Description
IIU Press and Research Centre Business Intelligence Journal
Business Intelligence Journal
Business Intelligence Journal
July, 2011 Vol.4 No.2
Volume 4 - Number 2 - July 2011 - Semiannual Publicaton
The Business Intelligence Journal (BIJ) is published by the Business Intelligence Service of London, UK (BIS) in collabo-
raton with the European Business School (Cambridge, UK) and the Business Management and Economics Depart-
ment at the School of Doctoral Studies of the European Union (Brussels, Belgium), as semiannual open access content
publicaton.
Editorial Note
202
Words from the Board of Editors 203
Profle of Authors Included in this Number
204
Information for Contributors
206
Articles Banking Proftability Determinants
209
James W. Scott, José Carlos Arias
The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
231
Walter Gerard Amedzro St-Hilaire
E-Banking Patronage in Nigeria: An Exploratory Study of Gender Difference
243
Asikhia, Olalekan
Foreign Direct Investment and its Effects on the Nigerian Economy
253
Omankhanlen Alex Ehimare
Artifcial Learning and Support Vector Machines: Default Risk Prediction
262
Fabio Franch
Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
273
Carlos Acosta-Calzado
Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
293
Monireh Panahi, Mohammad Pakeniat
Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
303
Akinyele, S.T.
Intuitive Managerial Decision Making in Malaysia and the United States
312
Isola Oluwabusuyi
Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants
in Nigeria
320
Olu Ojo, I. A. Irefn
Relationship between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
327
Nadia Sajjad Hafza, Syed Sohaib Shah, Khalid Zaman, Humera Jamsheed
Comparing Bankruptcy Prediction Models in Iran
335
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
Continued on back cover
© Copyright 2011: IIU Press and Research Centre A.C.
This is an open access content publication recognized by the Scholarly Publishing and Academic Resources Coalition
of the European Union (SPARC Europe) and distributed worldwide by the Directory of Open Access Journals
(DOAJ) of the Lund University Libraries (Lund, Sweden). Printed in Canada.
ISBN: 978-1-4251-8179-6
ISSN: 1918-2325
Vol. 4 No. 2
July 2011
Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors 343
Continued From Front
Cover
Adebisi Sunday A. , Babatunde Bayode O.
The Practicability of Activity-Based Costing (ABC) in the Nigerian Retail Banks 351
Ahmed Audu Maiyaki
Social Support as Mental Health Improver for Managerial Women in the Organizational Work
Environment
355
Jocelyn Sackey, Mohammed-Aminu Sanda
Case Studies Three Case Studies On Bank Nagari Padang (Indonesia) Headquarters And Main Branch 362
Heryanto
Organizational Communication as an Important Factor of Company Success: Case Study of Bosnia And
Herzegovina
390
Kenan Spaho
Short
Communications
A Short Communication on - How a Leading Power Distribution Company Effectively Tracks business
areas like Safety, Finance and Operation for Region and Business wise for evaluating their KPI’s - Using
BusinessObjects Xcelsius Dashboards.
394
Rakesh Tej Kumar Kalahasthi, Benson Hilbert
Modeling Strategies for Financial Hedging 402
José Carlos Arias
Announcements 405
2011 Business Intelligence Journal 202
Business Intelligence Journal - July, 2011 Vol.4 No.2
SPARCEurope
DIRECTORY OF
JOURNALS
OPEN ACCESS
Business Intelligence Journal by Business Intelligence Service is licensed under a Creative Commons Attribution 2.0 UK: England & Wales License.
Further tips for using the supplied HTML and RDF are here:http://creativecommons.org/learn/technology/usingmarkup
In collaboration with the European Business School, Cambridge, UK and the School
of Doctoral Studies of the European Union based in Brussels, Belgium, the Business
Intelligence Service of London, UK (BIS) publishes the Business Intelligence Journal
(BIJ) as a semi-annual scientific and academic open access journal, which includes research analysis and inquiry into issues of importance to
the international business community. Articles within the BIJ examine emerging trends and concerns, among others, in the areas of general
management, business law, public responsibility and ethics, marketing theory and applications, business finance and investment, general business
research, business and economics education, production/operations management, organizational behavior and theory, strategic management
policy, management organization, statistics and econometrics, personnel development and industrial relations, technology and innovation, case
studies and management information systems. The goal of the BIJ is to broaden the knowledge of business professionals and academicians by
including valuable insight to business-related information, by fostering cutting-edge research work and by providing open access to original
research, information and ideas contained within the journal’s pages. All articles in the BIJ are peered reviewed; the BIJ has been granted the Seal
of the Scholarly Publishing and Academic Resources Coalition of the European Union (SPARC Europe). The BIJ is an open access publication
distributed worldwide by the Directory of Open Access Journal (DOAJ) of the Lund University Libraries, Sweden for the Business Intelligence
Research Centre of London.
EDITORIAL NOTE
European Business School,
School of Graduate Studies
Cambridge, UK.
European Business School,
Singapore.
Brussels, Belgium
The Oxford Association of
Management
Ibkan Consultants
Investment Bankers
School of Doctoral Studies
(European Union) Isles Internationale
Université Brussels, Belgium.
Business Intelligence Service
London, UK
Lund University Libraries Sweden.
Seal of the Scholarly Publishing
and Academic Resources Coalition
(Granted to the Business Intelligence
Journal on the 20
th
day of August, 2008)
Editorial Board Editorial Department
Reviewers Coordinators
Editorial Design
Publisher
Associate Editors:
Dr. Katherinna Kurlenko
European Union Analogue Standards Certification
Committee, Brussels, Belgium
Prof. José Carlos Arias
Business Intelligence Research Centre of London, UK
Dr. Jünger Albinger
School of Doctoral Studies of the European Union,
Brussels, Belgium
Prof. Eric Palme
European Business School, Cambridge, UK
Dr. Zeng Lisheng
European Business School, Singapore
Dr. Mathew Bock
The Cambridge Association of Managers, Cambridge, UK
Dr. Henk Haegeman
The Oxford Association of Management, Oxford, UK
Michael Summers
Susan G. Boots
Martin A. Miller
Kenneth C. Michaels
Anita Peters
Roger Puig
Robert Miller
Pablo Gamez Olivo
Eric Veloz
IIU Press and Research Centre, A.C.
ISSN 1918 2325http://www.saycocorporativo.com/saycouk/BIJ/journals.
html
The Business Intelligence Journal is an Open Access journal
Printed in Canada
© Copyrights: IIU Press and Research Centre, A.C.
The Business Intelligence Journal (BIJ)
Published by the Business Intelligence Service, London, UK (BIS) for the Business Intelligence Research Centre
145 St. John Street, 2nd Floor,London EC1V 4PY, United Kingdom
[email protected]
Editor: Dr. Anne Surrey
Managing Editor: Dr. Paul Stenzel
Accounting and Finance: Prof. Ira Joubert
Human Resources: Prof. Beverly Lanting
Marketing: Prof. Frans Cooper
Operations and Production: Prof. Guy Le Roy
Information and Knowledge Management: Prof. Mitsuaki Uno
Leadership and Corporate Policy: Prof. Tui Unterthiner
Change, Conflict and Crisis: Prof. Takako Iwago
Mathematical and Quantitative Methods: Prof. Maria Nicklen
Microeconomics: Prof. Carlo Grunewald
Macroeconomics and Monetary Economics: Prof. Stephen L. Freinberg
International Economics: Prof. Mitchell Alvarez
Economic Thought and Methodology: Prof. Vincent Haidinger
Business Intelligence Journal
2011 Business Intelligence Journal 203
Business Intelligence Journal - July, 2011 Vol.4 No.2
The Business Intelligence Journal has come a long way since its frst issue appearance on July 2008 as it reaches the last
number of its fourth Volume with this July 2011 edition.
Because of obvious needs of higher quality and low budget, at the beginning the Business Intelligence Journal’s collaboration
were admitted basically when authors’ origin was the Business Intelligence Centre of London, the European Business School
of Cambridge or the School of Doctoral Studies of the European Union in Brussels, Belgium.
Though this certainly ensured our publication’s quality of content at a very reasonable cost on reviewing process, it also
denied the opportunity to authors emerging from many other sources around the world, which research and content quality
was at least as good as that of our originally chosen authors’ work.
With this number, the Business Intelligence Journal closes its 4th year of publishing, having received, since 2007 over 782
articles submitted from authors with intention to become admitted. Along these years, seven numbers in four volumes have
been published, containing only 60 admitted of the 782 research articles, case studies and book reviews, presented by 123
authors and coauthors from 27 different countries and fve continents.
Along these years we have also increased our printing copies production from 800 to 4,200, meant to grow physical
distribution to suit demand at 386 Universities and 108 Libraries in Europe, Asia, Africa, Oceania and the Americas.
All research articles published during this four years period in the Business Intelligence Journal derive from original
research and have been certifed at the highest quality rank for cutting edge research by the European Union Analogue
Standards Certifcation Committee (EUASC) in Brussels, being granted the Golden Seal at this certifcation process; has
also been qualifed as highest quality of scientifc and academic content by the Directory of Open Access Journals (DOAJ)
in the Lund University Libraries of Sweden, who also works as exclusive media distribution tool for its content worldwide;
and has been granted Highest Quality of Open Access Scholarly information by The Scholarly Publishing and Academic
Coalition of Europe (SPARC Europe), at the Bodleian Library in Oxford, UK, granting our Journal with the SPARC Europe
SEAL.
In time, our format has been slightly modifying to gain space and to suit our authors, reviewers and readers needs better. Its
appearance will also experience mandatory changes as we reach our ffth volume on 2012 in order to better adapt it towards
new market tendencies.
So, may this communication be useful to properly close our Business Intelligence Journal frst development stage and to
prevent our readers, distributors, reviewers and authors on its new look starting 2012, without disregard of our traditional
and strict policy of internal and peer reviewing process, also demanding of highest quality cutting edge scientifc and
academic content.
