (1997) The PIMS Principles, New York: Free Press.
Callen, J. L. and M. Morel (2001) ‘Linear Accounting Valuation When Abnormal
Earning are AR (2)’, Review of Quantitative Finance and Accounting 16: 191–203.
Callen, J. L. and D. Segal (2002) ‘An Empirical Test of the Feltham-Ohlson (1995)
Model’, Rotman Accounting Working Papers Series, no. 5.
Caparelli, F. (2004) Economia del Mercato mobiliare, Milan: McGraw-Hill.
Capizzi, V. (2001) ‘Il capital asset pricing model e le operazioni di corporate e invest-
ment banking’, SDA Bocconi, working paper, no. 57.
Carbo, S., E. P. M. Gardener, P. Molyneux and J. Williams (2000a) ‘Adaptive Strategies
by European Savings Banks’, in J. Falzon and E. P. M. Gardener (eds), Strategic
Challenges in European Banking, London: Macmillan, pp. 181–210.
Carbo, S., E. P. M. Gardener and J. Williams (2000b) ‘Efficiency and Technical Change
in Europe’s Savings Banks Industry’, Revue de la Banque 6: 381–94.
Carbo, S. and D. B. Humphrey (2004) ‘Opening the Black Box: Finding the Source of
Cost Inefficiency’, paper presented at the annual conference of the European
Association of University Teachers in Banking and Finance, Krakow University,
September.
Carbo, S. and J. Williams (2000) ‘Stakeholder Value in European Savings Banks’, in
L. Schuster (ed.), Shareholder Value Management in Banks, London: Macmillan,
pp. 134–66.
Carlesi, F. (1999) ‘Il Value Based Management in Banca’, APB News 3: 39–62.
Carman, J. M. (1990) ‘Consumer Perceptions of Service Quality: an Assessment of the
SERVQUAL Dimensions’, Journal of Retailing 65: 33–55.
Carretta, A. (1998) ‘Modelli di corporate governance ad azionariato diffuso. Il sistema
informativo e gli indicatori di performance per il consiglio di amministrazione’, in
G. Airoldi and G. Forestieri (eds), Corporate Governance, Milan: ETAS Libri.
Carretta, A. (ed.) (2002) Il governo del cambiamento culturale in banca, Rome: Bancaria
Editrice.
Carretta, A., F. Fiordelisi, G. Forestieri and P. Schwizer (1998) ‘Corporate Governance’,
SDA Bocconi, Research Division, no. 17/98, Milan.
Castellani, G., M. De Felice, F. Moriconi and C. Mottura (1993) Un corso sul controllo del
rischio di tasso di interesse, Bologna: Il Mulino.
Casu, B. and C. Girardone (2004a) ‘An Analysis of the Relevance of OBS Items in
Explaining Productivity Change in European Banking’, paper presented at EFMA
Basel Meeting.
Casu, B. and C. Girardone (2004b) ‘Financial Conglomerates: Strategic Drive and
Impact on Bank Efficiency and Productivity’, Applied Financial Economics 14, 10:
687–96.
Casu, B., C. Girardone and P. Molyneux (2004) ‘Productivity Change in Banking: a
Comparison of Parametric and Non-Parametric Approaches’, Journal of Banking and
Finance 28: 2521–40.
Cavalieri, E. (2000) Economia aziendale, 2 vols, Torino: Giappichelli.
References 311
Caves, D. W., L. R. Christensens and W. E. Diewert (1982) ‘The Economic Theory of

Index Numbers and the Measurement of Input, Output and Productivity’,
Econometrica 50: 1393–414.
Cebenoyan, A. S, E. S. Cooperman, C. A. Register and S. C. Hudgins (1993) ‘The
Relative Cost Efficiency of Stock versus Mutual S&Ls: a Stochastic Cost Frontier
Approach’, Journal of Financial Services Research 7: 151–70.
Chaffai, M. E., M. Dietsch and A. Lozano-Vivas (2001) ‘Technological and
Environmental Differences in the European Banking Industries’, Journal of Financial
Services Research 19: 147–62.
Chang, J. (1998) ‘The Decline in Value Relevance of Earnings and Book Values’,
working paper University of Pennsylvania (January).
Charnes, A., W. W. Cooper, Z. M. Huang and D. B. Sun (1990) ‘Polyhedral Cone Ratio
DEA Models with an Illustrative Application to Large Commercial Banks’, Journal of
Econometrics 46: 73–91.
Charnes, A., W. W. Cooper, A. Lewin and L. M. Seiford (1994) Data Envelopment
Analysis: Theory, Methodology and Application, Boston: Kluwer.
Charnes, A., W. W. Cooper and E. Rhodes (1978) ‘Measuring for Efficiency of Decision
Making Units’, European Journal of Operational Research 2: 429–44.
Charreaux, G. and P. Desbrières (2001) ‘Corporate Governance : Stakeholder Value
versus Shareholder Value’, Journal of Management and Governance 5: 107–28.
Cleland, A. S. and A. B. Bono (1997) ‘Building Customer and Shareholder Value’,
Strategy and Leadership 25: 22–8.
Coelli, T. (1995) ‘Estimators and Hypothesis Tests for a Stochastic Frontier Function:
a Monte Carlo Analysis’, Journal of Productivity Analysis 6: 247–68.
Coelli, T. (1996) ‘A Guide to DEAP Version 2.1: a Data Envelopment Analysis
(Computer) Program’, CEPA working paper, no. 8.
Coelli, T. (1998) ‘A Multi-stage Methodology for the Solution of Oriented DEA Models’,
CEPA working paper, no. 1.
Coelli, T., D. S. Prasada Rao and G. E. Battese (1997) An Introduction to Efficiency and
Productivity Analysis, Boston, Dordrecht and London: Kluwer.
Cohen, R. D. (2004a) ‘The Optimal Capital Structure of Depository Institutions’,
Wilmott Magazine, March, 1–11.
Cohen, R. D. (2004b) ‘An Analytical Process for Generating the WACC Curve and
Locating the Optimal Capital Structure’, Wilmott Magazine, November: 1–11.
Cole, D. W. (1972) ‘Return on Equity Model for Banks’, Banker’s Magazine 55: 40–7.
Collins, D. W., E. L. Maydew and I. S. Weiss (1997) ‘Changes on the Value-relevance of
Earnings and Book Value over the Past Forty Years’, Journal of Accounting and
Cook, W. D., M. Kress and L. M. Seiford (1996) ‘Data Envelopment Analysis in the
Presence of Both Quantitative and Qualitative Factors’, Journal of Operational Research
Society 47: 945–53.
Cooper, W. W., L. M. Seiford and K. Tone (2000) Data Envelopment Analysis, Boston:
Kluwer.
Copeland, T., T. Koller and J. Murrin (2000) Valuation: Measuring and Managing the
Value of Companies, New York: John Wiley & Sons.
Cornett, M. M., G. Hovakimian and D. Palia (2003) ‘The Impact of the Manager-
Shareholder Conflict on Acquiring Bank Returns’, Journal of Banking and Finance
27: 103–31.
Costabile, M. (1996) ‘La misurazione del valore per il cliente: Aspetti metodologici e
implicazioni per la gestione dei processi di scambio’, UTET, Torino.
Costi, A. (1994) L’ordinamento bancario Italiano, Bologna: Il Mulino.
312 References
Credit Suisse Financial Products (1997) Credit Risk: a Credit Risk Management Framework,
Technical Document, Credit Suisse.
Croning, Jr. J. J. and S. A. Taylor (1992) ‘Measuring Service Quality: a Reexamination
and Extension’, Journal of Marketing 56: 55–68.
Croning, Jr. J. J. and S. A. Taylor (1994) ‘SERVPERF versus SERVQUAL: Reconciling
Performance-Based and Perception-Minus-Expectations Measurement of Service
Quality’, Journal of Marketing 58: 125–31.
Crosbie, P. J. (1999) Modelling Default Risk, San Francisco: KMV Corporation.
Cruickshank, D. (2000) Competition in UK Banking, Report to the Chancellor of the
Exchequer, London: HMSO.
Cryer, J. D. and R. B. Miller (1994) Statistics for Business: Data Analysis and Modelling,
Duxburry Series in Business Statistics and Decision Sciences.
Cummins, D, G. Turchetti and M. Weiss (1995) ‘Productivity and Technical Efficiency
in the Italian Insurance Industry’, working paper, Wharton Financial Institution
Center, University of Pennsylvania.
Cummins, D. and H. Zi (1995) ‘Measuring Economic Efficiency of the US Life
Insurance Industry: Econometric and Mathematical Programming Techniques’,
Wharton Financial Institutions Center, University of Pennsylvania, working paper.
Cybo-Ottone, A. and M. Murgia (2000) ‘Mergers and Shareholder Wealth in European
Banking’, Journal of Banking and Finance 24: 831–59.
Cyert, R. M. and J. G. March (1963) A Behavioural Theory of the Firm, Englewood Cliffs,
NJ: Prentice-Hall.
Dacrema, P. and P. De Sury (1987) ‘La funzione obiettivo dell’impresa’, in R. Ruozi
(ed.), Profitto e dimensioni nelle banche italiane, Torino: UTET.
Dalborg, H. (1999) ‘Shareholder Value in Banking’, paper presented to the May 1999
session of the Institut International d’Etudes Bancaire.
Dallocchio, M. (2004) Finanza Aziendale, Milan: EGEA.
Damodaran, A. (1994) Damodaran on Valuation, New York: John Wiley & Sons.
Damodaran, A. (1996) Investment Valuation, New York: John Wiley & Sons.
Damodaran, A. (1999a) ‘Value Creation and Enhancement: Back to the Future’,
Leonard N. Stern School of Business, Department of Finance, Working Paper Series,
no. 18 (FIN-99-018).
Damodaran, A. (1999b) ‘Estimating Risk Parameters’, Leonard N. Stern School
of Business, Department of Finance, Working Paper Series, no. 19 (FIN-99-019).
Damodaran, A. (1999c) ‘Estimating Risk Premiums’, Leonard N. Stern School
of Business, Department of Finance, Working Paper Series, no. 19 (FIN-99-021).
Damodaran, A. (1999d) ‘Estimating Risk Free Rates’, Leonard N. Stern School of
Business, Department of Finance, Working Paper Series, no. 19 (FIN-99-021).
Damodaran, A. (2001) The Dark Side of Valuation, New Jersey: Financial Times, Prentice
Hall.
Damodaran, A. (2005) Applied Corporate Finance, New York: John Wiley & Sons.
Davidson, R. and J. G. McKinnon (1993) Estimation and Inference in Econometrics,
Oxford: Oxford University Press.
Davis, D. and K. Lee (1997) ‘A Practical Approach to Capital Structure for Banks’,
Journal of Applied Corporate Finance 10, 1: 33–43.
De Laurentis, G. (2001) Rating interni e credit risk management, Rome: Bancaria Editrice.
De Laurentis, G. (ed.) (2004) Performance Measurement Frontiers in Banking and Finance,
Milan: EGEA.
De Young, R. (1997) ‘A Diagnostic Test for the Distribution-Free Efficiency Estimator:
an Example using US Commercial Bank Data’, European Journal of Operational
Research 98: 243–9.
References 313
De Young, R. (1998) ‘Management Quality and X-inefficiency in National Banks’,

Journal of Financial Services Research 13: 5–22.
Debreu, G. (1951) ‘The Coefficient of Recourse Utilisation’, Econometrica 19: 273–92.
Dechow, P. M., A. P. Hutton and R. G. Sloan (1999) ‘An Empirical Assessment of the
Residual Income Valuation Model’, Journal of Accounting and Economics 26: 1–34.
Dekimpe, M. G., J-B. E. M. Steenkamp and M. Mellens (1997) ‘Decline and Variability
in Brand Loyalty’, International Journal of Research in Marketing 14: 405–20.
DeLong, G. L. (2001) ‘Stockholder Gains from Focusing versus Diversifying Bank
Mergers’, Journal of Financial Economics 59: 221–52.
Dermine, J. (2002) ‘European Banking: Past, Present and Future’, paper presented at the
second ECB conference on ‘The transformation of the European financial system’,
Frankfurt am Main, 24 and 25 October.
Di Antonio, M. (1999) ‘Vecchi e nuovi indicatori di performance economico-
finanziaria in banca: un inquadramento concettuale’, APB News 3: 63–92.
Di Antonio, M. (2002) Creazione di valore e controllo strategico nella banca, Rome:
Bancaria Editrice.
Di Antonio, M. (2004) ‘The Measurement, Control and Management of Operating
Costs in Banking’, in G. De Laurentis (ed.), Performance Mesurement Frontiers in
Banking and Finance, Milan: EGEA, pp. 197–231.
Dietsch, M. and A. Lozano-Vivas (2000) ‘How the Environment Determines Banking
Efficiency: a Comparison between French and Spanish Industries’, Journal of Banking
and Finance 24: 985–1004.
Dillon, R. D. and J. E. Owers (1997) ‘EVA as a Financial Metric: Attributes, Utilization,
and Relationship to NPV’, Financial Practice and Education Spring/Summer: 32–40.
Dorman, K. G. (1994) ‘Mystery Shopping Results Can Shape Your Future’, Bank
Marketing 26: 17–21.
Dougherty, C. (1992) Introduction to Econometrics, Oxford: Oxford University Press.
Dowling, G. R. and M. Uncles (1997) ‘Do Customer Loyalty Programs Really Work?’
Sloan Management Review 38: 71–82.
Drake, C., A. Gwynne and N. Waite (1998) ‘Barclays Life Customer and Loyalty
Tracking Survey: a Demostrantion of Customer Loyalty Research in Practice’,
International Journal of Bank Marketing 16, 7: 287–92.
Drake, L. Howcroft (1993) ‘A Study of Relative Efficiency of UK Bank Branches’,
Loughborough University of Technology, Economic Research Paper, no. 13.
Drucker, P. F. (1995) ‘The Information Executives Truly Need’, Harvard Business Review
73, 1: 54–62.
Easton, P. and T. Harris (1991) ‘Earnings as an Explanatory Variable of Returns’, Journal
of Accounting Research 29: 19–36.
Easton, P. D. (1998) ‘Discussion of Revalued Financial, Tangible and Intangible Assets:
Association with Share Prices and Non-market-based Value Estimates’, Journal of
Accounting Research 36 (supplement): 235–47.
Easton, P. D. and G. A. Sommers (2003) ‘Scale and the Scale Effect in Market-based
Accounting Research’, Journal of Business, Finance and Accounting 30, 1: 25–55.
ECB (European Central Bank) (2000) EU Banks’ Income Structure, April.
ECB (European Central Bank) (2004) EU Banking Sector Stability, Banking Supervision
Committee (BSC) of EU System of Central Banks, November.
Eisenbeis, R. A., G. D. Ferrier and S. H. Kwan (1996) ‘An Empirical Analysis of the
Informativeness of Programming and SFA Efficiency Scores: Efficiency and Bank
Performance’, University of North Carolina, working paper, Chapel Hill, NC (April).
Elton, E. J. and M. J. Gruber (1995) Modern Portfolio Theory and Investment Analysis, New
York: John Wiley & Sons.
314 References
Esho, N. and I. G. Sharpe (1996) ‘X-efficiency of Australian Permanent Building