Most cordially,
The Editorial Board
Words from the Board of Editors
Business Intelligence Journal
Profile of authors included in this number
204 Business Intelligence Journal July
Business Intelligence Journal - July, 2011 Vol.4 No.2
Article 1: Banking Proftability Determinants
Author: 1 - James W. Scott – (PhD) Researcher at the Business Intelligence Research Centre of hosted at the Business Intelligence
Service, London, UK.
email: [email protected]
2 - José Carlos Arias – (PhD) In Science by the school Robert de Sorbon, Paris, France, Professor of Management Science at
the European Business School, Cambridge, UK.
email: [email protected]
Article 2: The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Author: Walter Gerard Amedzro St-Hilaire – (PhD) Researcher on Strategy and Governance of Public Organizations in HEC
Montréal, Researcher for the Research Center on the Governance of Natural Resources.
email: [email protected]
Article 3: E-Banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Author: Asikhia, Olalekan – Senior Lecturer, Department of Business Studies, College of Business and Social Sciences, Covenant
University, Ota, Canaan land.
email: [email protected]
Article 4: Foreign Direct Investment and its Effects on the Nigerian Economy
Author: Omankhanlen Alex Ehimare – Lecturer at the Department of Banking and Finance, Covenant University, Ota, Ogun State,
Nigeria.
email: [email protected]
Article 5: Artifcial Learning and Support Vector Machines: Default Risk Prediction
Author: Fabio Franch – (PhD) from West Virginia University and studied at University of Trento, where he successfully completed his
studies (BA and MA) in 2007. He was also an exchange student at the Abo Akademi in Turku, Finland. His interests are machine
learning and the dynamic analysis of economic phenomena.
e-mail: [email protected]
Article 6: Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Author: Carlos Acosta Calzado – (MBA) BS in Industrial Engineer from ITESM, MBA with focus on Finance and Economics from
NYU Stern. He is currently partner at Ibkan Consultants, Investment Banking.
e-mail: [email protected]
Article 7: Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Author: 1 - Monireh Panahi – (PhD) candidate in Strategic Management in Shahid Beheshti University. Marketing consultant; author
of articles on marketing and strategic management in Iranian journals. School of Management & Accounting, Shahid Beheshti
University, Daneshjoo Blvd., Evin,
email: [email protected]
2 - Mohammad Pakeniat – Researcher in SRRC, Sharif University of Technology. MBA from School of Management &
Economics, Sharif University of Technology. Business consultant and researcher in the felds of strategic management in
developing countries and service marketing. Shahid Rezaii Research Center, Sharif University of Technology, Azadi Avenue,
Tehran, Iran
email: [email protected]
Article 8: Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Author: Akinyele, S.T. – (PhD) School of Business, Covenant University,Ota-Nigeria.
email: [email protected]
Article 9: Intuitive Managerial Decision Making in Malaysia and the United States
Author: Isola Oluwabusuyi – (MBA, MA, DBA) School of Business and Accounting Brown Mackie College, Atlanta.
email: [email protected]
2011 Business Intelligence Journal 205
Business Intelligence Journal - July, 2011 Vol.4 No.2
In order to make contact with any of the Authors referred to above, please forward your request to: [email protected],
including BIJ’s edition (BIJ Volume 4, Number 2, July 2011), article’s and author’s names with your requirement. BIJ’s Editor will be
glad to submit your requests or inquiries before authors.
Article 10: Franchising And Organizational Performance: Empirical Investigation Of Selected Fast Food Restaurants In Nigeria
Author: 1 - Olu Ojo – Department Of Business Administration College of Management and Social Sciences Osun State University P.
M. B. 2008, Okuku Osun State, Nigeria
email: [email protected]
2 - I. A. Irefn – Technology Planning and Development Unit Obafemi Awolowo University Ile-Ife, Osun State, Nigeria.
email: [email protected]
Article 11: Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Author: 1 - Khalid Zaman – Assistant Professor, Department of Management Sciences, COMSATS Institute of Information Technology,
Abbottabad, Pakistan.
email: [email protected], [email protected]
2 - Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed – MS Scholar, Department of Management Sciences,
COMSATS Institute of Information Technology, Abbottabad, Pakistan.
Article 12: Comparing Bankruptcy Prediction Models In Iran
Author: 1 - Ali Ebrahimi Kordlar – Faculty of management, Tehran university, Tehran, Iran.
2 - Nader Nikbakht – Faculty of management, Tehran University, Tehran, Iran.
email: [email protected]
Article 13: Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors
Author: 1 - Adebisi Sunday A. – (PhD) Department of Business Administration, Faculty of Management sciences
University of Ado Ekiti, P.M.B 5363, Ekiti State Nigeria.
email: [email protected]
2 - Babatunde Bayode O. – Department of Business Administration, College of Management and Social Sciences, Osun State
University, P.M.B 2008, Okuku, Osun State, Nigeria.
email: [email protected]
Article 14: The Practicability of Activity-Based Costing (ABC) in the Nigerian Retail Banks
Author: Ahmed Audu Maiyaki – Department of Business Administration Bayero University Kano, Nigeria.
email: [email protected]
Article 15: Social Support as Mental Health Improver for Managerial Women in the Organizational Work Environment
Author: 1 - Jocelyn Sackey, MPhil – Department of Business Administration, Technology and Social Sciences
Luleå University of Technology.
2 - Mohammed-Aminu Sanda – (PhD) Division of Industrial Work Environment, Department of Business Administration,
Technology and Social Sciences, Luleå University of Technology.
email: [email protected]
Case Studies 1: Three Case Studies On Bank Nagari Padang (Indonesia) Headquarters And Main Branch
Author: Heryanto – (PhD) Chief of RfD of the Chamber of Commerce and Industry; Professor of Mangement Science in Indonesia.
email: [email protected]
Case Studies 2: Organizational Communication As An Important Factor Of Company Success: Case Study Of Bosnia And Herzegovina
Author: Kenan Spaho – Center for Techological and Economical Development. Sarajevo, Bosnia and Herzegovina.
email: [email protected]
Short Communi-
cations 1:
A Short Communication on - How a Leading Power Distribution Company Effectively Tracks Business Areas Like Safety,
Finance and Operation for Region and Business Wise for Evaluating their Kpi’s - Using Businessobjects Xcelsius Dashboards.
Author: 1 - Rakesh Tej Kumar Kalahasthi – SAP BI Practice, Bangalore, India.
email: [email protected]
2 - Benson Hilbert – SAP BI Practice, Bangalore, India.
email: [email protected]
Short Communi-
cations 2:
Modeling Strategies for Financial Hedging
Author: José Carlos Arias – (PhD) In Science by the school Robert de Sorbon, Paris, France, Professor of Management Science at the
European Business School, Cambridge, UK.
email: [email protected]
Business Intelligence Journal - July, 2011 Vol.4 No.2
206 Business Intelligence Journal July
INFORMATION FOR CONTRIBUTORS
Electronic submission of manuscripts is strongly
encouraged, provided that the text, tables, and fgures
are included in a single Microsoft Word fle (preferably
in Times New Roman, 12 size font)
Submit manuscript as e-mail attachment to the
BIJ Editorial Offce at: edit.bij@saycocorporativo.
com. A manuscript number will be mailed to the
corresponding author within the following 7 days.
The cover letter should include the corresponding
author’s full address and telephone/fax numbers and
should be in an e-mail message sent to the Editor,
with the fle, whose name should begin with the frst
author’s surname, as an attachment. The authors may
also suggest two to four reviewers for the manuscript
(BIJ may designate other reviewers).
BIJ will only accept manuscripts submitted as
e-mail attachments.
Article Types
Three types of manuscripts may be submitted:
Regular Articles: These should describe new and
carefully confrmed fndings, and research methods
should be given in suffcient detail for others to verify
the work. The length of a full paper should be the
minimum required to describe and interpret the work
clearly.
Short Communications: A Short Communication
is suitable for recording the results of complete
small investigations or giving details of new models,
innovative methods or techniques. The style of main
sections need not conform to that of full-length papers.
Short communications are 2 to 4 printed pages (about
6 to 12 manuscript pages) in length.
Reviews: Submissions of reviews and perspectives
covering topics of current interest are welcome and
encouraged. Reviews should be concise and no longer
than 4-6 printed pages (about 12 to 18 manuscript
pages). Reviews manuscripts are also peer-reviewed.
Review Process
All manuscripts are reviewed by an editor and
members of the Editorial Board or qualifed outside
reviewers. Decisions will be made as rapidly as
possible, and the journal strives to return reviewers’
comments to authors within 3 weeks. The editorial
board will re-review manuscripts that are accepted
pending revision. It is the goal of the BIJ to publish
manuscripts within the following BIJ edition after
submission.
Regular Articles
All portions of the manuscript must be typed
double-spaced and all pages numbered starting from
the title page.
The Title should be a brief phrase describing the
contents of the paper. The Title Page should include
the authors’ full names and affliations, the name of
the corresponding author along with phone, fax and
e-mail information. Present addresses of authors
should appear as a footnote.
The Abstract should be informative and completely
self-explanatory, briefy present the topic, state the
scope of the work, indicate signifcant data, and point
out major fndings and conclusions. The Abstract
should be 100 to 200 words in length. Complete
sentences, active verbs, and the third person should
be used, and the abstract should be written in the past
tense. Standard nomenclature
Should be used and abbreviations should be
avoided. No literature should be cited.
Following the abstract, about 3 to 10 key words that
will provide indexing references to should be listed.
A list of non-standard Abbreviations should be
added. In general, non-standard abbreviations should
be used only when the full term is very long and used
often. Each abbreviation should be spelled out and
introduced in parentheses the frst time it is used in
the text.
The Introduction should provide a clear statement
of the problem, the relevant literature on the subject,
and the proposed approach or solution. It should be
understandable to colleagues from a broad range of
disciplines.
Materials and methods should be complete enough
to allow possible replication of the research. However,
only truly new research methods should be described in
Business Intelligence Journal - July, 2011 Vol.4 No.2
2011 Business Intelligence Journal 207
detail; previously published methods should be cited,
and important modifcations of published methods
should be mentioned briefy. Capitalize trade names
and include the manufacturer’s name and address.
Subheadings should be used. Methods in general use
need not be described in detail.
Results should be presented with clarity and
precision. The results should be written in the past
tense when describing author’s fndings. Previously
published fndings should be written in the present
tense. Results should be explained, but largely without
referring to the literature. Discussion, speculation and
detailed interpretation of data should not be included
in the Results but should be put into the Discussion
section.
The Discussion should interpret the fndings in view
of the results obtained in this and in past studies on the
topic. State the conclusions in a few sentences at the
end of the paper. The Results and Discussion sections
can include subheadings, and when appropriate, both
sections can be combined.
The Acknowledgments of people, grants, funds,
etc should be brief.
Tables should be kept to a minimum and be
designed to be as simple as possible. Tables are to be
typed double-spaced throughout, including headings
and footnotes. Each table should be on a separate
page, numbered consecutively in Arabic numerals and
supplied with a heading and a legend. Tables should
be self-explanatory without reference to the text. The
details of the research methods should preferably be
described in the legend instead of in the text. The same
data should not be presented in both table and graph
form or repeated in the text.
Figure legends should be typed in numerical order
on a separate sheet. Graphics should be prepared using
applications capable of generating high resolution
GIF, TIFF, JPEG or PowerPoint before pasting in
the Microsoft Word manuscript fle. Tables should
be prepared in Microsoft Word. Use Arabic numerals
to designate fgures and upper case letters for their
parts (Figure 1). Begin each legend with a title and
include suffcient description so that the fgure
is understandable without reading the text of the
manuscript. Information given in legends should not
be repeated in the text.
References: In the text, a reference identifed by
means of an author’s name should be followed by the
date of the reference in parentheses. When there are
more than two authors, only the frst author’s name
should be mentioned, followed by ‘et al’. In the
event that an author cited has had two or more works
published during the same year, the reference, both in
the text and in the reference list, should be identifed
by a lower case letter like ‘a’ and ‘b’ after the date to
distinguish the works.
Examples:
Smith (2000), Wang et al. (2003), (Kelebeni, 1983),
(Usman and Smith, 1992), (Chege, 1998; Chukwura,
1987a,b; Tijani, 1993, 1995), (Kumasi et al., 2001)
References should be listed at the end of the paper
in alphabetical order. Articles in preparation or articles
submitted for publication, unpublished observations,
personal communications, etc. should not be included
in the reference list but should only be mentioned in
the article text (e.g., A. Kingori, University of Nairobi,
Kenya, personal communication). Journal names are
abbreviated according to Chemical Abstracts. Authors
are fully responsible for the accuracy of the references.
Examples:
Papadogonas TA (2007). The fnancial performance of
large and small frms: evidence from Greece. Int. J.
Financ. Serv. Manage. 2(1/2): 14 – 20.
Mihiotis AN, Konidaris NF (2007). Internal auditing:
an essential tool for adding value and improving
the operations of fnancial institutions and
organizations. Int. J. Financ. Serv. Manage. 2(1/2):
75 – 81.
Gurau C (2006). Multi-channel banking in Romania:
a comparative study of the strategic approach
adopted by domestic and foreign banks Afr. J.
Financ. Servic. Manage. 1(4): 381 – 399.
Yoon CY,Leem CS (2004).Development of an
evaluation system of personal e-business
competency and maturity levels Int. J. Electron.
Bus. 2(4): 404 – 437.
Business Intelligence Journal - July, 2011 Vol.4 No.2
208 Business Intelligence Journal July
Short Communications
Short Communications are limited to a maximum
of two fgures and one table. They should present
a complete study that is more limited in scope
than is found in full-length papers. The items of
manuscript preparation listed above apply to Short
Communications with the following differences: (1)
Abstracts are limited to 100 words; (2) instead of a
separate Materials and Methods section, research
methods may be incorporated into Figure Legends and
Table footnotes; (3) Results and Discussion should be
combined into a single section.
Proofs and Reprints
Electronic proofs will be sent (e-mail attachment)
to the corresponding author as a PDF fle. Page proofs
are considered to be the fnal version of the manuscript.
With the exception of typographical or minor clerical
errors, no changes will be made in the manuscript at
the proof stage. Because BIJ will be published online
without access restrictions, authors will have electronic
access to the full text (PDF) of the article.
Authors can download the PDF fle from which
they can print unlimited copies of their articles.
Copyright
Submission of a manuscript implies: that the work
described has not been published before (except in the
form of an abstract or as part of a published lecture, or
thesis) that it is not under consideration for publication
elsewhere; that if and when the manuscript is accepted
for publication, the authors agree to automatic transfer
of the copyright to the publisher.
Costs for Authors
Revision, edition and publishing costs will be
totally paid by the IIU Press & Research Centre A.C.
and authors’ sole contribution will be providing BIS
with their invaluable work.
Publication Decisions
Decisions by the editor of all submitted manuscripts
refect the recommendations of members of the
Editorial Board and other qualifed reviewers using
a “blind” review process. Reviewers’ comments
are made available to authors. Manuscripts that
are inappropriate or insuffciently developed may
be returned to the authors without formal review
for submission to a more suitable journal or for
resubmission to BIJ following further development.
Manuscripts submitted will be judged primarily
on their substantive content, though writing style,
structure and length will also be considered. Poor
presentation is suffcient reason for the rejection of
a manuscript. Manuscripts should also be written as
concisely and simply as possible, without sacrifcing
clarity or meaningfulness of exposition. Manuscripts
will be evaluated by the editor when frst received,
on their contribution-to-length-ratio, meaning that
manuscripts with strong contributions will be assigned
more pages than those making narrower contributions.
Papers intended to make very extensive contributions
(over 35 double-space pages, using one inch margins
and Times New Roman twelve-pitch font) will, at
discretion of the editor, be allotted additional space.
Authors are expected to get and use feedback from
colleagues prior to submitting a manuscript for formal
review.
2011 209
BANKING PROFITABILITY DETERMINANTS
James W. Scott (PhD)
Researcher at the Business Intelligence Research Centre of hosted at the Business Intelligence Service, London, UK
2nd Floor, 147 St. John Street, London, UK
Email: [email protected]
José Carlos Arias (PhD, DBA)
Professor of Management Science at the European Business School, Cambridge, UK,
2nd Floor, 147 St. John Street, London, UK
Email: [email protected]
Abstract
Notwithsanding the enormously complex and dynamic nature of the environment in which they compete, there is a growing body
of evidence that suggests it is possible to discern relevant indicators of profitability for the banking industry today. The purpose of this
study was to develop an appropriate econometric model whereby the primary determinants of profitability of the top five bank holding
companies in the United States could be examined and understood. To accomplish this purpose, an econometric model based on internal
aspects of the banking organizations as they related to their return on assets and external aspects of the environment in which they
compete as measured by growth in GDP was developed based on guidance provided by economists and industry experts to determine
the impact of the external national economy of these five leading banks according to their size as measured by total assets. A critical
review of the relevant peer-reviewed, scholarly and organizational literature is followed by an analysis of the statistical data for these
bank holding companies using the econometric model. A summary of the research and salient conclusions are provided in the concluding
chapter. Key words: Finance, Economics, Banking, Profits, Profitability Measurements.
The banking industry in general has experienced some
profound changes in recent decades, as innovations in
technology and the inexorable forces driving globalization
continue to create both opportunities for growth and
challenges for banking managers to remain proftable in
this increasingly competitive environment. Most of the
studies concerning bank proftability to date, including
Short (1979), Bourke (1989), Molyneux and Thornton
(1992), Demirguc-Kunt and Huizinga (2000) and Goddard,
Molyneux, and Wilson (2004b), have employed different
linear models to estimate the impact of various factors
that could be signifcant in terms of explaining profts.
According to Athanasoglou and his colleagues (2005),
these studies were seminal in demonstrating the feasibility
of conducting a meaningful analysis of the determinants
of bank proftability, but some of the methods used by
these studies failed to take into account the robust and
dynamic nature of the economic environment in which they
competed. Moreover, the studies to date have primarily
considered determinants of proftability at the bank and/
or industry level, with the choice of variables used lacking
internal consistency in some instances; in addition, there
has been a dearth of research concerning the potential
infuence of the macroeconomic environment, due in
part to the small time dimension of the panels used in the
estimation (Athanasoglou et al., 2005).
Other factors that have constrained the research to date
include the fact that the econometric methodology used
in the study was not adequately described and/or failed to
account for some features of bank profts which suggest
that the estimates obtained by these studies may have
been biased or inconsistent (Athanasoglou et al., 2005).
Nevertheless, studies have amply demonstrated the ability
of models to better understand what fuels proftability in
fnancial services organizations, and this represents the
purpose of the study which is described further below.
The general purpose of this study was to develop an
appropriate econometric model whereby the primary
determinants of proftability of the top fve bank holding
companies in the United States could be examined and
understood. The specifc purpose of this study was to
determine the impact of the external environment in which
the fve leading bank holding companies in the United
States (as of June 2007) as measured by real growth in GDP
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
210 Business Intelligence Journal July
compared to internal aspects of the respective organizations
as measured by their return on assets as an indicator of
proftability.
This study reviewed the peer-reviewed, scholarly
and organizational literature as it applied to the banking
industry in general and to the United States in particular
to identify recent trends, determinants of proftability for
fnancial services organizations, and to provide the requisite
background to developing a straight-forward econometric
model that could be used to analyze bank-level data for the
nation’s fve leading bank holding companies as measured
by total assets.
According to Goddard and his colleagues (2004a), in
spite of the growing body of research into determinants
of banking proftability, there remains a paucity of studies
that have investigated the specifc relationship between
organizational size and its impact on proftability. These
authors report that, “Previous studies of the dynamics of
growth on the one hand, and proft on the other, have in
the main developed separately, and followed contrasting
empirical methodologies. Nevertheless, there are several
theoretical arguments to suggest that these two performance
indicators are closely related. [However], few researchers
have tested for empirical relationships between growth and
proft directly” (Goddard et al., 2004a p. 1069). Moreover,
as Cover (1999) emphasizes, the need for identifying
determinants of proftability in the banking industry has
never been greater: “As banks move into the twenty-frst
century, they must focus more than ever before on creating
new streams of revenue in order to shareholder value.
Crucial to this effort is the need to assess and analyze the
proftability of the bank’s current customers, relationships,
services, and products. It is only through such analyses that
banks can determine which customers to fght for, which
customer relationships to expand, and which prospective
customers to pursue” (p. 78).
The rationale that guided this study was that the larger
the bank holding company, the more impervious it would
be to downturns in the external environment as measured
by real growth in GDP. Larger fnancial services companies
enjoy a wide range of advantage by virtue of economies of
scale, accumulated tacit knowledge and expertise as well
as the resources needed to continue their expansions into
foreign markets where barriers to entry may be prohibitive
for smaller players in the market. For example, according to
Goddard, Molyneux and Wilson (2004), “Previous studies
of the determinants of concentration have proposed various
explanations as to why some frms grow and attain large
size. These include economies of scale or scope, effciency
gains attained through size, the adoption of entry-deterring
strategies, or the exercise of other forms of market power”
(p. 1069). Likewise, as noted above, a number of studies
have demonstrated the feasibility of using bank-level data
to investigate proftability determinants. In addition, the
research to date also suggests that banking frms are one of
the best places to test for the effects of various proftability
factors such as technological innovation. In this regard,
Berger and Deyoung (2006) emphasize that, “Banks
have embraced substantial advances in both physical and
fnancial technologies during the past two decades, and
the broader industry category of which banking is a part,
Depository and Nondepository Financial Institutions, is
the most information technology-intensive industry in the
United States” (p. 1483).