Societies, 1974–1990’, The Economic Record 72: 246–59.
Fama, E. F. (1976) Foundation of Finance, New York: Basic Books.
Fama, E. F. (1980) ‘Agency Problems and the Theory of the Firm’, Journal of Political
Economy 88: 288–307.
Fama, E. F. and M. C. Jensen (1983) ‘Separation of Ownership and Control’, Journal of
Law and Economics 26: 301–25.
Fama, E. F. and R. K. French (1999) ‘The Corporate Cost of Capital and the Return on
Corporate Investment’, Journal of Finance 54: 1939–67.
Fama, E. F. and R. K. French (2002) ‘The Equity Premium’, Journal of Finance 57:
637–59.
Fare, R. S., S. Grosskopf and C. A. K. Lovell (1994) Production Frontiers, Cambridge:
Cambridge University Press.
Fare, R. S. and C. A. K. Lovell (1978) ‘Measuring the Technical Efficiency of
Production’, Journal of Economic Theory 19: 150–62.
Farrell, M. J. (1957) ‘The Measurement of Productive Efficiency’, Journal of the Royal
Statistical Society, Series A, 120, Part 3: 253–90.
Favero, C. A. and L. Papi (1995) ‘Technical Efficiency and Scale Efficiency in the Italian
Banking Sector: a Non-parametric Approach’, Applied Economics 27: 385–95.
Fecher, F., D. Kessler, S. Perelman and P. Pestieau (1993) ‘Productive Performance of
the French Insurance Industry’, Journal of Productivity Analysis 4: 77–93.
Fecher, F. and P. Pestieau (1993) ‘Efficiency and Competition in OECD Financial
Services’, in H. O. Fried, C. A. K. Lovell and S. S. Schmidt (eds), The Measurement of
Productive Efficiency, New York: Oxford University Press, pp. 374–85.
Feltham, G. and J. A. Ohlson (1995) ‘Valuation and Clean Surplus Accounting for
Operating and Financial Activities’, Contemporary Accounting Research 11: 689–731.
Fernández, P. (2002) ‘A Definition of Shareholder Value Creation’, University of
Navarra, IESE, research paper no. 448.
Ferrier, G. D. and J. G. Hirschberg (1997) ‘Bootstrapping Confidence Intervals for
Linear Programming Efficiency Scores: With an Illustration Using Italian Banking
Data’, Journal of Productivity Analysis 8: 19–33.
Ferrier, G. D., K. Kerstens and P. Vanden Eeckaut (1994) ‘Radial and Non-radial
Technical Efficiency Measures on a DEA Reference of Technology: a Comparison
Using Banking Data’, Researches Economiques de Louvain, 60: 449–79.
Ferrier, G. D. and C. A. K. Lovell (1990) ‘Measuring Cost Efficiency in Banking:
Econometric and Linear Programming Evidence’, Journal of Econometrics 46: 229–45.
Filotto, U. (2000) E-finance e E-commerce: Banche e nuovi competitors, Rome: Bancaria
Editrice.
Filotto, U. (2002) La ‘nuova’ web bank: Cultura, organizzazione e tecnologia dopo la crisi di
Internet, Rome: Bancaria Editrice.
Fiordelisi, F. (1997) ‘La qualità dei servizi bancari alla clientela privata’, Lettera market-
ing 17, 2: 14–21.
Focarelli, D. and A. F. Pozzolo (2001) ‘The Patterns of Cross-border Bank Mergers and
Shareholdings in OECD Countries’, Journal of Banking and Finance 25: 2305–37.
Forbes, W. and P. Molyneux (1994) ‘Corporate Control and the Restructuring of
European Banking’, Journal of International Financial Markets, Institutions and Money 3:
101–16.
Forestieri, G. (1998) ‘La corporate governance negli schemi interpretativi della lettera-
ture’, in G. Airoldi and G. Forestieri (eds), Corporate Governance, Milan: ETAS Libri.
Fornell, C. (1992) ‘A National Customer Satisfaction Barometer: the Swedish Experi-
ence’, Journal of Marketing 56: 6–21.
References 315
Fornell, C. and B. Wernerfelt (1987) ‘Defensive Marketing Strategy by Consumer

Complaint Management: a Theoretical Analysis’, Journal of Marketing Research
24: 337–46.
Fornell, C. and B. Wernerfelt (1998) ‘A Model for Customer Complaint Management’,
Marketing Science 7: 271–86.
Førsund, F. R. (1996) ‘On the Calculation of the Scale Elasticity in DEA Models’, Journal
of Productivity Analysis 7: 283–302.
Førsund, F. R. (2001) ‘Categorical Variable in DEA’, International Centre for Economic
Research (ICER) working paper, March.
Forza, C. and R. Filippini (1998) ‘TQM Impact on Quality Conformance and Customer
Satisfaction: a Causal Model’, International Journal of Production Economics 55, 1: 1–20.
Foster, G. and M. Gupta (1997) ‘The Customer Profitability Implication of Customer
Satisfaction’, Stanford University and Washington University, working paper.
Francis, J. and K. Schipper (1999) ‘Have Financial Statements Lost their Relevance?’
Journal of Accounting Research 37: 319–52.
Frankel, R. and C. M. C. Lee (1998) ‘Accounting Valuation, Market Expectations and
Cross-Sectional Stock Returns’, Journal of Accounting Economics 25: 283–319.
Fried, H. O., C. A. K. Lovell and S. S. Schmidt (1993) The Measurement of Productive
Efficiency, New York: Oxford University Press.
Fried, H. O., C. A. K. Lovell and S. Yaisawarng (1999) ‘The Impact of Mergers on Credit
Union Service Provision’, Journal of Banking and Finance 23: 367–86.
Friend, I. and I. Tokutsu (1987) ‘The Cost of Capital to Corporations in Japan and the
USA’, Journal of Banking and Finance 11: 313–27.
Gale, B. (1992) Monitoring Customer Satisfaction and Market Perceived Quality, Worth
Repeating Series, no. 922CS01, American Marketing Association, Chicago.
Gallant, A. R. (1981) ‘On the Bias in Flexible Functional Form Forms and Essentially
Unbiased Form: the Fourier Flexible Form’, Journal of Econometrics 15: 211–45.
Gallant, A. R. (1982) ‘Unbiased Determination of Production Technologies’, Journal of
Econometrics 20: 285–324.
Gardener, E. P. M. (1997) ‘The Challenge of Deregulation for the European Banks’,
Institute of European Finance, University of Wales Bangor, working paper, no. 3.
Garman, M. and J. Ohlson (1980) ‘Information and the Sequential Valuation of Assets
in Arbitrage Free Economies’, Journal of Accounting Research 18: 420–40.
Garvey, G. T. and T. T. Milbourn (2000) ‘EVA versus Earnings: Does it Matter Which
is More Highly Correlated with Stock Returns?’, Journal of Accounting Research
38 (supplement): 209–45.
Garvin, D. A. (1988) Managing Quality: the Strategic and Competitive Edge, New York: Free
Press.
Gjesdal, F. (1981) ‘Accounting for Stewardship’, Journal of Accounting Research 19:
208–31.
Glass, J. C. and D. G. McKillop (2000) ‘A Post-deregulation Analysis of the Sources
of Productivity Growth in UK Building Societies’, The Manchester School 68, 3:
360–85.
Glassman, D. M. (1997) ‘Contracting for Value: EVA and the Economics of
Organisations’, Journal of Applied Corporate Finance 10, 2: 110–23.
Golany, B. and J. E. Storbeck (1999) ‘A Multiperiod DEA Study of Branches of a Large
US Bank’, Interfaces 29: 14–26.
Gordon, M. J. and J. Shapiro (1956) ‘Capital Equipment Analysis: the Residual Rate of
Profit’, Management Science 3: 102–10.
Gordon, R. A. (1945) ‘Business Leadership in the Large Corporation’, working paper,
Brookings Institution.
316 References
Gordon, R. A. (1948) ‘Short Period Price Determination in Theory and Practice’,

American Economic Review (June): 265–88.
Grant, J. L. (1996) ‘Foundations of EVA for Investment Managers’, Journal of Portfolio
Management (fall): 41–8.
Greene, W. H. (1993) ‘The Econometric Approach to Efficiency Analysis’, in
H. O. Fried, C. A. K. Lovell and S. S. Schmidt (eds), The Measurement of Productive
Efficiency, New York: Oxford University Press.
Green, W. H. (1997) LIMDEP Version 7.0: User’s Manual and Reference Guide, New York:
Econometric Software Inc.
Greenspan, A. (1996) ‘Remarks at the Annual Convention of the American Bankers
Association’, Honolulu, Hawaii, October 5.
Griffell-Tatjé, E. and C. A. K. Lovell (1995) ‘A Note on the Malmquist Productivity
Index’, Economics Letters 47, 18: 169–75.
Griffell-Tatjé, E. and C. A. K. Lovell (1997) ‘The Source of Productivity Change in
Spanish Banking’, European Journal of Operational Research 98: 364–80.
Grilliches, Z. (1996) ‘The Discovery of Residual: a Historical Note’, Journal of Economic
Literature 34, 3: 1324–30.
Grönroos, C. (1983) ‘Service Management and Marketing in the Service Sector’,
Marketing Science Institute, report no. 104, Cambridge, MA.
Grönroos, C. (1984) ‘A Service Quality Model and its Marketing Implications’, European
Journal of Marketing 19: 36–44.
Grönroos, C. (1990) Service Management ands Marketing, Lexington: Lexington Books.
Grosskopf, S. (1993) ‘Efficiency and Productivity’, in H. O. Fried, C. A. K. Lovell and
S. S. Schmidt (eds), The Measurement of Productive Efficiency, New York: Oxford
University Press.
Grosskopf, S. (1996) ‘Statistical Inference and Non-parametric Efficiency: a Selective
Survey’, Journal of Productivity Analysis 7: 161–76.
Group of Ten (2001) ‘Consolidation in the Financial Sector’, January, available at
www.oecd.org.
Gualandri, E. (2004) ‘Il quadro normativo e di vigilanza sulle istituzioni creditizie’, in
Onado (ed.), La banca come impresa, Bologna: Il’Mulino.
Guatri, L. (1991) La teoria di creazione del valore: una via europea, Milan: EGEA.
Guatri, L. (1998) Trattato sulla valutazione delle aziende, Milan: EGEA.
Guatri, L. and L. Sicca (2000) Strategie, leve del valore, valutazione delle aziende, Milan:
EGEA.
Habermayer, L. (1997) ‘Agency Conflicts and Managerial Discretion’, Simon School of
Business, working paper, no. 21–12.
Hall, P., W. Hardle and L. Simar (1995) ‘Iterated Bootstrap with Application to Frontier
Models’, Journal of Productivity Analysis 6: 63–76.
Hall, R. L. and C. J. Hitch (1939) ‘Price Theory and Business Behaviour’, Oxford
Economic Papers, May (translated in Italian ‘Teoria del prezzo e comportamnento
d’impresa’, in G. Zanetti (ed.), Contributi per un’analisi economica dell’impresa, Naples:
Liguori).
Hallowell, R. (1996) ‘The Relationship of Customer Satisfaction, Customer Loyalty and
Profitability: an Empirical Study’, International Journal of Service Industry Management
7, 4: 27–42.
Hamilton, R. (1777) An Introduction to Merchandize. Edinburgh.
Hancock, D. (1986) ‘A Model of the Financial Firm with Imperfect Asset and Deposit
Elasticities’, Journal of Banking and Finance 10: 37–54.
Hand, J. R. M. and W. R. Landsman (1998) ‘Testing the Ohlson Model: v or not v,
that is the Question’, working paper, University of North Carolina at Chapel Hill,
available at http://ssrn.com/abstract=126308.
References 317
Hanke, E. (1993) ‘Through the Looking Glass’, Credit Union Management, August: 42:
45–6.
Hannan, T. H. (1991) ‘Bank Commercial Loan Markets and the Role of the Market
Structure Evidence from Surveys of Commercial Lending’, Journal of Banking and
Finance 15: 133–49.
Haugen, R. A. (2000) Modern Investment Theory, 5th edition, New Jersey: Prentice Hall.
Hauser, J. R., D. I. Simester and B. Wernerfelt (1994) ‘Customer Satisfaction Incentives’,
Marketing Science 13, 4: 327–50.
Hawawini, G. A. and I. Swary (1990) Mergers and Acquisitions in the US Banking Industry:
Evidence from Capital Markets, Amsterdam, North-Holland: Elsevier Science Publishers
B.V.
Hayn, C. (1995) ‘The Information Content of Losses’, Journal of Accounting and
Haynes, M. and S. Thompson (1999) ‘The Productivity Effects of Bank Mergers:
Evidence from the UK Building Societies’, Journal of Banking and Finance 23: 825–46.
Helfert, E. A (1999) Techniques of Financial Analysis: a Guide to Value Creation, New York:
McGraw-Hill/Irwin.
Heskett, J. L., T. O. Jones, G. W. Loveman, W. E. Sasser and L. A. Schlesinger (1994)
‘Putting the Service Profit Chain to Work’, Harvard Business Review (March–April):
164–74.
Heskett, J. L., W. E. Sasser and L. A. Schlesinger (1997) The Service Profit Chain,
Hesselink, M. and T. Van der Wiele (2003) Mystery Shopping: In-depth Measurement of
Customer Satisfaction, ERIM Report series in Research in Management, ERS-2003-020-
ORG.
Hicks, J. (1935) ‘Annual Survey of Economic Theory: the Theory of Monopoly’,
Econometrica 3, 1: 1–20.
Hill, N. (1996) Handbook of Customer Satisfaction Measurement, Aldershot, UK: Gower.
Hirschey, M. (1985) ‘Market Structure and Market Value’, Journal of Business
58: 89–98.
Hjalmarsson, L., S. C. Kumbhakar and A. Heshmati (1996) ‘DEA, DFA and SFA: a
Comparison’, Journal of Productivity Analysis 7: 303–27.
Hoffman, G. (1993) ‘Customers Can Hone Mystery Shopping’, Bank Marketing
25: 36–7.
Holliday, K. K. (1994) ‘Mutual Funds’, Bank Marketing 26: 23–31.
Holthausen, R. W. and R. L. Watts (2001) ‘The Relevance of the Value-Relevance
Literature for Financial Accounting Standard Setting’, Journal of Accounting and
Hörther, S. (2000) ‘Shareholder Value: Changing the Face of the European Financial
Industry’, in L. Schuster (ed.), Shareholder Value Management in Banks, London:
Macmillan, pp. 13–35.
Hotchkiss, D. A. (1995) ‘What Do Your Customers Really Think?’, Bank Marketing
27: 13–20.
Houghaard, J. L. (1999) ‘Fuzzy Scores of Technical Efficiency’, European Journal of
Operational Research 115: 529–41.
Houston, J. F., C. M. James and M. D. Ryngaert (2001) ‘Where Do Merger Gains Come
From? Bank Mergers from the Perspective of Insiders and Outsiders’, Journal of
Financial Economics 60: 285–331.
Houston, J. F. and M. Ryngaert (1994) ‘The Overall Gains from Large Bank Mergers’,
Houston, J. F. and M. Ryngaert (1997) ‘Equity Issuance and Adverse Selection: a Direct
Test Using Conditional Stock Offers’, Journal of Finance 52: 197–219.
318 References
Hudgins, S. C. and B. Seifert (1996) ‘Stockholders and International Acquisitions of

Financial Firms: an Emphasis on Banking’, Journal of Financial Services Research 10:
163–80.
Hughes, J. P., W. Lang, L. J. Mester and C. G. Moon (1999) ‘The Dollars and Sense of
Bank Consolidation’, Journal of Banking and Finance 23: 291–324.
Humphrey, D. B. (1987) ‘Cost Dispersion and the Measurement of Economies in
Banking’, Federal Reserve Bank of Richmond, Economic Review (May): 24–38.
Humphrey, D. B. and L. B. Pulley (1997) ‘Bank Responses to Deregulation: Profits,
Technology and Efficiency’, Journal of Money, Credit and Banking 29: 73–93.
International Accounting Standards Board (2004) IAS39: Financial Instruments:
Recognition and Measurement, London: International Accounting Standards Board.
Itami, H. and T. Roehl (1987) Mobilizing Invisible Assets, Cambridge, MA: Harvard
University Press.
Ittner, C. and D. F. Larckner (1996) ‘Measuring the Impact of Quality Initiatives on
Firm Financial Performance’, in S. Ghosh and D. Fedor (eds), Advances in the
Management of Organisational Quality, Volume 1, Greenwich, CT: JAI Press, pp. 1–37.
Ittner, C. D. and D. F. Larcker (1998) ‘Are Non-financial Measures a Leading Indicator
of Financial Performance? An Analysis of Customer Satisfaction’, Journal of
Accounting Research 36, supplement: 1–35.
Jacobson, R. and D. A. Aaker (1987) ‘The Financial Information Content of Perceived
Quality’, Journal of Marketing Research 31: 191–201.
Jensen, M. C. and W. H. Meckling (1976) ‘The Theory of the Firm: Managerial
Behaviour, Agency Costs and Ownership Structure’, Journal of Financial Economics 3,
4: 305–60.
Jensen, M. C. and R. S. Ruback (1983) ‘The Market for Corporate Control’, Journal of
Johnston, R. (1995) ‘The Determinants of Service Quality: Satisfiers and Dissatisfiers’,
International Journal of Service Industry Management 6, 5: 53–71.
Johnston, R. (1997) ‘Identifying the Critical Determinants of Service Quality in Retail
Banking: Importance and Effect’, International Journal of Bank Marketing 15,
4: 111–16.
Jondrow, J., C. A. K. Lovell, I. V. Materov and P. Schmidt (1982) ‘On Estimation of
Technical Inefficiency in the Stochastic Frontier Production Function Model’, Journal
of Econometrics 19: 233–8.
Jones, T. O. and W. E. Sasser (1995) ‘Why Satisfied Customers Defect’, Harward Business
Review (November–December): 89–9.
Kane, E. J. (2000) ‘Incentives for Banking Megamergers: What Motives Might Central
Bank Economists Infer from Event-Study Evidence?’, Journal of Money, Credit and
Banking 32: 671–99.
Kantor, J. and S. Maital (1999) ‘Measuring Efficiency by Product Group: Integrating
DSEA with Activity Based Accounting in Large Mideast Banks’, Interfaces 29: 27–36.
Karceski, J., S. Ongena and D. C. Smith (2004) ‘The Impact of Bank Consolidation on
Commercial Borrower Welfare’, FRB International Finance Discussion Paper, no. 679.
Available at http://ssrn.com/abstract=244071).
Kester, W. C. and T. Luehrman (1989) ‘Real Interest Rates and the Cost of Capital:
a Comparison of the United States and Japan’, Japan and the World Economy 1,
3: 279–301.
Kester, W. C. and T. Luehrman (1992) ‘What Makes You Think US Capital is
So Expensive?’, Continental Bank Journal of Applied Corporate Finance 5: 29–41.
Kim, M. (1985) ‘Scale Economies in Banking: a Methodological Note’, Journal of Money,
Credit and Banking 17: 96–102.
References 319
Kimball, R. C. (1998) ‘Economic Profit and Performance Measurement in Banking’,

New England Economic Review, Federal Researve Bank of Boston, July: 35–53.
Kneip, A., U. Park and L. Simar (1996) ‘A Note on the Convergence of Non-parametric
DEA Efficiency Measures’, Université Catholique de Louvain, Center for Operations
Research and Econometrics (CORE), Discussion Paper, no. 39.
Kneip, A. and L. Simar (1996) ‘A General Framework for Frontier Estimation with Panel
Data’, Journal of Productivity Analysis 7: 187–212.
Knight, J. A. (1997) Value Based Management, Developing a Systematic Approach to Create
Shareholder Value, New York: McGraw-Hill.
Kock, W. T. and S. S. McDonald (2002) Bank Management, 5th edition, Fort Worth:
South-Western College Publishing.
Koop, R. J., V. K. Smith and W. J. Vaughan (1982) ‘Stochastic Cost Frontiers
and Perceived Technical Inefficiency’, in V. K. Smith (ed.), Advances in Applied Micro-
economics, vol. 2, Greenwich, Conn: JAI Press.
Koopmans, T. C. (1951) ‘An Analysis of Production as an Efficient Combination
of Activities’, in T. C. Koopmans (ed.), Activity Analysis of Production and Allo-
cation, Cowles Commission for Research in Economics, Monograph, no. 13,
New York.
Koopmans, T. C. (1957) Three Essays on the State of Economic Science, New York:
McGraw-Hill.
Koska, M. T. (1990) ‘High Quality Care and Hospital Profits: Is There a Link?’, Hospitals
5 (March): 62–3.
Kothari, S. P. (2001) ‘Capital Market Research in Accounting’, Journal of Accounting and
Kramer, J. K. and G. Pushner (1997) ‘An Empirical Analysis of Economic Value-Added
as a Proxy for Market Value Added’, Financial Practice and Education (Spring/Summer):
41–9.
Kumbhakar, S. C., A. Lozano-Vivas, C. A. K. Knox Lovell and I. Hasan (1999) ‘The
Effect of Deregulation on the Performance of Financial Institutions: the Case of
Spanish Savings Banks’, New York University, Stern Business School, Finance
Working Paper Series, FIN 99-064.
Kwan, S. H. and R. Eisenbeis (1999) ‘Mergers of Publicly Traded Banking Organizations
Revisited’, Federal Reserve Bank of Atlanta Economic Review 84: 26–37.
LaBarbera, P. and D. Mazursky (1983) ‘A Longitudinal Assessment of Consumer
Satisfaction/Dissatisfaction: the Dynamic Aspect of the Cognitive Process’, Journal of
Marketing Research (November): 393–404.
Lambert, R. A. (1998) ‘Customer Satisfaction and Future Financial Performance:
Discussion of “Are Non-Financial Measures Leading Indicator of Financial
Performance?” An Analysis of Customer Satisfaction’, Journal of Accounting Research
36 (supplement): 37–64.
Land, K., C. A. K. Lovell and S. Thore (1993) ‘Chance Constrained Data Envelopment
Analysis’, Managerial and Decision Economics 14: 541–54.
Lang, G. and P. Welzel (1996) ‘Efficiency and Technical Progress in Banking: Empirical
Results for a Panel of German Cooperative Banks’, Journal of Banking and Finance
20: 1003–23.
Lassar, M. W., C. Manolis and R. D. Winsor (2000) ‘Service Quality Perspectives and
Satisfaction in Private Banking’, Journal of Service Marketing 14, 3: 244–71.
Le, T. and K. Sheehan (1998) ‘Measuring the Relative Marginal Cost of Debt and
Capital for Banks’, Federal Deposit Insurance Corporation, Division of Research and
Statistics, unpublished paper. A summary is available in FRBNY Economic Policy
Review, October 1998.
320 References
Leamer, E. E. and H. B. Leonard (1983) ‘Reporting the Fragility of Regressions

Estimates’, Review of Economics and Statistics 65: 306–17.
Lee, C. M. C. (2001) ‘Market Efficiency and Accounting Research: a Discussion of
Capital Market Research in Accounting by S. P. Kothari’, Journal of Accounting and
Lee, H., Y. Lee and D. Yoo (2000) ‘The Determinants of Perceived Service Quality and
its Relationship with Satisfaction’, Journal of Service Marketing 14, 3: 217–31.
Lee, J., J. Lee. and L. Freick (2001) ‘The Impact of Switching Costs on the Customer
Satisfaction-Loyalty Link: Mobile Phone Service in France’, Journal of Service Marketing
15, 1: 35–8.
Leeds, B. (1992) ‘Mystery Shopping Offers Clues to Quality Service’, Bank Marketing
24: 24–6.
Leeds, B. (1995) ‘Mystery Shopping: From Novelty to Necessity’, Bank Marketing
27: 17–23.
Lehen, K. and A. K. M. Makhija (1997) ‘EVA, Accounting Profits and CEO Turnover:
an Empirical Examination, 1985–1994’, Journal of Applied Corporate Finance 10,
2: 90–7.
Lehen, K. and A. K. M. Makhija (1996) ‘EVA and MVA as Performance Measures and
Signals for Strategic Change’, Strategy and Leadership (May–June): 34–7.
Lehman Brothers (1999) Internet Banking: the Swedish Model, London: Lehman Brothers.
Lehman, B. (1993) ‘Earnings, Dividend Policy, and Present Value Relations: Building
Blocks of Dividend Policy Invariant Cash Flows’, Review of Quantitative Finance and
Accounting 3: 263–82.
Lester, R. A. (1946) ‘Shortcomings of Marginal Analysis for Wage Employment
Problems’, American Economic Review (March): 63–82.
Lester, R. A. (1947) ‘Marginalism, Minimum Wages and Labour Markets’, American
Economic Review (March): 135–48.
Lev, B. (1989) ‘On the Usefulness of Earnings and Earnings Research: Lessons and
Directions from Two Decades of Empirical Research’, Journal of Accounting Research:
Supplement: 153–92.
Lindblom, C. E. (1959) ‘The Science of Muddling Through’, Public Administrative Review
19: 79–88.
Lings, I. N. (2000) ‘Internal Marketing and Supply Chain Management’, Journal of
Service Marketing 14, 1: 27–43.
Lloyd-Williams, D. M., P. Molyneux and J. D. Thornton (1994) ‘Market Structure and
Performance in Spanish Banking’, Journal of Banking and Finance 18: 433–43.
Lo, K. and T. Lys (2000) ‘The Ohlson Model: Contribution to Valuation Theory,
Limitations, and Empirical Applications’, Journal of Accounting, Auditing and Finance
15, 3: 337–67.
Loderer, C. and P. Zgranggen (1999) ‘When Shareholders Choose Not to Maximise
Value: the Union Bank of Switzerland’s 1994 Proxy Fight’, Journal of Applied Corporate
Finance 12, 3: 91–102.
Lovell, C. A. K. (1993) ‘Production Frontier and Productive Efficiency’, in H. O. Fried,
C. A. K. Lovell and S. S. Schmidt (eds), The Measurement of Productive Efficiency, New
York: Oxford University Press, pp. 3–67.
Lovell, C. A. K. and J. T. Pastor (1999) ‘Radial DEA Models Without Input or Without
Outputs’, European Journal of Operational Research 118: 46–51.
Lozano-Vivas, A. (1997) ‘Efficiency of the Spanish Saving Banks from the Profit
Function’, European Journal of Operational Research 98: 381–94.
Lozano-Vivas, A. (1998) ‘Efficiency and Technical Change for Spanish Banks’, Applied
References 321
Maccario, A., A. Sironi and C. Zazzara (2002) ‘Is Banks’ Cost of Equity Capital Different
Across Countries? Evidence from the G10 Countries’ Major Banks’, SDA Bocconi,
working papers, no. 77.
Machlup, F. (1946) ‘Marginal Analysis and Empirical Research’, American Economic
Review (September): 519–54.
Machlup, F. (1947) The Economics of Sellers’ Competition, Baltimore: Johns Hopkins
University Press.
Malmquist, S. (1953) ‘Index Numbers and Indifference Surfaces’, Trabajos de Estatica
4: 209–42.
March, J. G. and H. A. Simon (1958) Organizations, New York: John Wiley & Sons.
Markowitz, H. M. (1952) ‘Portfolio Selection’, Journal of Finance 7, 1: 77–91.
Marris, R. (1964) The Economic Theory of Managerial Capitalism, London: Macmillan.
Marshall, A. (1890) Principles of Economics, London and New York: Macmillan.
Maudos, J. and J. M. Pastor (2001) ‘Cost and Profit Efficiency in Banking: an
International Comparison of Europe, Japan and the USA’, Applied Economic Letters
8: 383–7.
McAllister, P. H. and D. A. McManus (1993) ‘Resolving the Scale Efficiency Puzzle in
Banking’, Journal of Banking and Finance 17: 389–405.
McKillop, D. G. and C. J. Glass (1994) ‘A Cost Model of Building Societies as Producers
of Mortgage and Other Financial Products’, Journal of Business Finance and Accounting
21: 1031–46.
McKillop, D. G., C. J. Glass and C. Ferguson (2002) ‘Investigating the Growth
Performance of UK Credit Unions Using Radial and Non-Radial Efficiency Measures’,
Journal of Banking and Finance 26, 8: 1563–91.
McKillop, D. G., C. J. Glass and Y. Morikawa (1996) ‘The Composite Cost Function and
Efficiency in Giant Japanese Banks’, Journal of Banking and Finance 20: 1651–71.
McTaggart, J. and S. Gillis (1998) ‘Setting Targets to Maximise Shareholder Value’,
Strategy and Leadership 26, 2: 18–22.
Meeusen, W. and J. Van den Broeck (1977) ‘Efficiency Estimation from Cobb-Douglas
Production Functions with Composed Error’, International Economic Review 18:
435–44.
Messori, M. (2001) ‘The Consolidation of the Italian Banking System: Effects
on Competitiveness and Ownership Structures’, paper presented at the Tenth Inter-
national ‘Tor Vergata’ Conference on Banking and Finance, Rome, 5–7 December.
Mester, L. J. (1993) ‘Efficiency in the Savings and Loan Industry’, Journal of Banking and
Finance 17: 267–86.
Mester, L. J. (1996) ‘A Study of Bank Efficiency Taking into Account Risk Preferences’,
Miller, M. H. and F. Modigliani (1961) ‘Dividend Policy, Growth, and the Valuation of
Shares’, Journal of Business 4: 411–33.
Miller, Stephen M. and Athanasios G. Noulas (1996) ‘The Technical Efficiency of Large
Bank Production’, Journal of Banking and Finance, 20, 3: 495–509.
Mitchell, K., N. M. Onvural (1996) ‘Economies of Scale and Scope at Large Commercial
Banks: Evidence from the Fourier Flexible Functional Form’, Journal of Money, Credit
and Banking 28: 178–99.
Modigliani, F. and H. M. Miller (1958) ‘The Cost of Capital, Corporation Finance and
the Theory of Investment’, American Economic Review 48: 261–97.
Monsen, R. J. and A. Downs (1965) ‘A Theory of Large Managerial Firms’, Journal of
Political Economy (June): 2221–36.
Morel, M. (1998) ‘Endogenous Parameter Time Series Estimation: Ohlson versus Nested
Cash Flow Models’, Rutgers University, working paper.
322 References
Morel, M. (1999) ‘Multi-lagged Specification of the Ohlson Model’, Journal of