This study used a fve-chapter format to achieve the
above-stated research goals. The frst chapter introduced
the topics under consideration and provided a statement
of the importance of the study, its importance and scope,
as well as its supporting rationale. The second chapter of
the study consists of a critical review of the relevant peer-
reviewed, scholarly and organizational literature. The third
chapter more fully describes the econometric methodology
used to analyze the proftability of the top fve banks in the
United States today, and chapter four presents the analysis
of the data. Finally, chapter fve provides a summary of the
research, salient conclusions and recommendations.
Review of Related Literature
Background and Overview.
On the one hand, the banking industry today enjoys a
number of advantages compared to past years that would
appear to contribute to their ability to generate profts.
According to Berger and Deyoung (2006), the banking
industry in the United States has been in a constant process
of geographic expansion in recent years, both within
nations and across nations. These authors report that, “At
one time, nearly all customers were served by locally
based institutions. In contrast, it is now much more likely
that the bank or branch providing services is owned by an
organization headquartered a substantial distance away,
perhaps in another state, region, or nation” (p. 1483).
As an example, these authors note that between 1985
and 1998 the distance between the largest bank and the
other affliate banks in U.S. multibank holding companies
(MBHCs) increased by over 50 percent on average, from
123.35 to 188.91 miles, as a number of MBHCs acquired
2011 211
banks in other states and regions (Berger & Deyoung, 2006).
Moreover, the banking industry, like any other industry, will
experience potential diseconomies to geographic expansion
in the form of agency costs associated with monitoring
junior managers in a distant locale; however, innovations
in information processing and telecommunications may
lessen these agency costs by improving the ability of senior
managers located at the organization’s headquarters to
monitor and communicate with staff at distant subsidiaries
(Berger & Deyoung, 2006).
In the modern banking industry, technologies such as
ATM networks and transactional Internet websites allow
banks to interact more effciently with their customers
regardless of geographic proximity; furthermore, recent
innovations in fnancial technologies provide the capacity
to provide these services using long-distance interfaces with
customers. According to Berger and Deyoung, “Greater
use of quantitative methods in applied fnance, such as
credit scoring, may allow banks to extend credit without
geographic proximity to the borrower by ‘hardening’ their
credit information” (p. 1483). Likewise, new product mixes
of fnancial engineering, such as derivative contracts, may
provide banks of all sizes to unbundle, repackage, or
hedge risks at lower costs without regard to the geographic
proximity to the other party (Berger and Deyoung, 2006).
These fnancial innovations may also provide senior banking
managers with the ability monitor the decisions made by loan
offcers and managers at distant affliate banks more easily,
and to evaluate and manage the contributions of individual
affliate banks to the organization’s overall returns and risk
more effciently as well (Berger & Deyoung, 2006).
On the other hand, these same trends have introduced
some additional constraints on the banking industry as
they seek to maintain their existing market share and grow
their companies along multinational lines. In this regard,
Saunders and Walter (1998) report that, “Financial systems
that are deemed ineffcient or incomplete are characterized
by a limited range of fnancial services and obsolescent
fnancial processes. Both static and dynamic effciency are
of obvious importance from the standpoint of national and
global resource allocation, not only within the fnancial
services industry itself but also as it effects users of
fnancial services” (p. 19). In other words, because fnancial
services can be regarded as being “inputs” to the overall
production process of a country, the level of national output
and income -- as well as its rate of economic growth -- are
directly affected by the effciency characteristics of the
fnancial services sector (Saunders and Walter, 1998).
A “retarded” fnancial services industry, in this sense,
can represent a major impediment to a nation’s overall
real economic performance. Such retardation represents
a burden on the fnal consumers of fnancial services and
potentially reduces the level of private and social welfare. It
also represents a burden on producers, by raising their cost
structures and diminishing their competitive performance in
domestic and global markets (Saunders and Walter, 1998).
Therefore, any such retarded fnancial services industry
skews the trends involved in the resource allocation in the
national economy.
Financial system ineffciencies can be traced to a number
of factors:
1. Regulations that prevent fnancial frms from complete
access to alternative sources of funding or the full range
of borrowers and issuers;
2. Taxation imposed at various stages of the fnancial
intermediation process, including securities transfer
taxes, transactions taxes, etc.;
3. Lack of competition that reduces incentives to cut
intermediation costs and promote innovation;
4. Lack of market discipline imposed on owners and
managers of fnancial intermediaries, leading to poor
risk management, agency problems, and increased
costs; and,
5. Existence of major information imperfections among
contracting parties (Saunders and Walter, 1998).
National fnancial systems that are statically and/
or dynamically ineffcient tend to be disintermediated.
Borrowers or issuers in a position to do so seek foreign
markets or offshore markets that offer lower costs or a
more suitable range of products. Investors likewise seek
markets abroad that offer higher rates of return or improved
opportunities to construct more effcient portfolios.
These types of systems can be described as being
either “uncompetitive” or “unattractive” as venues for
fnancial intermediation in the context of global markets;
nevertheless, individual fnancial services institutions may
be able to cross-subsidize foreign activities from abnormal
profts earned domestically for a period of time (Saunders
and Walter, 1998).
The delivery of fnancial services can be conceptualized
as a market structure that combines three principal
dimensions in the delivery of fnancial services in terms of
the clients served (C), the geographic arenas where business
is done (A), and the products supplied (P) (Saunders and
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
212 Business Intelligence Journal July
Figure ___. The Client-Arena-Product (C-A-P) Matrix.
Source: Saunders and Walters, 1998 at p. 20.
The inherent attractiveness of each cell to the suppliers
of fnancial services is inextricably related to the size of
the prospective risk-adjusted returns that can be extracted
from it; in addition, the durability of returns generated will
depend on the ability of new players to enter the cell and the
development of substitute products over time (Saunders and
Walters, 1998). Based in large part on recent innovations in
technology and fnancial deregulation, competitors in the
fnancial services industry are confronted with expanded
potential access to each dimension of the C-A-P opportunity
set (Saunders and Walters, 1998)
To achieve maximum proftability, these authors
recommend that fnancial services companies allocate their
available resources to those C-A-P cells in Figure 1 that
promise to provide them with the highest risk-adjusted
returns. According to these authors, though, “In order to do
this, they will also have to allocate capital, costs, returns, and
risks appropriately across cells. Beyond this, however, the
economics of supplying fnancial services internationally
is jointly subject to economies of scale and economies
of scope. The existence of both types of economies have
strategic implications for players in the industry; indeed,
economies of scale suggest an emphasis on deepening the
activities of individual frms within a cell, or across cells
in the product dimension” (emphasis added) (Saunders and
Walter, 1998 p. 21).
Moreover, economies of scope indicate there has been
an effort made to broaden fnancial services activities
across cells; in other words, a bank can produce a given
level of output in a given cell more inexpensively or market
it more effectively than institutions that are less active
across such multiple cells; however, this depends on the
benefts and costs of linking cells together in a coherent
fashion (Saunders and Walter, 1998).
In this regard, regulation of fnancial services can be
viewed as having an important infuence in terms of:
1. Accessibility of geographical arenas;
2. Accessibility of individual client groups by players
originating in different sectors of the fnancial services
business; and,
3. Substitutability among fnancial products in meeting
personal, corporate, or government fnancial needs
(Saunders and Walter, 1998).
Financial intermediaries are clearly sensitive to
incremental competition in C-A-P cells as illustrated in
Figure __ above, particularly where economic entry barriers
are limited (Saunders and Walter, 1998). Furthermore,
“Market penetration by competitors can erode indigenous
players' returns and raises protectionist motivations.
Given the economic interests involved, banks and other
fnancial institutions are in an excellent position to convert
them into political power in order to achieve protection
against potential rivals. They are often exceedingly well
connected politically, and their lobbying power motivated
by protectionist drives can be awesome” (Saunders and
Walter, 1998 p. 21).
Competitive distortions in the fnancial services industry
take the form of entry barriers and operating restrictions. In
terms of Figure 1 above, entry barriers tend to restrict the
movement of fnancial services frms in the lateral "arenas"
dimension of the matrix; a fnancial services organization
that is locked out of a particular national market does not
enjoy the same level of lateral opportunity set that excludes
the relevant segment of "client" and "product" cells
(Saunders and Walter, 1998).
Assuming that a fnancial services organization has
successfully gained access to a particular arena, there are
a number of operating restrictions that carry the potential
to constrain either the depth of service it can supply
to a particular cell (e.g., lending limits, staffng limits,
restrictions on physical location) or in the feasible set of
cells within the tranche (e.g., limits on services banks or
securities frms are allowed to supply and the client groups
Walter, 1998). In this regard, Figure 1 below illustrates
these dimensions in the form of a matrix of C X A X P cells
wherein individual cell characteristics can be analyzed in
terms of conventional competitive structure criteria.
2011 213
Table 1.Banking Structure in the United States.
Year
No. of
Banks
Total Assets
($ Billion)
No. of Banks
w/Assets>$10
Billion
% of Assets Held
by Large Banks
2000 8,315 $6.2 82 70%
1980 14,769 $1.9 18 34%
Source: FDIC cited in Gup, 2003 at p. 2.
Figure 2. No. of Banks in the United States: 2000 versus 1980
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
No. of Banks
2000 1980
Source: Based on tabular data cited in Gup, 2003 at p. 2.
$0.00
$1.00
$2.00
$3.00
$4.00
$5.00
$6.00
$7.00
Total Assets ($ billion)
2000 1980
Figure 3. Total Assets of Banks in the United States (in $
billion): 2000 versus 1980.
Source: Based on tabular data cited in Gup, 2003 at p. 2.
they are allowed to serve). Operating restrictions in turn
can be subclassifed in terms of whether they place limits
on the kinds of fnancial services that may be sold locally
(Type A) or the specifc client-groups that may be served
(Type B). According to these authors, “Operating limits
may severely reduce proftability associated with the arena
concerned, while creating signifcant excess returns for the
protected industry; regulators may also tolerate a certain
amount of anticompetitive, cartel-like behavior on the part
of domestic fnancial institutions” (Saunders and Walter,
1998 p. 21).
Economies of scope and scale may be signifcantly
constrained by entry and operating restrictions in a
particular market, indicating the importance of the impact
of competitive distortions on horizontal integration in
the fnancial services industry. Under universal banking
structures the entire C-A-P matrix is available to fnancial
institutions in terms of their competitive positioning and
execution, providing the potential -- absent anticompetitive
behavior -- for maximum static and dynamic effciency
in the fnancial system. Restrictions on entry by "ft and
proper" players, regardless of whether they are domestic
or international, as well as limits on the business that may
be done and the clients that may be served, represent a
signifcant threat for eroding the domestic and international
performance of fnancial institutions and of the national
economy as a whole (Sanders and Walters, 1998 p. 22).