Accounting, Auditing and Finance 12: 147–61.
Moriconi, F. (1998) Matematica finanziaria, Bologna: Il Mulino.
Morrall, K. (1994) ‘Mystery Shopping Tests Service and Compliance’, Bank Marketing
26: 13–23.
Munari, L. (2000) ‘Customer satisfaction e redditività nelle banche’, Banche e Banchieri
3: 195–225.
Munari, L. (2004) ‘Commercial Performance and Profitability: the Case of the Italian
Banks’, in G. De Laurentis (ed.), Performance Measurement Frontier in Banking and
Finance, Milan: EGEA.
Myers, J. N. (1999) ‘Implementing Residual Income Valuation with Linear Information
Dynamics’, Accounting Review 74: 1–28.
Myers, R. (1996) ‘Metric Wars’, CFO: the Magazine for Senior Financial Executives 12,
10: 41–50.
Natale, E and S. Sarrocco (2001) Value based management nelle banche, Milan: Il Sole
24 ORE Libri.
Nelson, E., R. T. Rust, A. J. Zahorik, R. L. Rose, P. Batalden and B. A. Siemanski (1992)
‘Do Patient Perceptions of Quality Relate to Hospital Financial Performance?’, Journal
of Health Care Marketing 13: 1–13.
Nelson, R. A. (1984) ‘Regulation, Capital Vintage and Technical Change in the Electric
Utility Industry’, Review of Economics and Statistics 66: 59–69.
O’Byrne, S. F. (1996) ‘EVA and Market Value’, Journal of Applied Corporate Finance 9,
1: 116–25.
Ohlson, J. A. (1990) ‘A Synthesis of Security Valuation Theory and the Role of
Dividends, Cash Flows and Earnings’, Contemporary Accounting Research 6: 648–76.
Ohlson, J. A. (1995) ‘Earnings, Book Values, and Dividends in Equity Valuation’,
Contemporary Accounting Research 11: 661–87.
Ohslon, J. A. (2001) ‘Earnings, Book Values, and Dividends in Equity Valuation: an
Empirical Perspective’, Contemporary Accounting Research 18, 1: 107–20.
Ohlson, J. A. and J. Liu (2000) ‘The Feltham-Ohlson (1995) Model: Empirical
Implications’, Journal of Accounting, Auditing and Finance 15: 321–31.
Olesen, O. B. and N. C. Petersen (1995) ‘Chance Constrained Efficiency Valuation’,
Management Science 41: 442–57.
Oliver, R. (1980) ‘A Cognitive Model of the Antecedents and Consequences of
Satisfaction Decisions’, Journal of Marketing 17: 460–9.
Olson, J. C. (1977) ‘Price as an Informational Cue: Effects in Product Evaluation’,
in A. G. Woodside, J. N. Sheth and P. D. Bennett (eds), Consumer and Industrial
Buying Behaviour, New York and Rome: North-Holland Publishing Company,
pp. 267–86.
Onado, M. (2004) La banca come impresa, Bologna: Il Mulino.
Ota, K. (2002) ‘A Test of the Ohlson (1995) Model: Empirical Evidence from Japan’,
SSRN working papers, International Journal of Accounting 37: 157–82.
Padoa-Schioppa, T. (2001) ‘Bank Competition: a Changing Paradigm’, European
Financial Review 5: 13–20.
Panda, Tapan Kumar (2003) ‘Creating Customer Life Time Value Through Effective
CRM in the Financial Services Industry’, Journal of Services Research 2: 157–71.
Parasuraman, A. and L. Colby, (2001) Techno-Ready Marketing: How and Why Your
Customers Adopt Technology, New York: Free Press.
Parasuraman, A., V. A. Zeithaml and L. L. Berry (1985) ‘A Conceptual Model
of Service Quality and its Implications for Future Research’, Journal of Marketing
49: 41–50.
References 323
Parasuraman, A., V. A. Zeithaml and L. L. Berry (1988) ‘SERVAQUAL: a Multiple-item

Scale for Measuring Consumer Perceptions of Service Quality’, Journal of Retailing
64: 12–40.
Parasuraman, A., V. A. Zeithaml and L. L. Berry (1990) Delivering Quality Services,
Parasuraman, A., V. A. Zeithaml and L. L. Berry (1991) Servire qualità, Milan: McGraw-
Hill.
Parasuraman, A., V. A. Zeithaml and L. L. Berry (1993) ‘More on Improving Service
Quality Measurement’, Journal of Retailing 69: 140–7.
Parasuraman, A., V. A. Zeithaml and L. L. Berry (1994) ‘Reassessment of Expectations as
a Comparison Standard in Measuring Service Quality: Implication for Further
Research’, Journal of Marketing 58: 111–24.
Pastor, J. M., F. Pérez and J. Quesada (1997) ‘Efficiency Analysis in Banking
Firms: an International Comparison’, European Journal of Operational Research 98:
395–407.
Paul, J. (1992) ‘On the Efficiency of Stock Based Compensation’, Review of Financial
Studies 5, 3: 471–502.
Pavarani, E. (2002) Analisi finanziaria, Milan: McGraw-Hill.
Penman, S. and T. Sougiannis (1998) ‘A Comparison of Dividend, Cash Flow
and Earnings Approaches to Equity Valuation’, Contemporary Accounting Research
15: 343–83.
Peristiani, S. (1997) ‘Do Mergers Improve the Cross-Efficiency and Scale of US Banks?
Evidence from the 1980s’, Journal of Money, Credit and Banking 29: 326–37.
Peschiera, F. (1999) ‘La creazione di valore nel sistema bancario italiano. Applicazione
della metodologia EVA alle banche quotate italiane nel periodo 1995–98’, APB News
3: 93–104.
Peters, T. (1988) Thriving on Chaos, New York: Knopf.
Peterson, P. P. and D. R. Peterson (1996) Company Performance and Measures
of Value Added, Research Foundation of the Institute of Chartered Financial
Analysts.
Pilloff, S. J. (1996) ‘Performance Changes and Shareholder Wealth Creation Associated
with Mergers of Publicly Traded Banking Institutions’, Journal of Money, Credit, and
Banking 28: 294–310.
Pilloff, S. J. and A. M. Santomero (1998) ‘The Value Effect of Bank Mergers and
Acquisitions’, in Y. Amihud and G. Miller (eds), Bank Mergers and Acquisitions,
Boston: Kluwer.
Podinovski, V. V. and A. D. Athanassopoulos (1998) ‘Assessing the Relative Efficiency
of Decision Making Units using DEA Models with Weight Restrictions’, Journal of
the Operational Research Society 49: 500–8.
Porter, M. E. (1980) Competitive Strategy: Techniques for Analyzing Industries and
Competitors, New York: Free Press.
Preinreich, G. A. D. (1938) ‘Annual Survey of Economic Theory: the Theory of
Depreciation’, Econometrica 6: 219–41.
Prosperetti, L. and G. Durante (2004) ‘La Gestione delle risorse umane nei nuovi
percorsi di ristrutturazione delle grandi banche europee’, Bancaria, 1: 12–26.
Prowse, S. (1997) ‘Il sistema di gestione aziendale nel settore bancario: quel’è lo stato
attuale delle nostre conoscenze in merito?’, in Quaderni di moneta e Credito, special
issue, ‘Proprietà, controllo e governo delle banche’, March.
Quey-Jen Yeh (1996) ‘The Application of Data Envelopment Analysis in Conjunction
with Financial Ratios for Bank Performance Evaluation’, Journal of the Operational
Research Society 47: 980–8.
324 References
Rajan, M. V. (2000) ‘Discussion of EVA versus Earnings: Does it Matter Which is

More Highly Correlated with Stock Returns?’, Journal of Accounting Research 38
(supplement): 247–54.
Rappaport, A. (1986) Creating Shareholder Value, New York: Free Press.
Rees, B. (1995) Financial Analysis, New Jersey: Prentice-Hall.
Rees, W. P. (1997) ‘The Impact of Dividends, Debt and Investments on Valuation
Models’, Journal of Business Finance and Accounting 24: 1111–40.
Regalli, M. (2003) Stock option e incentivazioni del management, Milan: Edizioni Il Sole
24 Ore.
Reichheld, F. (1996) The Loyalty Effect: the Hidden Force Behind Growth, Profits and
Lasting Value, Boston: Harvard Business School Press.
Reichheld, F. F. and W. E. Sasser (1990) ‘Zero Defection: Quality Comes to Services’,
Harvard Business Review (September–October): 105–11.
Reichheld, F. F. and T. Teal (2001) The Loyalty Effect: the Hidden Force Behind Growth,
Profits, and Lasting Value, Boston: Harvard Business School Press.
Reilly, F. K. and K. C. Brown (2002) Investment Analysis and Portfolio Management
7th edn, Fort Worth: South-Western College Publishing.
Resti, A. (1997a) ‘Evaluating the Cost Efficiency of the Italian Banking System: What
Can be Learned from the Joint Application of Parametric and Non-parametric
Techniques’, Journal of Banking and Finance 21: 221–50.
Resti, A. (1997b) ‘How Should We Measure Bank Efficiency? A Comparison of Classic
and Recent Techniques Based on Simulated Data’, Università degli studi di Bergamo,
Quaderni del Dipartimento di Matematica, Statistica, Informatica ed Applicazioni,
no. 16.
Resti, A. (1997c) ‘Linear Programming and Econometric Methods for Banking Efficiency
Evaluation: an Empirical Comparison Based on a Panel of Italian Banks’, University of
Wales Bangor, Institute of European Finance (IEF) research paper, no. 3.
Resti, A. (1998) ‘Regulation Can Foster Mergers, Can Mergers Foster Efficiency?
The Italian Case’, Journal of Economics and Business 50: 157–69.
Resti, A. (1999) ‘La creazione di valore in banca: canali endogeni e canali esogeni’,
APB News 3: 19–38.
Resti, A. and L. Galbiati (2004) ‘Competitività e MandA: le aggregazioni bancarie
creano valore? Il punto di vista del mercato’, in G. Bracchi and D. Masciandaro (eds),
La competitività dell’industria bancaria, Rome: Edibank.
Revell, J. (1995) Deregulation and Bank Efficiency, University of Wales, Bangor,
Institute of European Finance (IEF), Research Monograph in Banking and Finance,
M95/1.
Richmond, J. (1974) ‘Estimating the Efficiency of Production’, International Economic
Review 15: 515–21.
Ross, S. A., R. W. Westerfield and J. Jaffe (2002) Corporate Finance, McGraw-Hill College
Division.
Rothschild, K. W. (1947) ‘Price Theory and Oligopoly’, Economic Journal 57: 299–320.
Ruozi, R. (ed.) (1986) Profitto e dimensioni nelle banche italiane: Quaderni del Banco di
Santo Spirito, Torino: UTET.
Ruozi, R. (2002) ‘La cultura aziendale e i processi di fusione e di acquisizione’, in
A. Carretta (ed.), Il governo del cambiamento culturale in banca, Rome: Bancaria
Editrice, pp. 72–90.
Rust, R. T., C. Moorman and P. R. Dickson (2002) ‘Getting Return on Quality:
Cost Reduction, Revenue Expansion, or Both?’, Journal of Marketing 66: 7–24.
Rust, R. T., A. J. Zahorik and T. L. Keiningham (1994) Return on Quality, Chicago: Irwin
Publishing.
References 325
Saita, F. (2000) Il risk management in banca: performance corrette per il rischio e allocazione
di capitale, Milan: EGEA.
Salvati, M. (1967) Una critica alle teorie dell’ impresa, Rome: Edizioni Ateneo.
Santorum, F. (2002) ‘Sviluppo dell’impresa e creazione di valore’, in E. Pavarani (ed.),
Analisi finanziaria, Milan: McGraw-Hill, pp. 343–84.
Sathye, M. (2001) ‘X-efficiency in Australian Banking: an Empirical Investigation’,
Saunders, A. (1997) Financial Institution Management, 2nd edn, Chicago: Irwin
Publishers.
Schaffnit, C., D. Rosen and J. C. Paradi (1997) ‘Best Practice Analysis of Bank Branches:
an Application of DEA in Large Canadian Banks’, European Journal of Operational
Research 98: 269–89.
Schmidt, P. and R. C. Sickles (1984) ‘Production Frontiers and Panel Data’, Journal of
Business and Economic Statistics 2: 367–74.
Schroeck, G. (2002) Risk Management and Value Creation in Financial Institutions,
London: John Wiley & Sons.
Schuster, L. (2000) Shareholder Value Management in Banks, London: Macmillan.
Schwizer, P. (1998) ‘I modelli di corporate governance negli intermediari finanziari ad
azionariato diffuso: Asimmetrie di ruolo degli azionisti ed efficacia del governo’,
in G. Airoldi and G. Forestieri (eds), Corporate Governance, Milan: ETAS Libri.
Seaver, B. L. and K. P. Triantis (1989) ‘The Implication of Using Messy Data to Estimate
Production-Frontier-Based Technical Efficiency Measures’, Journal of Business and
Economics Statistics 7: 51–9.
Seiford, L. M. (1996) ‘Data Envelopment Analysis: the Evolution of the State of the
Art (1978–1995)’, Journal of Productivity Analysis 7: 99–137.
Seiford, L. M. and R. M. Thrall (1990) ‘Recent Development in DEA: the Mathematical
Programming Approach to Frontier Analysis’, Journal of Econometrics 46: 7–38.
Seiford, L. M. and J. Zhu (1998) ‘Stability Regions for Maintaining Efficiency in Data
Envelopment Analysis’, European Journal of Operational Research 108: 127–39.
Seitz, W. D. (1996) ‘Efficiency Measures for Stream-Electric Generating Plants’,
Proceedings of the Thirty-Ninth Meeting of the Western Farm Economic Association,
pp. 143–51.
Sharpe, W. F. (1964) ‘Capital Asset Prices: a Theory of Market Equilibrium Under
Conditions of Risk’, Journal of Finance (September): 425–42.
Shaw, R. (2000) ‘Shareholder Value or Stakeholder Value? That is the Question’, in
L. Schuster (ed.), Shareholder Value Management in Banks, London: Macmillan,
pp. 2000: 36–52.
Shepard, W. G. (1953) Cost and Production Functions, Princeton: Princeton University
Press.
Shepard, W. G. (1970) Theory of Cost and Production Functions, Princeton: Princeton
University Press.
Shepard, W. G. (1986) ‘Tobin’s q and Structure Performance Relationship: Comment’,
American Economic Review 76: 1205–10.
Sherman, H. D. and F. Gold (1985) ‘Bank Branch Operating Efficiency: Evaluation with
Data Envelopment Analysis’, Journal of Banking and Finance 9: 297–315.
Sherman, H. D. and G. Lading (1995) ‘Managing Bank Productivity Using Data
Envelopment Analysis (DEA)’, Interfaces 25: 60–73.
Shleifer and Vishny (1996) ‘A Survey of Corporate Governance’, NBER working paper
no. 554, April.
Siems, T. F. (1996) Bank Mergers and Shareholder Wealth: Evidence from 1995 Megamerger
Deals, Federal Reserve Bank of Dallas, Financial Industry Studies.
326 References
Simar, L. (1992) ‘Estimating Efficiency from Frontier Models with Panel Data:
a Comparison and Semi-Parametric Methods With Bootstrapping’, Journal of
Simar, L. (1995) ‘Aspects of Statistical Analysis in DEA-type Frontier Models’, Université
Catholique de Louvain, Center for Operations Research and Econometrics (CORE),
discussion paper, no. 9561.
Simar, L. (1997) ‘Efficiency Analysis: the Statistical Approach’, Università degli studi di
Bergamo, Quaderni del Dipartimento di Matematica, Statistica ed Informatica ed
Applicazioni, no. 26.
Sinkey, J. F. (1998) Commercial Bank Financial Management, 5th edn, New Jersey:
Prentice Hall.
Sironi, A. (1995) Rischio e valore nelle banche, dai modelli di risk management alle politiche
di capital allocation, Milan: EGEA.
Sironi, A. (1999) ‘Estimating Banks’ Cost of Equity Capital: Evidence from an
International Comparison’, Quaderno Newfin, Università Commerciale, Milan:
Bocconi.
Sironi, A. (2004) ‘Estimating a Bank’s Cost of Equity Capital’, in G. De Lauretis
(ed.), Performance Measurements Frontiers in Banking and Finance, Milan: EGEA,
pp. 349–74.
Smirlock, M. (1985) ‘Evidence on the (Non)Relationship between Concentration and
Profitability in Banking’, Journal of Money, Credit and Banking 17: 69–83.
Smirlock, M., T. Galligan and W. Marshall (1984) ‘Tobin’s q and Structure-Performance
Relationship’, American Economic Review 76: 1050–60.
Smirlock, M., T. Galligan and W. Marshall (1986) ‘Tobin’s q and Structure-Performance
Relationship: Reply’, American Economic Review 76: 1211–13.
Spong, K., R. J. Sullivan and R. De Young (1995) ‘What Makes a Bank Efficient? A Look
at Financial Characteristics and Bank Management and Ownership Structure’, FRB of
Kansas City Review (December): 1–19.
Stafforf, M. R. (1996) ‘Demographic Discriminators of Service Quality in the Banking
Industry’, Journal of Service Marketing 10: 6–22.
Steigler, G. J. (1976) ‘The Xistence of X-efficiency’, American Economic Review 66,
1: 213–16.
Stern, J. M. and J. S. Shiely (2001) The EVA Revolution, London: John Wiley & Sons.
Stern Stewart (2004) http://www.sternstewart.com/content/performance/info.
Stewart, B. G. (1991) The Quest for Value, New York: HarperBusiness.
Stewart, B. G. (1996) ‘Roundtable Discussion of Current Issues in Commercial Banking’
(sponsored by the University of Florida and Stern Stewart and Co), Journal of Applied
Corporate Finance 9, 2: 24–51.
Stigler, G. J. (1947) ‘Professor Lester and the Marginalists’, American Economic Review
(March): 154–57.
Stovall, S. A. (1993) ‘Keeping Tabs on Customer Service’, Bank Marketing 25: 29–33.
Subrahmanyam, V., N. Rangan and S. Rosenstein (1997) ‘The Role of Outside Directors
in Bank Acquisitions’, Financial Management 26: 23–6.
Swamy, P. A. V. B., J. R. Barth, R. Y. Chou and J. S. Jahera (1995) ‘Determinants of
U.S. Commercial Bank Performance: Regulatory and Econometric Isssues’, FEDS
Paper, no. 29.
Taylor, W. M., R. G. Thompson, R. M. Thrall and P. S. Dharmapala (1997) ‘DEA/AR
Efficiency and Profitability of Mexican Banks: a Total Income Model’, European
Journal of Operational Research 98: 346–63.
Teas, R. K. (1993) ‘Expectations, Performance Evaluations, and Consumers’ Perception
of Quality’, Journal of Marketing 57: 18–34.
References 327
Teas, R. K. (1994) ‘Expectations as a Comparison Standard in Measuring Service

Quality: an Assessment and a Reassessment’, Journal of Marketing 58: 132–9.
Technical Assistance Research Programmes (2002). Available at www.tarp.com.
Tepper, G. C. (1994) ‘The Merits of Self-Testing’, Mortgage Banking 54: 76.
Thanassoulis, E. (1999) ‘Data Envelopment Analysis and its Use in Banking’, Interfaces
29: 1–13.
The Economist (1997) ‘A Star to Sail By?’, The Economist, 2 August, pp. 61–3.
The Economist (2001) ‘Marked by the Market’, The Economist, 29 November, pp. 31–2.
Thompson, R. G., E. Brinkmann, P. S. Dharmapala, M. D. Gonzalez-Lima and
R. M. Thrall (1997) ‘DEA/AR Profit Ratios and Sensitivity of 100 Large U.S. Banks’,
European Journal of Operational Research 98: 213–29.
Thompson, R. G., P. S. Dharmapala and R. M. Thrall (1993), ‘Importance for DEA of
Zeros in Data and Multipliers’, Journal of Productivity Analysis 4: 337–48.
Thompson, R. G., F. D. Singleton Jr., R. M. Thrall and R. M. Smith (1986) ‘Comparative
Site Evaluations for Locating a High-energy Physics Lab in Texas’, Interfaces 16: 35–49.
Timme, S. G. and W. K. Young (1991) ‘On Use of a Direct Measure of Efficiency on
Testing Structure Performance Relationships’, working paper, Georgia State
University.
Tinbergen, J. (1941) Econometrie, Gorinchem: J. Noorduijn en Zoon.
Toyne, M. F. (1998) ‘Interstate Bank Mergers and Their Impact on Shareholder Returns:
Evidence from the 1990s’, Quarterly Journal of Business and Economics 37: 48–58.
Tse, A. C. B. (2001) ‘How Much More are Consumers Willing to Pay for a Higher Level
of Service? A Preliminary Survey’ Journal of Service Marketing 15, 1: 11–17.
Tulkens, H. (1993) ‘On FDH Efficiency Analysis: Some Methodological Issues and
Application to Retail Banking, Courts and Urban Transit’, Journal of Productivity
Analysis 4: 183–210.
Tylecote, A. and S. Tarhan (2000) ‘Innovation in Banking: a Review from the Point of
View of Corporate Governance’, Sheffield University Research on Corporate
Governance and Product Innovation (COPI), Sheffield, UK.
Uyemura, D. G., G. C. Kantor and J. M. Pettit (1996) ‘EVA for Banks: Value Creation,
Risk Management and Profitability Measurement’, Journal of Applied Corporate Finance
9, 2: 94–105.
van der Wiele, T. and A. Brown (2002) ‘Quality Management over a Decade:
a Longitudinal Study’, International Journal of Quality and Reliability Management
19: 508–23.
van Iwaarden, J. T., van der Wiele, L. Ball and R. Millen (2003) ‘New Research:
Applying SERVQUAL to Websites: an Exploratory Study’, International Journal of
Quality and Reliability Management 20, 8: 919–35.
Veld, C. H. and Y. V. Veld-Merkoulova (2004) ‘Do Spin-offs Really Create Value? The
European Case’, Journal of Banking and Finance 28: 1111–35.
Velez-Pereja, I. (2000) ‘Value Creation and its Measurement: a Critical Look at EVA’,
unpublished paper, Universidad Javeriana, Bogotà, Colombia.
Venanzi, D. and B. Fidanza (2005) ‘Shareholder value o stakeholder value: qual’è
l’obiettivo di fondo delle decisioni di impresa’, Atti di ‘27° Congresso AIDEA – La
riconfiguzione dei processi decisional: nel quadro evolutivo della competizione,
Rome: Catania, AIDEA.
Verhoef, P. C., P. H. Franses and J. C. Hoekstra (1999) ‘The Impact of Satisfaction on
the Breadth of a Relationship with a Multi-service Provider’, RIBES Report 9955,
Erasmus University, Rotterdam.
Vesala, J. (1993) ‘Retail Banking in European Financial Integration’, Bank of Finland
Research Department, Finland.
328 References
Weaver, S. C. and J. F. Weston (2003) ‘A Unifying Theory of Value Based Management’,

Anderson Graduate School of Management and Finance, University of California, Los
Angeles, working paper 403.
Weissenrieder, F. (1997) ‘Value Based Management: Economic Value Added or Cash
Value Added?’, Department of Economics, Gothenburg University, study no. 3.
Wheeklock, D. C. and P. W. Wilson (1995) ‘Evaluating the Efficiency of Commercial
Banks: Does Our View of What Banks Do Matter?’, Federal Reserve Bank of St. Louis,
working paper, July.
Wheeklock, D. C. and P. W. Wilson (1999) ‘Technical Progress, Inefficiency and
Productivity Change in US Banking, 1984–1993’, Journal of Money, Credit and Banking
31: 213–34.
Wheeklock, D. C. and P. W. Wilson (2001) ‘Consolidation in U.S. Banking: Who
Are the Acquirers?’, Federal Reserve Bank of St. Louis, working paper, no. 3,
April.
Williams, J. (2001) ‘Financial Deregulation and Productivity Change in European
Banking’, Revue Bancarie et Financière 8: 470–7.
Williams, J. and J. Wilson (2000) ‘The Size and Growth of Banks: Empirical Evidence
from Four European Countries’, Applied Economics 32, 9: 1101–9.
Williams, R. T., van der Wiele and J. van Iwaarden (2004) ‘TQM: Why it Will Again
Become a Top Management Issue’, International Journal of Quality and Reliability
Management 21, 6: 603–11.
Williamson, O. E. (1963) ‘Managerial Discretion and Business Behaviour’, American
Economic Review 53: 1032–57.
Williamson, O. E. (1964) The Economics of Discretionary Behaviour: Managerial Objectives
in a Theory of the Firm, New Jersey: Prentice Hall.
Wilson, P. W. (1993) ‘Detecting Outliers in Deterministic Non-parametric Frontier
Models with Multiple Outputs’, Journal of Business and Economics Statistics 11: 319–23.
Wilson, P. W. (1995) ‘Detecting Influential Observations in Data Envelopment
Analysis’, Journal of Productivity Analysis 6: 27–45.
Woodside, A. G., J. N. Sheth and P. D. Bennet (eds) (1977) Consumer and Industrial
Buying Behaviour, New York and Rome: North Holland Publishing Company.
Yi, Y. (1990) ‘A Critical Evaluation of Customer Satisfaction’, in V. Zeithaml (ed.),
A Review of Marketing, Chicago: American Marketing Association.
Yildirim, H. S. and G. C. Philippatos (2002) ‘Efficiency of Banks: Recent Evidence from
the Transition Economies of Europe 1993–2000’, paper presented at the EFMA 2002
conference, London.
Yue, P. (1992) ‘Data Envelopment Analysis and Commercial Bank Performance: a
Primer with Application to Missouri Banks’, Federal Reserve Bank of St. Louis,
working paper, January–February, pp. 31–45.
Zanetti, G. (1980) Contributi per un’analisi economica dell’impresa, Naples: Liguori.
Zeithaml, V. A. (2000) ‘Service Quality, Profitability, and the Economic Worth of
Customers: What We Know and What We Need to Learn’, Journal of the Academy of
Marketing Science 28, 1: 67–85.
Zeithaml, V. A., L. L. Berry and A. Parasuraman (1993a) ‘The Behavioral Consequences
of Service Quality’, Journal of Marketing 60: 31–46.
Zeithaml, V. A., L. L. Berry and A. Parasuraman (1993b) ‘The Nature and Determinants
of Customer Expectations of Service’, Journal of the Academy of Marketing Science
21: 1–12.
Zeithaml, V. A., L. L. Berry and A. Parasuraman (1994) ‘The Nature and the
Determinants of Customer Expectations of Service’, Marketing Science Institute
Research Program Series, May, report no. 91–113.
References 329
Zeithaml, V. A., L. L. Berry and A. Parasuraman (1996) ‘The Behavioral Consequences