As noted above, consolidation by acquisition or
otherwise in the banking industry in the United States from
1980 to 2000 is one of the most signifcant changes to affect
the industry in recent years (Gup, 2003). In their recent
study, “Competition from Large, Multimarket Firms and the
Performance of Small, Single-Market Firms: Evidence from
the Banking Industry,” Berger and his colleagues (2007)
report that, “Over the last two decades, retail banking in
the U.S. has changed dramatically. Large banks that branch
across multiple local markets have signifcantly increased
their share of local markets, and small banks that operate
in a single market have experienced substantially reduced
local market shares” (p. 331). In 1982, large, multimarket
banks--banks with gross total assets (GTA) over $1 billion
(real 1994 dollars) with branch offces in more than one
local market held 23 percent of local deposits in U.S.
metropolitan markets; however, by 2000, this fgures had
almost tripled to 65 percent (Berger et al., 2007). During this
same time period, shares in these markets of small, single-
market banks (GTA [less than or equal to] $1 billion, offces
in only one market) decreased by approximately two-thirds
from 60 percent to 19 percent (Berger et al., 2007).
As shown in Table 1 below, the number of banks
decreased by more than 6,000, while the percentage of total
assets held by the largest banks doubled; as a result, 82 large
banks representing less than 1 percent of the total number
of banks now hold more than two-thirds of all bank assets:
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
214 Business Intelligence Journal July
Figure 4. No. of Banks in the United States with Assets Greater
than $10 billion: 2000 versus 1980.
0
20
40
60
80
100
No. of Banks w/Assets > $10 Billion
2000 1980
Source: Based on tabular data cited in Gup, 2003 at p. 2.
Figure 5. Percentage of Assets Held by Large Banks in the
United States: 2000 versus 1980.
0%
10%
20%
30%
40%
50%
60%
70%
% of Assets Held by Large Banks
2000 1980
Source: Based on tabular data cited in Gup, 2003 at p. 2.
According to Toyne and Tripp (1998), reductions
in regulatory restrictions on interstate bank mergers
and increased proftability in the banking industry have
resulted in higher levels of merger and acquisition activity
in the United States in recent years. The trend towards
consolidation in the banking industry, though, is certainly
not unique to the United States. For example, a study by the
central banks of the Group of Ten Nations (G-10) identifed
a high level of concentration in all of the 13 countries they
investigated (Gup, 2003). These consolidations have been
the result of the following:
1. Deregulation of geographic markets;
2. Shifts in fnancial technology (e.g., securitization and
derivatives);
3. Innovations in communications technology;
4. Changes in information technology;
5. The desire to achieve economies of scale;
6. High stock prices used as currency in mergers, and,
7. Other factors (Gup, 2003).
In general, the consolidation involves large banks
within one country, cross-border bank consolidations, and
consolidations between banks and other types of fnancial
institutions (Gup, 2003). Some of the “other factors”
identifed by Gup (2003) above include the need to respond
to new competition from smaller bank that are able to
market higher proft products more effciently than their
larger counterparts. These and others determinants of bank
performance are discussed further below.
Determinants of Bank Performance.
While virtually all of the studies reviewed herein
emphasize the need for yet more studies, there has been
a growing body of evidence concerning the ability of
researchers to identify accurate determinants of bank
performance in recent years. Most of the studies on the
determinants of bank’s interest margin and proftability
have focused whether on a particular country (Berger,
1995; Guru et al., 2002; Barajas et al., 2001; Ben Naceur
and Goaied, 2001) or on a panel of countries (Abreu and
Mendes, 2002; Demerguç-Kunt and Huizingha, 1999). The
empirical evidence for the United States has largely been
compiled by Berger (1995), Neeley and Wheelock (1997)
and Angbazo (1997) (all cited in Ben Naceur, 2003).
One study by Berger (1995) investigated the relationship
between the return on equity and the capital asset ratio for
a sample of banks in the United States for the period 1983-
1992; using the Granger causality model, this researcher
determined that the return of equity and capital to asset
ratio tend to be positively related. Similarly, Neeley and
Wheelock (1997) studied the proftability of a sample of
insured commercial banks in the US for the 1980-1995
period. These researchers determined that bank performance
is positively related to the annual percentage changes in
the state’s per capita income. Likewise, Anghazo (1997)
studied the determinants of bank net interest margins for a
representative sample of banks in the United States for the
period 1989-2003; the fndings of that study showed that
default risk, the opportunity cost of non-interest bearing
reserves, leverage and management effciency are also
2011 215
Author/Date/Title/
Publication
Key Findings Comments
Spindler, J. Andrew, Jonathan T. B. Howe, and
David F. Dedyo. (1990, May). “The Performance
of Internationally Active Banks and Securities
Firms Based on Conventional Competitiveness
Measures”. Federal Reserve Bank of New York,
Mimeo.
In terms of size, U.S. banks fell consistently throughout the
period, especially against their Japanese competitors. The same
approximate pattern emerged with respect to revenue growth.
(The authors note that while a signifcant component of this
“size efect” is attributable to the decline in the value of the dollar
against the yen and other major currencies over the period.) They
performed slightly below the sample mean on ROA, well below
their Swiss, British, and Japanese counterparts. The same was true
of ROE in comparison with Japanese and French banks. In terms
of productivity, U.S. banks fell into the middle of the range. On
capitalization, U.S. banks were high in the rankings on the frst
measure cited above, but well behind Japanese, German, and
Swiss banks on the second.
The authors report the results of a study by the Federal Reserve
Bank of New York attempted to assess the performance of 51
banks and securities frms based in various countries (the
United States, Canada, France, Germany, Japan, Switzerland,
and the United Kingdom) that were internationally active
during the second half of the 1980s. The performance
measures used include frm size (total assets and total revenue,
in real terms); proftability (return on assets [ROA] and return
on equity [ROE]); productivity (ratio of total revenue to total
non-interest expense); and capitalization (as measured by
the ratio of shareholder equity to total assets, and the ratio of
market capitalization to reported earnings).
Goddard, John, Phil Molyneux and John O.S.
Wilson. (2004). “Dynamics of Growth and
Proftability in Banking.” Journal of Money, Credit
& Banking, 36(6), 1069.
Empirical research concerning the dynamics of company
proftability is based on an account of the determinants of proft
that is an alternative to the essentially static Structure-Conduct-
Performance (SCP) paradigm; however, although the relevant
microtheory identifes SCP relationships applicable when markets
are in equilibrium, there is no certainty that a proft fgure observed
at any point in time represents an equilibrium value.
The hypotheses tested in the persistence of proft (POP)
literature are that entry and exit are sufciently free to
eliminate any abnormal proft quickly, and that all frms’ proft
rates tend to converge to the same long-run average value.
The alternative is that some frms possess special knowledge
or other advantages enabling them to prevent imitation or
block entry. If so abnormal proft may tend to persist from
year to year, and diferences in average proft rates may be
sustained indefnitely. Empirical tests of the POP hypothesis
in banking are few in number; however, recent studies have
presented extensive evidence of POP in U.S. banking.
Stiroh, Kevin J. (2004). “Diversifcation in Banking:
Is Noninterest Income the Answer?” Journal of
Money, Credit & Banking, 36(5), 853.
These researchers conclude that the banking industry is the
U.S. is steadily shifting away from traditional sources of revenue
such as loan making and toward activities that generate fee
income, service charges, trading revenue, and other types of
noninterest income; while noninterest income has always played
an important role in banking revenue, there is a clear trend that it
is gaining importance. By 2001, noninterest income represented
43% of net operating revenue (net interest income plus
noninterest income), an increase from just 25% in 1984. Also note
that, “banks with higher noninterest income shares have lower
proftability per unit of risk” (p. 853).
The shift toward noninterest income has contributed to
higher levels of bank revenue in recent years; there is also an
indication that it will lower the volatility of bank proft and
revenue and reduce risk.
Kaushik, Surendra K. and Raymond H. Lopez.
(1996) “Proftability of Credit Unions, Commercial
Banks and Savings Banks: A Comparative Analysis.”
American Economist, 40(1), 66.
Commercial banks and credit unions experienced growth in
assets faster than growth in loans during the 1990s; as a result,
their investment portfolios have been increasing in absolute and
relative size. “This tends to hold down proftability since margins
on loans are greater than those on investments” (p. 67).
Authors also determined that credit union loan portfolios have
grown more rapidly than either commercial banks’ or savings
institutions’ in recent years, and their net interest margins have
consistently been above banks.
Prendergast, C. (1993, March). “The Provision
of Incentives in Firms.” Journal of Economic
Literature, 7, 63.
Studies have shown that some incentive structures may result
in dysfunctional behavior; this may occur more frequently when
incentives within regulated fnancial services companies relate to
volume and create a clear bias towards writing business.
Bank managers may be rewarded by the volume of loans rather
than their risk-adjusted proftability; numerous instances of
bank distress have been related to inappropriate incentive
structures that created a bias in favor of the bank’s balance
sheet growth.
Grigorian, David A. and Vlad Manole. (2006).
“Determinants of Commercial Bank Performance
in Transition: An Application of Data Envelopment
Analysis.” Comparative Economic Studies, 48(3),
497.
One of the methods used in the literature to evaluate productivity
and performance of banks is the data envelopment analysis (DEA),
a non-parametric method that provides a framework in which it
is possible to account for a wide range of functions performed by
the banks.
This method compares relative performance of decision-
making units (DMU) (i.e., banks) by constructing a frontier
comprised of the most efcient DMUs and focusing on how
close other DMUs are to this frontier.
Hasan, Iftekhar and William C. Hunter. (1996).
“Management Efciency in Minority- and Women-
Owned Banks.” Economic Perspectives, 20(2), 20.
Studies comparing the economic performance of smaller minority-
and nonminority-owned banks have generally shown that the
minority-owned banks have tended to be smaller, somewhat less
proftable, and more expenditure prone than comparable groups
of nonminority banks.
Previous studies reported that smaller minority-owned banks
tended to operate with lower ratios of equity capital to assets,
to employ more conservative asset portfolio management
policies, and to post higher loan losses than their nonminority
peers.
Li, Shaomin, Yigang Pan and David K. Tse. (1999).
“The Impact of Order and Mode of Market Entry
on Proftability and Market Share.” Journal of
International Business Studies, 30(1), 81.
Large frms have more resources to invest in innovation, pursue
more aggressive expansion strategies, and perform better; in
addition, large frms beneft from economies of scale, scope, and
learning. “In short, large frms tend to perform better holding
other factors constant” (p. 81).