of Service Quality’, Journal of Marketing, 60, 31–46.
Zeithaml, V. A., A. Parasuraman and A. Malhotra (2000) ‘A Conceptual Framework for
Understanding e-Service Quality: Implications for Future Research and Managerial
Practice’, Marketing Science Institute, report 00-115.
Zenios, C. V., A. V. Zenios, K. Agathocleous and A. Soteriou (1999) ‘Benchmarks of the
Efficiency of Bank Branches’, Interfaces 29: 37–51.
Zimmer, S. A. and R. N. McCauley (1991) ‘Bank Cost of Capital and International
Competition’, Federal Reserve Bank of New York Quarterly Review 15: 35–9.
Zollo, M. and D. Leshchinkskii (2000) ‘Can Firms Learn to Acquire? Do Markets
Notice?’, Financial Institutions Center, Wharton School, Philadelphia, PA, working
paper. Available at http://fic.wharton.upenn.edu/fic/papers/00/0001.pdf.
Index
n.b. Page numbers in bold refer to figures or tables.
SHV = shareholder value
Aaker, D. A. 108–9 Almanij 79

AB Asesores 133 Alpha Bank 79
ABB (Belgium) 132 alternative profit efficiency 263, 271,
ABB Aros 133 277
Abbey National 78, 127, 132 see also profit efficiency
ABN (Netherlands) 132 altruism 26
ABN Amro Holdings 79, 133 Altunbas, Y. 116–18, 216–17
Accounting Horizons (journal) 85–6 Amalgamated Finance Ltd 177
accounting performance measures Ambroveneto 132
50–75, 52 Amel, D. 129, 130
Cash Flow Return on Investment AMEV + Mees Pierson 133
(CFROI) 57–9, 58 AMRO (Netherlands) 132
comparison among metrics 71–5, Anderson, E. W. 26, 109
71–2 Anglo-Saxon model 36
Discounted Cash Flow (DCF) 54–6, Annual Contribution Margin (ACM)
55 101–2
Economic Value Added (EVA) 61–70, Argentaria (BBVA) 132
67 Associazione Bancaria Italiana 10
Market Value Added (MVA) 60–1, 60 ATM transactions 30
Risk Adjusted (RAPM) 59–60, 59 Australia 118, 128, 130, 131
Shareholder Value Added (SVA) Austria 76, 78, 116, 128
56–7 ‘Available for Sale Securities’ (ASS) 153
see also performance measures
Accounting Review (journal) 85–8 B Sky B 33
Acheampong, Y. J. 90 BACOB (Belgium) 132–3
acquisitions, mergers and 129–44, Banca Agricola Mantovani SpA 185
130, 132–3, 140–2 Banca Carige SpA 186
Affinito, M. 31–2 Banca Commerciale Italiana SpA
AG (Fortis) 132 (COMIT) 35, 81, 185
AKB Privat – und Handelsbank 189 Banca CRT SpA (Banca Cassa di
Akbar, S. 94 Risparmio di Torino) 185
Akhavein, J. D. 222 Banca di Napoli 132
Al Ehrbar 54, 62, 67, 90–1, 147, 299n Banca Intesa SpA 77, 78, 81
Alarm, I. M. S. 118 Banca Monte dei Paschi di Siena SpA
Ali, A. I. 197 185
Alliance & Leicester 79 Banca Nazionale del Lavoro SpA (BNL)
Alliance & Leicester Group Treasury Plc 35, 79, 82
177 Banca Nazionale dell’Agricoltura SpA
Allianz 132 185
Allied Irish Banks 78 Banca Opi SpA 186
allocative efficiency 110, 195, 200, Banca popolare dell’Emilia Romagna
207–13 186
analysing determinants 258, 260, Banca Popolare di Bergamo – Credito
263, 268, 271, 273, 277 Varesino 185
330
Index 331
Banca Regionale Europea SpA 186 Barclays de Zoete Weld (BZW) 133
Banca Toscana SpA 186 Barings 133
Banco Ambrosiano Veneto SpA 185 Barker, R. 149
Banco Central 132 Basle Capital Accord 123–4
Banco de Bilbao 132 Battese, G. E. 117, 119, 195–6, 219, 282
Banco de Vizcaya 132 Bauer, P. W. 110
Banco di Brescia San Paolo Cab SpA – Baumol, W. J. 12
Banco di Brescia SpA 185 Bayerische Hypo-und Vereinsbank AG
Banco di Napoli 35 79
Banco di Roma 35, 79, 82, 132 Bayerische Hypobank (HBV) 132
Banco di Santo Spirito 132 Bayerische Vereinsbank AG 132
Banco di Sardegna SpA 185 BBL (Belgium) 133
Banco di Sicilia SpA (BdS) 185 BBV (Spain) 132
Banco Espirito Santo 79 BBVA (Spain) 79
Banco Hispano 132 BCH (Spain) 132
Banco Popolare di Brescia Scarl 185 BCI (Italy) 132
Banco Popolare di Verona e Novara BCP (Portugal) 79, 132
185 Beccalli, E. 119, 216–17
Banco Popular 78 Becher, D. A. 141, 143
Banco Santander 79, 132 behaviouralist approach 4, 10, 13
Banco Santander Central Hispano Beitel, P. 139, 141, 141–2, 144
(BSCH) 133 Belgium 78, 79, 115, 118, 128, 130,
Banerjee, A. 141, 143 131, 132
Banesto 132 Berenberg Bank (Joh. Berenberg, Gossler
Bank Austria-Creditanstalt 133 & Co.) 189
bank efficiency 112–15 Berger, A. N. 6–7, 196, 215–16, 222,
Bank for International Settlement 128 278, 280–1
Bank of Ireland 78 creating SHV and 110–11, 117–18,
bank mergers, domestic 132 120, 137, 139
bank performance see performance Berle, A. A. 11
measures Berliner Volksbank eG 189
Bank of Scotland 77, 79, 132, 177 BHF (Germany) 133
Bank van Roeselaere 132 Biddle, G. C. 89, 91, 147, 155–6
Bank of Wales Plc 178 BIL (Luxembourg) 133
bank-based model 36 Bipop (Capitalia) 132
Banker, R. D. 197 Bipop-Carire 80
Bankgesellschaft Berlin 80 BMW Bank GmbH 189
Bankhaus H. Aufhäuser 189 BNP Paribas 78, 132, 181
banking sector restructure 128 Bono, A. B. 18
Banking Supervision Committee 125 borrower satisfaction (BS) 261, 264,
Bankinter 80 269, 271, 273, 276–7
banks in Europe 31–2 borrowers 10
Bankscope 160, 162, 201, 203, 227, Bos, J. W. B. 223
245 Boston Consulting Group (BCG) 50,
Banque de Polynesie 181 91, 297n
Banque Nationale de Paris BPA (Portugal) 132
Intercontinentale (BNPI) 182 BPSM (Portugal) 132
Banque Paribas 181 Braxton Associates 50
Banque Paribas Pacifique 181 BRED Banque Populaire 181
Barclays Bank Plc 77, 78, 82, 123, 127, Brison 133
132, 177 Bristol & West International 178
332 Index
British Linen Bank Ltd 177 Champalimaud 133

Brown, A. 109 Charnes, A. 197–8, 200
Brown, S. 93 Charterhouse Securities 133
building society sector (UK) 117 Chase 133
business activities optimal mix 124–9, Cheltenham & Gloucester Plc 132,
125–6, 128 177
Business Value, Theory of 40–1 Christiania 133
Buzzell, R. D. 109 Citigroup 133
Cleland, A. S. 18
CAB & Banco San Paolo di Brescia Clydesdale Bank Plc 177
Combined 185 Coelli, T. 117, 119, 195–6, 198, 219,
Caisse régionale de Crédit Agricole 282
mutuel Alsace Vosges 182 Cofiri 35
Caja de Barcelona 132 Cole, D. W. 51
Canada 128, 130, 131 Commercial Bank of Greece 79
Capital Asset Pricing Model (CAPM) commercial banks 117, 124–6, 133,
40–1, 65, 152 245–52, 253
Capital at Risk (CAR) 59–60, 124 cost efficiency 201, 202–5, 207–13,
capital charge (CAPCHG) 156, 171 214, 216–17
capital invested (CI) 73–5, 149, 176, performance measures 163, 176–9,
179, 184, 188 181–2, 183–7, 184–6, 188–93
Carbo, S. 116, 119, 216–17 productivity changes 227, 228,
Cariplo (Intesa) 132 229–36, 236–7, 238–9
Cariverona Banca SpA 185 profit efficiency 219, 220–1, 222
Casden Banque Populaire 181 SHV efficiency 283, 284, 285–6, 287
Cash Flow, Discounted (DCF) 43, Commerzbank AG 79, 127
53–6, 55, 59, 71–4, 71–2, 95 Compagnie Bancaire 181
Cash Flow Return on Investment concepts 38–40, 278–81
(CFROI) 53, 57–9, 58 confirmation 26
Cash Flow to Equity (CFE) 55 consolidation, financial 129–44
Cash Flow to Firm (CFF) 55–6 Constant Returns to Scale (CRS) 197,
Cassa Depositi e Prestiti 35 225
Cassa di Risparmio della provincia di model 200
Viterbo SpA 186 consumer associations 10
Cassa di Risparmio di Biella e Vercelli Contemporary Accounting Research
(BIVERBANCA) 186 (journal) 85–6
Cassa di Risparmio di Padova e Rovigo contracting costs 295n
SpA 186 Coopcredito SPA 35
Cassa di Risparmio di Parma e Piacenza cooperative banks 245–51
SpA 185 cost efficiency 201, 202–5, 207–12,
Cassa di Risparmio di Roma 132 214
Cassamarca (Cassa di Risparmio della performance measures 163, 176,
Marca Trivigiana SPA) 186 179, 181–2, 183, 184–6, 187,
Casu, B. 119–20, 216, 219, 223, 237, 188–93
239, 240, 259 productivity changes 227, 228,
CCF (France) 133, 181 229–34, 236–7, 236, 238–9
CDC Ixis (France) 181 profit efficiency 219, 220–1, 222
CERA (Belgium) 132 SHV efficiency 283, 284, 285–6, 287
CGER (Belgium) 132 Cooperman, E. 141, 143
CGER/SNCI (Belgium) 133 Copeland, T. 42
Chaffai, M. E. 118 core business strategies 126
Index 333
Cornett, M. M. 141, 143 Credit Suisse 79, 132

corporate capital structure 296–7n Credit Suisse First Boston (CSFB) 133
cost of capital (CC) 61, 149 credit unions (UK) 119
cost efficiency 195–218 Credito Bergamasco 186
analysing determinants 263, 268, Credito Industriale Sardo 35
269–70, 271, 276–8, 277 Credito Italiano 35
components 273 Credito Romagnolo (Rolo) 132
estimating methodology 195–201, cross-customer communication 26
197 Cumulative Abnormal Returns (CuAR)
results 201–18, 204–5, 207–13 139, 143
sample description 201, 202–3 Customer Life Time Value (CLTV)
Cost of Equity 65, 95 101–2, 104
‘cost leadership’ 46 Customer Loss Rate (CLR) 102, 242
cost ‘X’-efficiency (X-EFF) 261, 266, Customer Retention Rate (CRR) 242
270–1, 271, 275, 277, 278, 282, 287 customer satisfaction 25–6, 25,
Coutts & Co 177 97–109, 266, 272
creating shareholder value (SHV) achieving 97–109
95–145, 96 definition of 7, 98–9
endogenous channels 97–129, 97 improving 103, 104–8, 105–6
business activities optimal mix performance and 108–9
124–9 SHV and 99–101, 100
customer satisfaction 97–109 value of customer 101–4
efficiency and productivity measuring 239–54, 241, 243, 245–52
improvement 109–20 customer value approach 17–18, 22–4
optimising financial structure customers 9
120–4 Cybert, R. M. 13
exogenous channels 129–44
mergers and acquisitions (M&A) Daimler-Benz (Germany) 36
performance 139–44 Dalborg, H. 27, 34
trends in 129–39 Damodaran, A. 56, 66–7, 73, 121, 152
Crediop 132–3 Danske Bank 78
Crédit Agricole Alpes Provence 182 Data Envelope Analysis (DEA) 6,
Crédit Agricole d’Ile-de-France 181 118–20, 223, 281, 283
Crédit Agricole du Finistère 181 analysing determinants 260–1, 263,
Crédit Agricole du Morbihan 182 267–72, 268, 271, 273, 276–8,
Crédit Agricole du Nord 181 277
Crédit Agricole du Nord Est 181 cost efficiency and 195, 197–8, 201,
Crédit Agricole du Pas-de-Calais 181 202–3, 206–18, 207–13
Crédit Agricole Indosuez 132 models 198, 200, 225–6
Crédit Agricole Loire Haute-Loire 182 Datastream 160, 162
Crédit Agricole Loire-Atlantique 182 Davis, D. 124
Crédit Agricole Pyrenees Gascogne 181 De Long (author) 141, 143
Crédit Communal 133 De Young, R. 117, 141, 143
Crédit du Nord 181 definitions
Crédit Industriel d’Alsace et de Lorraine business activities optimal mix 124–9
(Banque CIAL) 181 customer satisfaction 7, 98–9
Crédit Industriel et Commercial (CIC) Economic Value Added (EVA) 63
181 performance measures 147–54
Credit Italiano (UniCredito) 132 productivity and efficiency 109–10
Crédit Local 133 SHV 2–3
Crédit Lyonnais 79, 133 of stakeholders 15
334 Index
Den Danske Bank 132 Earnings Per Share (EPS) 42, 61–2,
Denmark 76, 78, 114–15, 116, 132 92–3
depositor satisfaction (DS) 10, 261, Easton, P. D. 93–4
264, 269, 271, 273, 276–7 ECB see European Central Bank (ECB)
deposits and loans 246–52 Economic Efficiency see ‘X’-efficiency
Dermine, J. 132–3 ‘economic management’ 22, 22
determinants of shareholder value economic objectives 9–37
(SHV) 256–88, 257 rationality debate 12–13
analysing 256–78 SHV approach 29–36
listed banks results 262–70, stakeholder models 13–29
263–4, 266–9 theory foundations 10–12
listed and non-listed banks results Economic Value Added (EVA) 4–5, 21,
270–8, 271, 273–7 258, 281
methodology 256–61, 258 literature 43, 50, 54, 61–84 passim,
sample description 261–2, 283 89–91
SHV efficiency 278–81 adjustments 67–70
measuring 281–8, 282 components 63–7
results 284–8, 285–6 definition 63
Deutsche Bank SpA AG 27, 76, 78, filter criteria 68
127, 133, 185, 187 spectrum 67
Dexia 78, 132–3 success 62–3
Di Antonio, M. 72, 241 performance measures 146–51,
Dietsch, M. 116, 216 156–9, 157, 165, 167
Dillon Read 133 SHV 171, 184, 188–90
Discounted Cash Flow (DCF) 43, 53–6, Economist, The 49–50
55, 59, 71–4, 71–2, 95 effectiveness 22, 22
Distribution Free Analysis (DFA) efficiency
116–17, 119, 222–3 change 225, 259–60
diversification 126 productivity and 109–20
DNB Holdings (Norway) 78 bank 111–20, 112–15
Downs, A. 12 defining 109–10
Dresdner Bank AG 77, 78, 132–3 scores 285–6
drivers SHV and 110–11, 278–88
actions and strategies 96 Efficient-Structure (ES) 136–7
measuring 194–255 EFG Eurobank Ergasias 79
cost efficiency 195–218 empathy 99
customer satisfaction 239–54, empirical studies 75–94
241, 243, 245 SHV in Europe 76–83, 77–80
information content 263–4, 271, value relevance literature 83–92,
277 85–8
productivity changes 223–39 methodological issues 92–4
profit efficiency 218–23 Enron (USA) 53
Drucker, P. F. 61 equity market value (MVE) 147
Durbin-Watson test 171, 262, 270 Erste Bank 78
dynamic efficiency studies 139, European Banking Briefing 127
140–1 European Central Bank (ECB) 31,
125–6, 125, 129
Earnings Before Extraordinary Item European Company performance survey
(EBEI) 89 (Financial Times) 75–7, 77–80, 81
Earnings Before Interest and Tax (EBIT) European continental model 36
51, 55, 64, 69–70 event studies 139, 140–1
Index 335
Fama, E. F. 40 General Electric 61

Fare, R. S. 119, 230 Generale Bank 133
Farrell, M. J. 110, 119, 197 Germany 3, 6–7
FCE Bank Plc 177 analysing determinants 261–2, 270,
Feltham, G. 76 275, 277, 278
FG (Spain) 133 cost efficiency 201–6, 202–5,
Fiat (Italy) 36 209–10, 214–17
financial structure 120–4 creating SHV 113–14, 116, 118–20,
Financial Times (FT), European 127, 128, 130, 132, 142
Company performance survey customer satisfaction 240, 244, 245,
75–7, 77–80, 81 248–9, 253
Findomestic Banca SpA 185 economic objectives 20, 30, 36
Finland 113, 115, 116, 127, 128, 132 literature review 76–7, 78–80
Fiordelisi, F. 105–6 performance measures 146, 152,
Firm value 73–5 160, 163
First Union Corporation 34 SHV 171–5, 172–4, 183, 187–90,
five forces model 47 188–90
Foreningssparbanken 79 productivity changes 226–8, 227,
Fornell, C. 109 231–2, 236–7, 238
Fortis 78, 132, 133 profit efficiency 219–23, 220–1
Fourier Flexible functional form SHV efficiency 282–7, 283, 285–6
116–17, 119 Girardone, C. 119, 216, 219, 223
France 3, 6–7, 76–7, 78–9 Girobank Plc 177
analysing determinants 261–2, 270, Glass, J. C. 117, 119, 216
275, 277 Glass-Steagall Act 299n
cost efficiency 201–8, 202–5, 207–8, ‘global value’ 48
214–16 Gontard & Metallbank AG 189
creating SHV 116, 118–20, 128, 130, ‘goodwill’ 297n
131, 132 Gordon, R. A. 11
customer satisfaction 240, 244, Gota Bank 132
245–7, 253 government 10
economic objectives 20, 30, 33, 36 Granville Bank Ltd 178
performance measures 146, 152–3, Greece 79–80, 113
160, 163 Greenspan, A. 34
SHV 171–5, 172–4, 179, 180–4, Gresham Trust Plc 177
181–3 Gross-Gerauer Volksbank EG 189
productivity changes 226–30, 227, Group of Ten 129, 131, 134–6
229–30, 236–7, 238 Groupe Banques Populaires 181
profit efficiency 219–23, 220–1 Groupe Caisse d’Epargne 181
SHV efficiency 278, 282–7, 283,
285–6 Habermayer, L. 20
Free Cash Flow to Equity (FCFE) 95 Halifax 79, 132
FTSE World index 76 Hall, R. L. 11
funds-suppliers 10 Hambros 133
funds-takers 10 Hamburger Sparkasse 189
Hamilton, R. 2
Gale, B. T. 109 Handelsbanken 117
Gardener, E. P. M. 217 Hannan, T. H. 137
Garvey, G. T. 90–1 Hanseatic Bank GmbH & Co 189
General Accepted Accounting Principles Hatch, Jim 34
(GAAP) 53, 61, 68 Hawawini, G. A. 141
336 Index
HBOS Treasury Services Plc 177 literature review 53, 76–7, 78–80,
HFC Bank Plc 177 81
Hitch, C. J. 11 performance measures 146, 152–3,
Hoare Govett 133 160, 163
HOLT Value Associates 297n SHV 171–5, 172–4, 180, 183,
Holthausen, R. W. 88 184–6
Hörther, S. 127 privatisation 35
Hotelling’s Lemma 196 productivity changes 226–8, 227,
Houston, J. F. 141, 143 233, 236–7, 239
HSBC Bank Plc 33, 78, 133, 177 profit efficiency 219, 220–1, 222
HSBC Holdings PLC 82–3 SHV efficiency 282–4, 283, 285–6
Humphrey, D. B. 111, 118–20, 196, Ittner, C. D. 100, 109
215–16, 222
Hypovereinsbank 133 Jacobson, R. 108–9
Japan 20, 112, 114, 116, 128, 130, 131
IMI 35, 132 Jensen, M. C. 141
inbound logistics 49 Johnston, R. 99
incentive system 20–1 Journal of Accounting, Auditing and
income, non-interest 125 Finance 85–6
incremental association studies 83–4 Journal of Accounting and Economics
independent variables 159, 159 85–8
India 114 Journal of Accounting Research 85–7
Indosuez Belgium 132 Journal of Business Finance and
industrial countries 130 Accounting 87
Information Technology (IT) 108 Journal of Financial Statement Analysis
ING Bank (Netherlands) 132–3 86, 88
ING BHF-BANK AG (Germany) 190 Journal of International Financial
Institut International d’Etudes Bancaire Management and Accounting 87
27, 34 Journal of Risk and Insurance 87
interest costs (IC) 148 J. P. Morgan 152
interest margin (IM) 146, 148, 156, 159 Jyske Bank 78
interest revenues (IR) 148
intermediation margin (IDM) 146, KB (Belgium) 132
148, 159, 302n KBC (Belgium) 78, 132
Internal Rate of Return (IRR) 53, 57 Kimbal, R. C. 51
International Standard Organisation Kleinwort Benson 133
(ISO) 18 Koopmans, T. C. 110, 198
internet banking 30, 33, 126, 296n KOP (Finland) 132
Intesa 132 Koska, M. T. 108
invested capital 271, 273–7 Kreissparkasse Biberach 189
investment firms (UK) 119
Ireland 33, 76, 78, 118 La Caixa 132
Italy 3, 6–7, 10, 30, 36 labour market 21
analysing determinants 261–2, 270, Larcker, D. F. 100, 109
275, 277, 278 Lee, K. 124
cost efficiency 201, 202–5, 206, Lehen, K. 89
211–12, 214–17 Lehman Brothers 33
creating SHV 113–14, 119–20, 128, LEK/Alcar 91
130, 132 Leshchinkskii, D. 141, 143
customer satisfaction 240, 244, 245, linear programming (LP) model
250–1, 253 198–200
Index 337
listed banks 216–17, 287 market-based model 36, 139

analysing determinants 263–4, marketing and sales 49
266–9, 270, 271, 273–7 Markowitz, H. M. 40
performance measures 146–7, 160, Marris, R. 12, 16, 21
161–4, 168, 171–91 Marshall, A. 2
literature review 38–94 MBNA Europe Bank Ltd 177
accounting performance measures Means, J. C. 11
50–75 measuring company performance,
concepts 38–40 techniques of 5
empirical studies 75–94 Medicredito Centrale 35
metrics 49–50 Medicredito Fondiaro Centro-Italia 35
theory 40–9 Medicredito Friuli Venezia Giulia 35
Lloyds Bank Plc 132, 177 Medicredito Lombardo 35
Lloyds TSB 79, 127 Medicredito Toscano 35
Lloyds TSB Group PLC 82–3 Meliorbanca 35
Lloyds TSB Scotland Plc 177 merchant banks 133
loans 246–52 mergers and acquisitions (M&A)
Loderer, C. 19 129–44, 130, 132–3, 140–2
Lovell, C. A. K. 109, 119, 223, 225 Merita Bank 132–3
Lozano-Vivas, A. 116, 216, 223 MeritaNordbanken 28–9
Luxembourg 118 Merrill Lynch 133
Messori, M. 35
Machlup, F. 12 Mester, L. J. 6–7, 118, 120, 196, 278,
McKillop, D. G. 117, 119, 216 280–1
McKinsey 50 methodology
Makhija, A. K. M. 89 cost efficiency 195–201, 197
Malmquist Total Factor Productivity (TFP) determinants of SHV 256–61, 257
index 117–18, 120, 215, 225–6, 237 Distribution Free Analysis (DFA) 117
MAM (merchant bank) 133 performance measures 147–60
management model 16–17, 20–1 relevance literature 92–4
managerial capitalist theory 4, 10–12 Total Factor Productivity (TFP)
managers 9 225–6, 226
March, J. G. 12–13 metrics 49–50
marginal information content studies metrics comparison 71–5
84 Milbourn, T. T. 90–1
marginalists 4, 10, 13 Miller, Merton H. 5, 40–1, 134
Market Value Added (MVA) 54, 60–1, Miller-Modigliani framework 42–5
60, 68, 95, 147 Miller, S. M. 223
empirical studies 77, 81–2, 89–91 MKG Bank GmbH 189
performance ranking (Stern Stewart) models
75 Anglo-Saxon (market-based) 36, 139
Market Value of Equity (MVE) 91, 156, Capital Asset Pricing (CAPM) 40–1,
158 65, 152
Market-Adjusted Returns (MAR) 5–6, Constant Returns to Scale (CRS) 200
92 Data Envelope Analysis (DEA) 198,
analysing determinants 257, 262, 200
263–4, 266–9, 270 European continental (bank-based)
performance measures 147, 155, 36
158, 160 five forces 47
information content 164, 165–71, linear programming (LP) 198–200,
166, 168–70 226, 228
338 Index
models – continued Northern Rock Plc 78

performance measures 154–9 Norway 76, 78, 112–13, 115, 128, 142
social responsibility 15–16, 19 Noulas, A. G. 223
stakeholder 13–29
technical efficiency effects 117 objectives see economic objectives
‘untouchable’ management 16–17, O’Byrne, S. F. 84, 89, 91, 147, 156
20–1 O’Connor (bank) 133
Variable Returns to Scale (VRS) 197, Off-Balance Sheet (OBS) items 196–7
200 Ohlson, J. A. 76
see also methodology; Stochastic operating cash flow (CFO) 89
Frontier Analysis (SFA) operating performance studies 139,
Modigliani, F. 5, 40–1 140–1
Miller-Modigliani framework 42–5 operational activities 49
Monsen, R. J. 12 Ordinary-Least-Squares (OLS) 91, 94,
Monte del Paschi di Siena 78 155
Monte Paschi Siena 81 outbound logistics 49
Morgan Grenfell 133 Overall Efficiency (OE) see ‘X’-efficiency
Morgan Stanley 123 overall mean profits 172
Morgan Stanley Dean Witter 133
Munari, L. 103 Padoa-Schioppa, T. 30
Panasonic 33
National Bank of Greece 79 parametric approaches 111
National Westminster Bank Plc Parasuraman, A. 98, 107
(NatWest) 132, 177 Paribas (France) 132–3
Nelson, E. 108 Paribas Belgium 132
neoclassical theory 4, 10–12 Parmalat (Italy) 53
net commission and fee income Paul, J. 90
(NetCFI) 156, 171 Performance 1000 database 90
Net Income (NI) 89, 146, 148, 158–9, performance measures 71–2, 146–98
165 customer satisfaction and 108–9
Net Operating Profit After Tax (NOPAT) information content 163–71, 164,
110 168–70
literature review and 51, 63–4, listed and non-listed banks 171–91
67–70, 73, 84, 89 methodology 147–60
performance measures and 148–54, definition of 147–54
156, 171 model 154–9
Net Present Value (NPV) 53–4, 58, sample description 160–3
74–5 see also accounting performance
Netherlands 79, 128, 130, 131, 132 measures
New Basle Capital Accord 124 performance studies 139, 140–1
NIBID (Greece) 80 Peterson D. R. 84, 91, 147
Nissan Bank GmbH 190 Peterson P. P. 84, 91, 147
NMB (Netherlands) 132 Philippatos, G. C. 223
non-listed banks 270, 271, 273–7, Pilloff, S. J. 139, 141, 143
287–8 Piraeus Bank 80
performance measures 146–7, 160, Porter, M. E. 5, 45
161–4, 171–91 Porter framework 45–9, 47–8
non-parametric approaches 111 Portugal 79, 116, 132
Nordbanken 123, 132–3 Postbank 132
Nordea 78, 127 Price Efficiency 298n
Northern Bank Limited 177 privatisation 35
Index 339
productive efficiency 22, 22 performance measures and 146,