Financial services’ proftability and market share performance
are determined by a number of diferent factors and the size of
frms has long been of interest to business researchers.
Table 2.Recapitulation of Proftability Determinants in the Banking Industry.
all positively associated with bank interest spread (Ben
Naceur, 2003).
Table 2 below provides a recapitulation of comparable
studies of proftability indicators in the banking industry in
general, and in the United States in particular.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
216 Business Intelligence Journal July
Author/Date/Title/
Publication
Key Findings Comments
Cover, Jerry. (1999). “Proftability Analysis - A
Necessary Tool for Success in the 21st Century.”
ABA Banking Journal, 91(2), 78.
“Pareto’s Curve,” or the so-called 80/20 rule, holds that 80
percent of all business activity results from 20 percent of current
customers; however several recent studies reveal that in the
banking business, the ratio is even more extreme. One study found
that 15 percent of a bank’s customer base is responsible for 85
percent of its proftability. In the small business banking industry,
the ratio is even more pronounced with fewer than 10 percent of
a bank’s relationships produce 90 percent of its profts. In a typical
retail portfolio, 20 percent of accounts contribute profts equaling
200 percent of the overall return, while up to half of the accounts
generate losses.
“The real winners will be those banks who take advantage of
this window of opportunity. Clearly, it is a window, because
other banks and non-bank competitors are focusing on those
same customers” (p. 78).
Frieder, Larry A. (1991). “Determinants of Bank
Acquisition Premiums: Issues and Evidence.”
Contemporary Policy Issues, 9(2), 14.
This study found that both the internal factors of returns on equity
(ROE) of the target organization and the external factor of the
market growth rate of the target’s state (as measured by deposit
growth and projected population growth) were signifcant positive
determinants of a bank’s capacity for expansion and growth.
It is also possible to measure a fnancial services company’s
proftability by using either return on assets (ROA), ROE, the
ratio of operating earnings to assets, or the net interest spread
to assets.
Berger, Allen N., Astrid A. Dick, Lawrence
G. Goldberg and Lawrence J. White. (2007).
“Competition from Large, Multimarket Firms and
the Performance of Small, Single- M a r k e t
Firms: Evidence from the Banking Industry.”
Journal of Money, Credit & Banking, 39(2-3), 331.
Under the efciency hypothesis, technological progress in
the 1990s signifcantly improved the performance of large,
multimarket banks relative to small, single-market banks;
therefore, a greater presence of large, multimarket banks exerted
more competitive pressure and had more deleterious efects on
the performance of small, single-market banks in their markets in
the second period, 1991-2000, than in the frst period, 1982-90.
The more intense competition from large, multimarket banks in
the second time period may be manifested in decreased revenues
for small, single-market banks (e.g., lower fees or rates on loans,
lower fees on deposits) and/or increased expenses (e.g., higher
rates on deposits, additional expenses on advertising or quality to
retain customers).
Authors note that the relevant research on bank size and
performance in the U.S. includes studies of cost and revenue
performance, as well as the abilities of banks of diferent sizes
to provide retail services in which both large and small banks
compete, such as loans to small businesses and deposits.
Having established some of the important determinants
of proftability for the banking industry today, a review of
the leading bank holding companies in the U.S. today is in
order, and this review is provided below.
Review of Top Five Bank Holding Companies in the
United States Today.
Rank Institution Name Location Total Assets
1 Citigroup Inc. New York, NY $2,220,866,000
2 Bank of America
Corporation
Charlotte, NC $1,535,684,280
3 JPMorgan Chase & Co. New York, NY $1,458,042,000
4 Wachovia Corporation Charlotte, NC $719,922,000
5 Wells Fargo &
Company*
San Francisco,
CA
$539,865,000
Note: As of June 30, 2007, Taunus Corporation was listed as the 5
th
largest bank holding company by the Federal Reserve; however, in
2003, Taunus changed its organization from a Bank Holding Company
to a Financial Holding Company – Domestic (seehttp://www. ffec.
gov/nicpubweb/ nicweb/InstitutionHistory.aspx?parID_RSSD=
2816906&parDT_END =99991231), and was therefore was replaced
with the next top BHC, Wells Fargo & Company, for the purposes of this
analysis; a list of the top ffty bank holding companies in the U.S. as of
June 2007 is provided at Appendix A.
Source: Federal Reserve National Information Center, 2007.
Table 3. United States' Largest Bank Holding Companies (as of
June 30, 2007).
Figure 6. United States' Largest Bank Holding Companies (as
of June 30, 2007).
$0
$500,000,000
$1,000,000,000
$1,500,000,000
$2,000,000,000
$2,500,000,000
Total Assets
Citigroup BAC JPMorgan Wachovia Wells Fargo & Co.
Source: Based on tabular data from Federal Reserve National
Information Center, 2007.
The top fve banking holding companies in the United
States today by consolidated assets are, respectively,
Citigroup Inc., Bank of America Corporation, JPMorgan
Chase & Co., Wachovia Corporation and Wells Fargo &
Company. These companies’ respective rank in the banking
industry, city and state of their headquarters, and total, are
shown in Table 3 and Figure 6 below.
2011 217
Overview of Top Five Bank Holding Companies in the
United States (as of June 2007).
An overview and brief description of the top fve bank
holding companies in the United States as of June 2007 is
provided below, followed by a recapitulation of management
summaries from each bank holding company’s most
recent Form 10-Q fling with the Securities and Exchange
Commission (SEC).
Citigroup Inc.
Citigroup, Inc. (hereinafter “Citigroup” or alternatively,
“the bank”) was established in 1812 and is headquartered
New York City (Citigroup, 2007). Citigroup provides a wide
range of fnancial services to its domestic and international
customers through the following business segments.
Business Segment Description
Global Consumer segment This segment features banking, lending,
insurance, and investment services; as of
March 31, 2007, this business unit boasted
a network of 8,140 branches, about 19,100
automated teller machines, 708 automated
lending machines, and the Internet.
Markets and Banking This segment is responsible for the
provision of various investment and
commercial banking services and
products; these services consist of
investment banking and advisory services,
debt and equity trading, institutional
brokerage, foreign exchange, structured
products, derivatives, and lending
(Citigroup, 2007). The bank also provides
cash management and trade fnance for
corporations and fnancial institutions;
custody and fund services to insurance
companies and pension funds; clearing
services to intermediaries; and depository
and agency/trust services to multinational
corporations and governments (Citigroup,
2007).
Global Wealth Management This segment of the bank ofers investment
advice, fnancial planning, and brokerage
services to afuent individuals, companies,
and non-proft organizations (Citigroup,
2007). This segment also provides wealth
management services for its high net
worth clients, including investment
management, such as investment funds
management, capital markets solutions,
trust, fduciary, and custody services;
investment fnance that comprises credit
services, such as real estate fnancing,
commitments, and letters of credit; and
banking services, which consist of deposit,
checking, and savings accounts, as well
as cash management and other banking
services (Citigroup, 2007).
Alternative Investments The bank’s Alternative Investments
segment manages products across fve
asset classes, such as private equity, hedge
funds, real estate, structured products, and
managed futures (Citigroup, 2007).
Bank of America Corp.
Bank of America Corporation was established in 1874
and is currently headquartered in Charlotte, North Carolina
(Bank of America, 2007). Bank of America Corporation
competes today through its banking and non-banking
subsidiaries as a provider of fnancial services and products
throughout the United States and in selected international
markets (hereinafter “the bank” or alternatively, “the
bank” or “Bank of America”). Through its banking and
nonbanking subsidiaries, the bank and its subsidiaries
provide a diverse range of fnancial services and products
throughout the U.S. and in selected international markets
(Bank of America, 2007). As of June 30, 2007, the Bank of
America operated its banking activities under two primary
charters: (a) Bank of America, National Association (Bank
of America, N.A.) and (b) FIA Card Services, N.A. (Form
10-Q, 2007).
The bank competes through three business segments as
shown in Table 5 below.
Table 5. Bank of American Business Segments.
Business Segment Description
Global Corporate and
Investment Banking
This segment provides commercial and
corporate bank loans, indirect consumer
loans, commitment facilities, real estate
lending products, and leasing and asset-
based lending products for banking
clients, middle market commercial clients,
multinational corporate clients, public
and private developers, homebuilders,
and commercial real estate frms; advisory
services, fnancing, and related products for
institutional investor clients in support of
their investing and trading activities; debt
and equity underwriting, merger-related
advisory services, and risk management
solutions; and treasury management, trade
fnance, foreign exchange, short-term credit
facilities, and short-term investing options for
correspondent banks, commercial real estate
frms, and governments.
Global Consumer and
Small Business Banking
This segment provides savings accounts,
money market savings accounts, certifcate
of deposits, individual retirement accounts,
and regular and interest-checking accounts;
consumer cards, business cards, debit cards,
international cards, and merchant services;
mortgage products for home purchase and
refnancing needs; insurance services; and
lines of credit and home equity loans.
Global Wealth and
Investment Management.
This segment features wealth management
and retail brokerage services, as well as asset
management services, including mutual
funds, liquidity strategies, and separate
accounts.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
218 Business Intelligence Journal July
As of July 30, 2007, the company operated approximately
5,700 retail banking offces and 17,000 automated teller
machines (Bank of America, 2007).
J. P. Morgan Chase & Company.
The company was established in 1823 and is currently
headquartered in New York, New York (JPMorgan, 2007).
According to its company profle, JPMorgan Chase & Co.
(hereinafter “the company” or alternatively, “the bank”),
through its subsidiaries, provides a range of fnancial
services worldwide (JP Morgan, 2007). The bank currently
operates through six segments as shown in Table 6 below.
Table 6. JP Morgan Business Segments.
Business Segment Description
Investment Bank This segment ofers investment banking
products and services, such as advising on
corporate strategy and structure, capital raising
in equity and debt markets, risk management,
market-making in cash securities and derivative
instruments, and research. It serves corporations,
fnancial institutions, governments, and
institutional investors.
Retail Financial Services This segment provides regional banking
services, including consumer and business
banking, home equity lending, and education
lending, as well as ofers mortgage banking and
auto fnance services.
Card Services This segment issues credit cards, and general-
purpose cards to individual consumers, small
businesses, and partner organizations, including
cards issued with AARP, Amazon, Continental
Airlines, Marriott, Southwest Airlines, Sony,
United Airlines, and Walt Disney Company
brands.
Commercial Banking This segment provides lending, treasury services,
investment banking, and asset management
services to corporations, municipalities, fnancial
institutions, and not-for-proft entities.
Treasury and Securities
Services
This segment provides transaction, investment,
and information services to institutional clients.