productivity 109–20 148, 158–9, 165, 167
changes 223–39, 224 SHV 175, 176, 179, 183, 184, 187,
results 228–39, 229–36 190
sample description 226–8, 227 Return on Invested Capital (ROIC) 149
Total Factor Productivity (TFP) Return on Investments (ROI) 50–1
methodology 225–6, 226 Return on Risk Adjusted Capital
profit efficiency 261, 266, 267, 278, (RORAC) 60
282, 287 Return on Sales (ROS) 89
estimating methodology 218–19, Review of Quantitative Finance and
219 Accounting (journal) 87
results 219–23, 220–1 Rheingauer Volksbank eG 190
sample description 219 Risk Adjusted Performance Measures
prosumers 295n (RAPM) 59–60, 59
PSA Finance Deutschland 189 Risk Adjusted Return on Capital
pure technical efficiency 225 (RAROC) 59–60
change 229–36, 240, 258–60, 264, Risk Adjusted Return on Risk Adjusted
267, 271, 274, 277 Capital (RARORAC) 60
Risk Free Rate (RFR) 65–6
Raiffeisen – Volksbank im Landkreis Risk Premium 65
Altoetting eG 189 Robert Fleming 133
Rappaport, A. 19, 22, 41, 53, 56–7, 67 Rolo Banca 76, 78
RAS (Review of Accounting Studies) 86–7 Rolo Banca 1473 S. P. A. 185
rationality debate 12–13 Rothschild, K. W. 12
Raw Market returns (RAWR) 147, 158 Royal Bank of Scotland Plc 77, 78,
RCI Banque (France) 181 132, 177
RealDanmark 132 Ruback, R. S. 141
regression 92–3, 262, 266–9, 268–70, Ryngaert, M. 141, 143
272–3, 273–6, 275
performance measures and 146, St. Wendeler Volksbank eG 189
158, 165, 167, 168–9, 171 San Paolo 132
Reichheld, F. F. 25, 103 San Paolo di Torino 35
relative association studies 83 San Paolo IMI 79, 81, 132
Relative-Market-Power (RMP) 136–7 Santomero, A. M. 139
reliability 98 Santorum, F. 150
Reliance Bank Ltd 178 Sasser, W. E. 25
research and development (R&D) 150 Sathye, M. 118
Residual Income (RI) 89, 146, 148, savings banks 245–51
159, 165 cost efficiency 201, 202–5, 207–12,
responsiveness 99 214, 216–17
Resti, A. 38, 124, 216–17 performance measures 163, 176–7,
restructuring of banking sector 128 179, 181, 183–8, 184–6, 188–9
Return on Assets (ROA) 50–3, 89, 119, productivity changes 227, 228,
139, 262 229–34, 236–7, 236, 238–9
performance measures and 146, profit efficiency 219, 220–1, 222
165, 175, 180, 183, 187 SHV efficiency 283, 284, 285–6, 287
methodology 148, 158–9 scale efficiency 136–7, 200, 207–13, 225
Return on Capital (ROC) 61, 64, 69 analysing determinants 258, 260,
Return on Equity (ROE) 40, 50–3, 89, 263, 268, 271, 273, 276–8, 277
139, 262 change 229–36, 240, 258–60, 264,
decomposition 52 267, 271, 275, 277
340 Index
Schiereck, D. 139, 141–2, 144 Stern Stewart & Company 50, 61–2,
Schmiedel, H. 223 68, 70, 77, 90–1, 150
Schroder 133 MVA performance ranking 75–6, 81–2
Schuster, L. 14 Stochastic Frontier Analysis (SFA) 6,
Scottish Provident 127, 132 116–17, 119, 218–19, 223
SE Banken (Sweden) 117 analysing determinants 263, 266,
Seiford, L. M. 197 271, 272, 276, 277
Seow, G. 155 cost efficiency and 195, 201, 202–5,
service value chain 4–5, 28 217
Shareholder Value Added (SVA) 53, SHV efficiency 281–4, 283
56–7 stock ownership 20
shareholders 9 strategies 96
Sharpe, W. F. 40 structural forces 46, 47
Shephard’s Lemma 196 Structure-Conduct-Performance (SCP),
Simon, H. A. 12 Traditional 136–7
Sinkey, J. F. 52 study of shareholder value (SHV) 1–8
Sironi, A. 151 motivation and key questions 2–3
Skandinaviska Enskilda Banken 79 Sullivan, M. W. 109
Smith New Court 133 supervisory authorities 10
SNCI ( Belgium) 132 suppliers 10
social responsibility model 15–16, 19 support 49
Société Générale 77, 78, 133, 181 Svenska Handelsbanken 78
Société Générale Calédonienne de Swary, I. 141
Banque (SGCB) 181 Sweden 33, 36, 76, 78–9
Société Marseillaise de Crédit 182 creating SHV 115, 116–17, 128, 130,
Sommers, G. A. 93–4 132
Spain 30, 78–80, 223 Swiss Bank Corp (SBC) 132–3
creating SHV 112–15, 116, 118–20, Switzerland 78, 114, 128, 130, 131, 132
128, 130, 132
Sparkasse Aachen 189 takeover 21
Sparkasse Krefeld 189 tangibles 98
Stadtsparkasse Düsseldorf 189 Tarhan, S. 33, 33
Stafforf, M. R. 99 Technical Assistance Research Program
stakeholder models 13–29 (TARP) 99
customer value approach 17–18, 22 technical change 225, 236
limits of 22–4 technical efficiency 110, 195, 207–13
framework 14–15, 14 analysing determinants 258, 263,
limits of 19 268, 271, 273, 277
SHV approach 18–19, 25, 28, 29–36 change 229–36, 240, 258, 260, 264,
non-zero sum game 24–9 271, 277
suitability 34–6, 35 technical efficiency effects model 117
trends 29–34, 31–3 technological change 229–35, 240
social responsibility model 15–16 analysing determinants 258, 264,
limits of 19 267, 271, 274, 277
‘untouchable’ management model telephone banking 30
16–17 theory 4–5
limits of 20–1 evolution of SHV 40–9
‘stakeholder symbiosis’ 4, 27, 34 Miller-Modigliani framework 42–5
stakeholders 9–10 Porter framework 45–9, 47–8
Standard Chartered Bank 79, 177 neoclassical v managerial capitalist
Stark, A. W. 94 10–12
Index 341
Theory of Business Value 40–1 United States of America (US) 2–3, 53,
Total Equity (TE) 148 89
Total Factor Productivity (TFP) 6–7, creating SHV 99, 112–15, 117–18,
119–20, 215, 228, 258, 278 128, 130, 131–43 passim, 142
change 229–35, 236–9, 240 economic objectives 20, 29–30, 36
analysing determinants 258–60, ‘untouchable’ management model
262, 264, 267, 269–75, 271, 16–17, 20–1
277 Uyemura, D. G. 53, 67–8, 89, 91, 147,
information content 267, 274 287
cumulative estimates 238–9
methodology 225–6, 226–7, valuation ratio 12, 16–17
229–36 Value at risk of competition (VARC)
Total Quality Management (TQM) 23, 101–2
105 Value at Risk (VAR) 59
Total Shareholder Return (TSR) 76–7, ‘value chain’ 48–9, 48
78–80 value relevance literature 83–92, 85–8
trade unions 10 methodological issues 92–4
Traditional Structure-Conduct- van der Weile, T. 109
Performance (SCP) 136–7 Variable Return to Scale 225, 260
training 150 Variable Returns to Scale (VRS) model
transaction costs 33 197, 200
translog functional form 116 Velez-Pareja, I. 149–50
TrygBaltica 132 Vereins-und Westbank AG 189, 190
TSB Bank Plc 132, 177 Verhoef, P. C. 109
Tunisia 114 Volksbank 132
Turkey 112 Volksbank Franken eG 189
TV banking 33 Volksbank Freiburg eG 189
Tylecote, A. 33, 33 Volksbank Hochrhein eG 190
Volksbank Offenburg eG 189
UBS (Swiss) 78, 132 Volksbank Raiffeisenbank Murr-Lauter
Unibank 132–3 eG 190
UniCredit Banca 185 Volksbank Speyer – Neustadt –
Unicredito Italiano 77, 78, 81 Hockenheim 190
Unidanmark 133 Volksbank in Stuttgart AG 189
Union Bank of Finland 132 VR Bank Suedpfalz eG 189
United Kingdom (UK) 3, 6–7, 282–4,
283, 285–6 Wagon Finance Ltd 177
analysing determinants 261–2, 270, Wallenbergs (Sweden) 36
275, 277, 278 Walrasian system 11
creating SHV 114, 117–20, 128, 130, Warburg 133
131, 132 Watts, R. L. 88
economic objectives 20, 30, 33, 36 web-based services 296n
literature review 76–7, 78–9, 81–2, Weighted Average Cost of Capital
94 (WACC) 55–6, 59, 65, 120–2,
measuring drivers 201–28 passim, 150–1
202–5, 213, 220–1, 227, 234–5 weighted least square regression (WLS)
customer satisfaction 239–40, 94
244, 245, 252, 253 Westfalenbank AG 190
performance measures 146, 152–3, Wetzstein, M. E. 90
160, 163 Williams, J. 217
SHV 171, 172–4, 175, 176–8, 180 Williamson, O. E. 12
342 Index
Winterthur (Switzerland) 132 measuring drivers 195, 201, 204–9,

Woolwich Guernsey 178 206, 214–16, 219, 222
Woolwich Plc 132, 177 ‘X’-inefficiency 216–17
workforce 33–4
Yildirim, H. S. 223
‘X’-efficiency 261, 267 Yorkshire Bank Plc 177
cost (X-EFF) 261, 266, 270–1, 271,
275, 277, 278, 282, 287 Zeithaml, V. A. 108–9
creating SHV 110, 116–19, 127, 134, Zgranggen, P. 19
136–8 Zollo, M. 141, 143

(Palgrave Macmillan Studies in Banking and Financial Institutions) Franco Fiordelisi, Philip Molyneux (Auth.) - Shareholder Value in Banking-Palgrave Macmillan UK (2006) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Palgrave Macmillan Studies in Banking and Financial Institutions) Franco Fiordelisi, Philip Molyneux (Auth.) - Shareholder Value in Banking-Palgrave Macmillan UK (2006) PDF

Uploaded by

Copyright:

Available Formats

Shareholder Value in Banking

Palgrave Macmillan Studies in Banking and Financial Institutions

Palgrave Macmillan Studies in Banking and Financial Institutions

Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire

To Del and my cheeky kids – Catrin, Lois,

Measuring shareholder value in Europe 171

2.1 Number of banks in European banking systems between 32

7.9 Incremental information content of cost and profit efficiency 275

2.1 A possible implementation of the stakeholder approach 14

5.6 Return on equity and EVA on capital invested in German 188

Franco Fiordelisi (MA, PhD) is Lecturer in Banking and Finance at the

Philip Molyneux is Professor in Banking and Finance and Director of the

Managing to create sustainable shareholder value is currently recognised by

1. Why a book on ‘shareholder value’?

contributions attempt to empirically assess the importance of different drivers

Motivation and key questions

• What is shareholder value (SHV) and how should it be accurately

shareholder value can be measured in a relatively straightforward manner

• What is the level of shareholder value created in European banking?

In order to answer this question, this book presents an empirical assessment of

Finding a strong relationship between shareholder value and its drivers

• Is it possible to develop new measures of efficiency? What is their role in

Banks’ efficiency is usually measured focusing on production, costs and

Structure of the text

specifics of banking (EVAbkg) outperforms all other performance measures since

index using DEA for estimating distance functions. We decompose TFP

• Shareholders who are interested in maximising the value of their

(i) Funds-suppliers, such as depositors, who wish to receive high interest

Although this list of stakeholders may be further extended, it is sufficient to

Economic foundations of the theory of the ﬁrm

Neoclassical vs. the managerial capitalist view

(1) Firms are managed by two groups: managers and owners.

Although the managerial capitalist views differ substantially on a wide

• Gordon (1945) identified several groups with an active role in the

Debate about rationality

ticised: when market competition is not perfect (perfect competition is a too

stakeholders’: as a consequence, the stakeholder approach is a subjective

A basic stakeholder framework

Figure 2.1: A possible implementation of the stakeholder approach

Threats and opportunities

Source: adapted from Schuster (2000: 5).

These approaches appear to be complementary, rather than alternatives, and

• Definition of the group of stakeholders. In this phase, it is necessary to identify

As a result, the stakeholder approach requires firms to conduct periodical

Social responsibility model

value maximisation. The importance of banks’ social responsibility has

The ‘untouchable’ management model

financed by external sources (a more realistic assumption), there is a trade-off

The ‘untouchable management’ model has recently received increasing public

Customer value approach

unprofitable if it fails to meet customer expectations of its products and

Limits of stakeholder models

Limits of the social responsibility model

Fortunately, several market incentives make shareholder-value maximisa-

Limits of the management model

(1) Stock ownership by management leads executives’ interest towards share-

Limits of the customer value approach

Figure 2.2: The relationship between economic management–efficiency–effectiveness

The shareholder value approach: a non zero-sum game

• reducing future marketing and transaction costs for

Figure 2.3: Shareholder value maximisation, customer satisfaction, customer loyalty

commercial offers from other competitors. In this sense, customer satisfac-

Figure 2.4: The service value chain