It also ofers custodian services and cash
management solutions, including trade fnance
and logistics solutions, wholesale card products,
and liquidity management services.
Asset Management This segment provides investment and wealth
management services to institutions, retail
investors, and high-net-worth individuals.
It ofers global investment management
services; trust, estate, and banking services; and
retirement services (JPMorgan, 2007).
Wachovia Corp.
The parent company, Wachovia Corporation, engages in
capital management, the general bank, wealth management,
and the corporate and investment bank businesses.
Wachovia (hereinafter “the company” or alternatively, “the
bank”), provides a range of commercial and retail banking,
as well as a range of trust services through its full-service
banking offces in the U.S. (Wachovia, 2007).
The bank provides its customers with a full range of
checking, savings, check card, foreign currency, annuities,
life insurance, brokerage account transfers, CAP accounts,
individual retirement accounts, credit cards, home equity,
mortgage, hazard and food insurance, escrow, taxes, private
mortgage insurance, education loans, online services, online
banking, online bill pay, and online brokerage services
(Wachovia, 2007). Wachovia was established in 1879 and
is currently headquartered in Charlotte, North Carolina
(Wachovia, 2007).
The company also provides various other fnancial
services, including mortgage banking, investment banking,
estate planning, investment advisory, asset management,
credit and debit card products, trust services, charitable
services, mortgage banking, asset-based lending, leasing,
insurance, and international and securities brokerage
services. In addition, it engages in equity and debt
underwriting activities, private equity investment activities,
derivative securities activities, investment and wealth
management advisory businesses, and brokerage activities.
As of June 1, 2006, Wachovia operated 3,159 offces in 16
states, as well as operated 40 offces internationally. The
company reports assets of approximately $541.8 billion, as
of June 1, 2006 (Wachovia, 2007).
On a fnal note, according to the company’s most recent
Form 10-Q fling:
In the second quarter of 2007, we announced a
realignment of some of our businesses and other corporate
functions. This included the combination of the General
Bank’s private advisory group into the Wealth Management
businesses and the General Bank’s commercial real
estate business into the Corporate and Investment Bank’s
investment banking line of business, which is expected to
occur by the end of the year. We are still evaluating how
these realignments may affect our segment reporting for
future periods; however, we expect Wealth Management
will remain a separate reporting segment. (Form 10-Q, p. 4)
Wells Fargo & Company.
Wells Fargo & Company was established in 1929 and is
currently headquartered in San Francisco, California (Wells
Fargo, 2007). According to its company profle, Wells Fargo
& Company (hereinafter “the company” or alternatively,
“the bank”), through its subsidiaries, provides banking and
fnancial products and services in the United States. The
bank operates in three segments as shown in Table 7 below.
2011 219
Table 7. Wells Fargo & Company Business Segments.
Business Segment Description
Community Banking This segment provides a comprehensive
group of deposit products, including checking
accounts, savings deposits, market rate
accounts, individual retirement accounts, time
deposits, and debit cards; this segment’s loan
portfolio includes: lines of credit; equity lines
and loans; equipment and transportation loans,
including recreational vehicle and marine;
education loans; origination and purchase
of residential mortgage loans; servicing of
mortgage loans; and credit cards (Wells Fargo,
2007). This segment also provides receivables
and inventory fnancing, equipment leases, real
estate fnancing, small business administration
fnancing, venture capital fnancing, cash
management, payroll services, retirement plans,
health savings accounts, and credit and debit
card processing services.
Wholesale Banking This segment provides commercial, corporate,
and real estate banking products and services
in the United States. This segment’s products
include traditional commercial loans and lines
of credit, letters of credit, asset-based lending,
equipment leasing, mezzanine fnancing, high-
yield debt, international trade facilities, foreign
exchange services, treasury management,
investment management, institutional
fxed income and equity sales, interest rate,
commodity and equity risk management,
online/electronic products, insurance, and
investment banking services (Wells Fargo, 2007).
Wells Fargo Financial This segment consists of consumer fnance and
auto fnance operations. It also provides credit
cards and lease, and other commercial fnancing
services; as of March 24, 2007, this segment
provided its services through approximately
6000 branches (Wells Fargo, 2007).
Table 8. Recapitulation of Citigroup’s Management Summary from Most Recent Form 10-Qs.
Date of Filing Organization Key Performance Metric Highlights Comments
August 3, 2007 Citigroup Income from continuing operations rose 18% to $6.226 billion and was the highest
ever recorded by the bank. Diluted EPS from continuing operations was also up
18%. Revenues were a record $26.6 billion, an increase of 20% over the previous
year to date, led by Markets & Banking, up 33%. The bank’s international operations
recorded revenue growth of 34% in the quarter, with International Consumer up
16%, International Markets & Banking up 50%, and International Global Wealth
Management more than doubling. U.S. Consumer revenues grew 3%, while
Alternative Investments revenues increased an impressive 77%.
Acquisitions represented approximately 4% of the
bank’s revenue growth.
August 7, 2007 Bank of America At June 30, 2007, the Corporation had $1.5 trillion in assets and approximately
196 thousand full-time equivalent employees. Net interest income on a FTE basis
decreased $145 million to $8.8 billion and $588 million to $17.4 billion for the three
and six months ended June 30, 2007 compared to the same periods in 2006. The
primary drivers of the decreases were the impact of the divestitures of certain foreign
operations in 2006 and the frst quarter of 2007, increased hedge costs, higher
cost of deposits, spread compression, reduced benefts from purchase accounting
adjustments and the negative impact of the adoption of FSP 13-2. These decreases
were partially ofset by a higher contribution from market-based activity, higher
levels of consumer and commercial domestic loans and increased ALM portfolio
levels.
In July 2007, the Corporation completed the acquisition
of U.S. Trust Corporation (U.S. Trust) for $3.3 billion in
cash.
The company reports that the competitive environment
varies for its far-fung operations. For example, the
bank’s most recent Form 10-Q fling reports that, “The
fnancial services industry is highly competitive. Our
subsidiaries compete with fnancial services providers,
such as banks, savings and loan associations, credit
unions, fnance companies, mortgage banking companies,
insurance companies, and money market and mutual fund
companies” (March 1, 2007 p. 3). Beyond this broad range
of competitors, the bank’s subsidiaries were also confronted
with a number of challenges on some new fronts as well:
“They also face increased competition from nonbank
institutions such as brokerage houses and insurance
companies, as well as from fnancial services subsidiaries
of commercial and manufacturing companies. Many of
these competitors enjoy fewer regulatory constraints and
some may have lower cost structures” (Form 10-Q, March
1, 2007 p. 3).
Recapitulation of Management Summaries from Most
Recent Form 10-Qs.
A recapitulation of the respective fve leading bank
holding companies’ management summaries from their
most recent Form 10-Q flings with the SEC and comments
are provided in Table 8 below.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
220 Business Intelligence Journal July
Date of Filing Organization Key Performance Metric Highlights Comments
August 9, 2007 JP Morgan The company reports $1.5 trillion in assets, $119.2 billion in stockholders’ equity and
operations worldwide. The company reported 2007 second-quarter Net income of
$4.2 billion, or $1.20 per share, compared with Net income of $3.5 billion, or $0.99
per share, for the second quarter of 2006. Return on common equity for the quarter
was 14% compared with 13% in the prior year.
Net income for the frst six months of 2007 was $9.0 billion, or $2.55 per share,
compared with $6.6 billion, or $1.85 per share, in the comparable period last year.
Return on common equity was 16% for the frst six months of 2007 compared with
12% for the prior-year period.
The second quarter of 2007 economic environment
was a contributing factor to the performance of the
Firm and each of its businesses. The overall economic
expansion, strong level of capital markets activity and
positive performance in equity markets helped to
drive new business volume and organic growth within
each of the company’s wholesale businesses; however,
weakness in the housing markets resulted in increased
losses in Retail Financial Services, causing in an increase
in provision related to the home equity portfolio.
July 30, 2007 Wachovia
Corporation
Wachovia’s net income in the frst six months of 2007 was $4.6 billion, up 29 percent
from the frst six months of 2006, and diluted earnings per common share were up 7
percent to $2.42. After-tax net merger-related and restructuring expenses amounted
to 1 cent per share in the frst six months of 2007 and 4 cents per share in the same
period of 2006. Tax-equivalent net interest income increased 24 percent in the frst
six months of 2007 from the frst six months of 2006, refecting a larger balance sheet.
The net interest margin declined 22 basis points to 2.97 percent, primarily due to
growth in lower-spread consumer and commercial loans, a shift in deposits toward
lower-spread categories, the impact of acquisitions and the efect of an inverted
yield curve.
Wachovia and certain of its subsidiaries are currently
involved in a number of judicial, regulatory and
arbitration proceedings concerning matters arising
from the conduct of its business activities; these
proceedings include actions brought against Wachovia
and/or its subsidiaries with respect to transactions in
which Wachovia and/or its subsidiaries acted as banker,
lender, underwriter, fnancial advisor or broker or in
activities related thereto. The actual costs of resolving
legal claims may be substantially higher or lower than
the amounts reserved for those claims.
March 1, 2007 Wells Fargo At December 31, 2006, the company reports assets of $482 billion, loans of $319
billion, deposits of $310 billion and stockholders’ equity of $46 billion; based on
total assets, WFC is the ffth largest bank holding company in the United States. At
December 31, 2006, the bank had 158,000 active, full-time equivalent team members.
The fling reports that the bank enjoyed record earnings in 2006 with record diluted
earnings per share of $2.49, record net income of $8.5 billion, both up 11%, and
exceptional, broad-based performance across the company’s 80+ businesses. The
report also emphasizes that, “Over the past twenty years, our annual compound
growth rate in earnings per share was 14% and our annual compound growth rate
in revenue was 12%. Our total annual compound stockholder return of 14% the past
fve years was more than double the S&P 500 — and at 15% almost double for the
past ten years. We far out-paced the S&P 500 the past 15 and 20 years with total
annual compound shareholder returns of 18% and 21%, respectively — periods with
almost every economic cycle and economic condition a fnancial institution can
experience” (p. 34).
All common share and per share disclosures in this
Report refect the two-for-one stock split in the form of
a 100% stock dividend distributed on August 11, 2006.
Sources: As indicated.
Stock Performance of Top Five Bank Holding
Companies in the U.S.: Past Five Years to Date.
A comparison of the top fve bank holding companies’
stock performance for the past fve years to date is provided
in Figure 7 below.
Figure 7. Stock Performance of Top Five Bank Holding
Companies in the U.S.: Past Five Years to Date
Source: Yahoo! Finance, 2007.
Key:
C = Citigroup Inc.
BAC = Bank of America
JPM = JPMorgan Chase & Co.
WB = Wachovia
WFC = Wells Fargo & Company
Methodology
Description of the Study Approach
This study used a mixed methodology to achieve the
above-stated research purpose. The frst component of the
research methodology consisted of a comprehensive review
of the relevant peer-reviewed, scholarly and organizational
literature. This approach is congruent with various social
researchers who report that such a literature review is the
natural place to begin such an endeavor (Neuman, 2003;
Wood and Ellis, 2003). According to one authority, “Both
the opinions of experts in the feld and other research
studies are of interest. Such reading is referred to as a
review of the literature” (Fraenkel and Wallen, 2001 p.
48). Similarly, Gratton and Jones (2003) emphasize that
2011 221
Likewise, Silverman (2005, p. 300) suggests that a
literature review should aim to answer the following
questions:
1. What do we know about the topic?
2. What do we have to say critically about what is already
known?
3. Has anyone else ever done anything exactly the same?
4. Has anyone else done anything that is related?
5. Where does your work ft in with what has gone before?
6. Why is your research worth doing in the light of what
has already been done?
It was with the above goals and guidance that the review
of the relevant peer-reviewed, scholarly and organizational
data was conducted. Both university and public libraries
were consulted for this purpose, as well as a range of
online resources including EBSCO, Questia, Edgar Online,
Yahoo! Finance and others.
The econometric approach used in the study is also
congruent with a number of economic researchers who
emphasize that it is common practice for economists to
describe or discuss a theory in terms of an equation or a
set of equations; moreover, even elementary economics
textbooks present postulated relationships between
economic variables in an algebraic form, and suggest
inferences by mathematical manipulations (Charemza,
Deadman & Elgar, 1997).
Therefore, extending the process to include assignation
of quantitative measures to these relationships represents a
logical step. According to Charemza and his colleagues, “The
most widely used tool of economists to determine empirical
forms of theoretical constructs is that of econometrics.
The likely originator of the term 'econometrics' defned
it as '...the unifcation of economic theory, statistics, and
mathematics...' Much of the early empirical work in
economics (in other words, prior to 1940) was concerned
with the measurement of demand elasticities, and the
representation of the business cycle” (p. 1).
To a great extent, these trends were refective of the
activity of economists in developing theory in these areas,
and the increasing availability of reasonably long runs
of statistical data on agricultural commodities, foreign
trade and various industries (Charemza et al., 1997).
Subsequent developments of national income accounting
in conjunction with Keynesian economic theory further
created new opportunities for the econometric analysis
of macroeconomic series, including complete models of
economies; these estimated macroeconomic models could
be used for economic policy purposes, such as forecasting
or simulation (Charemza et al., 1997).
Regardless of the objectives being considered, it is
clear that in order to undertake econometric analysis, the
following must be used:
1. A relevant economic theory;
2. Statistical data;
3. A method that allows for the expression of the economic
theory using the statistical data (in practice, a theory
of estimation stemming from econometric theory)
(six properties that would be considered desirable
in an estimated model: (a) relevance, (b) simplicity,
(c) theoretical plausibility, (d) explanatory ability, (e)
a critical reviewing of the timely literature is an essential
component of almost any type of research endeavor: “No
matter how original you think the research question may
be,” they advise, “it is almost certain that your work will
be building on the work of others. It is here that the review
of such existing work is important. A literature review is
the background to the research, where it is important to
demonstrate a clear understanding of the relevant theories
and concepts, the results of past research into the area, the
types of methodologies and research designs employed in
such research, and areas where the literature is defcient”
(p. 51).
In addition, Wood and Ellis (2003) identifed the
following as important outcomes of a well conducted
literature review:
1. It helps describe a topic of interest and refne either
research questions or directions in which to look;
2. It presents a clear description and evaluation of the
theories and concepts that have informed research into
the topic of interest;
3. It clarifes the relationship to previous research and
highlights where new research may contribute by
identifying research possibilities which have been
overlooked so far in the literature;
4. It provides insights into the topic of interest that are
both methodological and substantive;
5. It demonstrates powers of critical analysis by, for
instance, exposing taken for granted assumptions
underpinning previous research and identifying
the possibilities of replacing them with alternative
assumptions;
6. It justifes any new research through a coherent critique
of what has gone before and demonstrates why new
research is both timely and important.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
222 Business Intelligence Journal July
accuracy of coeffcients and (f) forecasting ability; a
“good” model will display all of these properties to
some degree, but the existence of a potentially large
number of theoretically plausible models which also
satisfy some or all of the criteria makes the model
choice problem a nontrivial one in practice;
4. A “know-how,” which guides how to apply the
estimation theory to the statistical data, and how to
decide whether this application has been successful; in
this case, the requisite “know hows” relate to methods
of defning “good” models, and in fnding them
(Charemza et al., 1997, p. 3).
Based on the foregoing guidance, this study used a
weighted average approach to determining the impact of
the external national economy on the fve leading bank’s
proftability according to their total assets. The econometric
model described further in Chapter 4 below that was
developed for this study was deemed to have satisfed all of
these criteria to some extent.
Data-Gathering Method and Database of
Study
Beyond the critical review of the relevant peer-reviewed,
scholarly and organizational literature concerning
proftability determinants in the banking industry today, this
study also used bank-level data for the top fve banks in the
United States for last fve years for its econometric analysis,
as well as fve years’ of data from various international
fnancial statistics for the U.S. banking industry in general.
As noted above, these data were collected from reliable
online resources such as the U.S. Federal Reserve System’s
National Information Center, Free Edgar, Yahoo! Finance
and comparable organizational and governmental Web
sites.
Data Analysis
Econometric Model.
In their study, “Proftability of Credit Unions,
Commercial Banks and Savings Banks: A Comparative
Analysis,” Kaushik and Lopez (1996) report that,
“Proftability is the measure of both performance of each of
the industries and the degree of competition among them”
(p. 66). On the one hand, a straightforward approach to
defning competitiveness at any industry level is in terms
of the percentage of market shares held domestically and
internationally (Saunders and Walter, 1998). While this is
fairly simple task when it comes to industries such as steel
or automobiles that provide comparatively homogeneous
goods and services, such an application to the fnancial
services industry is far more complicated because of
the wide range of mixes and national settings involved
(Saunders and Walter, 1998).
Moreover, defning competitiveness in these terms may
not be especially refective of an individual organization’s
proftability, which depends as well on a variety of market
structure and other cost factors such as scale and scope
economies (Saunders and Walter, 1998). On the other hand,
using market share is a commonly used indicator of the
outcome of the competitive process itself because in free
markets, companies that enjoy the lowest cost structure or
feature highest product quality generally gain market share
compared to those with a higher cost structure or lower
product quality (Saunders and Walter, 1998).
In terms of applying this method to the fnancial services
industry, the picture becomes more complex. For instance,
Saunders and Walter emphasize that, “The fnancial services
industry sells perhaps 50 more or less distinct services to
perhaps 20 more or less distinct client groups. They range
from credit card loans to lower middle income households
all the way to swap-driven repackaged synthetic securities
engineered for the largest multinational companies. Some
of these services are highly internationalized, indeed
globalized, while others are individually sold or mass-
marketed domestically” (p. 24). Further complicating any
such across-the-board analysis is the paucity of market share
data; in some cases, this information is reasonably available;
however, in other cases, there are no available market share
data whatsoever (Saunders and Walter, 1998). Moreover,
some information is also obscured or contaminated by
exchange rate changes that take place during the periods
of time that data are available to researchers (Saunders and
Walter, 1998).
Based on the foregoing constraints, a better indicator
of proftability was required for the purposes of this
investigation. As noted above, the empirical test used in
this study is concerned with the determinants proftability
of these fve leading bank holding companies in the United
States as of June 2007. For this purpose, the measure of
proftability of each bank was defned as the return on assets
(ROA); the ROA is a ratio that is calculated by dividing
the net income over total assets. The macro-economic
variable GDP per capita growth was also used in the model
as estimated by the CIA World Factbook for 2006 and the
International Monetary Fund for the years 2004 and 2005.
2011 223
The GDP per capita growth rate was expected to have
a positive impact on bank’s performance according to the
well documented literature on the association between
economic growth and fnancial sector performance by
virtue of economies of scale that accrue to larger banking
organizations whose assets run into the billions. This is
congruent with the fndings of Berger et al. (2007) who
report, “Early research on bank cost scale economies
using data on U.S. banks from the 1980s generally fnds
very little scale economies or diseconomies except at very
small sizes, typically well under $1 billion in assets. Later
research suggests that there may be more extensive cost
scale economies in the 1990s, with average costs declining
up to asset sizes of $25 billion or more” (p. 331).
The rationale in support of this model was found
extensively throughout the body of evidence concerning
relevant proftability indicators for the banking industry
that clearly demonstrated economies of scale for larger
fnancial institutions, particularly as they compete across
international borders, notwithstanding any substantive
changes in the external operating environment to the
contrary. To gauge the soundness of this rationale, the
degree of proftability for each bank holding company
was determined by using a return on investment weighted
average approach. In this regard, Li, Pan and Tse (1999)
report that, “Proftability is measured by returns on assets
(ROA)” (p. 81). According to the editors of MoneyTerms,
“A weighted average is more heavily infuenced by some
of the numbers it is calculated from than others. It is
calculated by multiplying each number by a weight, adding
these together and then dividing the total by the sum of the
weights” (Weighted Average, 2007 p. 3). The formula for
this model is shown below:
ROA (net income/total assets) * GDP per capita growth
/ (ROA*GDP per capita growth/ROA+GDP per capita
growth)
The return on assets calculations internal aspect for the
fve respective banks compared with the GDP per capital
growth external factor rate is provided in Tables 9 through
13 below, as well as graphic representations of the data in
the accompanying fgures; the percentage differential is
also shown.
Table 9. Bank of America: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 21,133,000 16,465,000 14,143,000
Total Assets 1,459,737,000 1,291,803,000 1,110,457,000
Return on Assets
(ROA)
1.45% 1.27% 1.27%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 1.00% 0.95% 0.98%
Percentage
Diference ROA-
Weighted Avg.
0.45% 0.33% 0.30%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 10. JP Morgan: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 14,444,000 8,483,000 4,466,000
Total Assets 1,351,520,000 1,198,942,000 1,157,248,000
Return on Assets
(ROA)
1.07% 0.71% 0.39%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 0.80% 0.59% 0.35%
Percentage
Diference ROA-
Weighted Avg.
0.27% 0.11% 0.03%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
224 Business Intelligence Journal July
Table 11. Citigroup: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 21,538,000 24,589,000 17,046,000
Total Assets 1,884,318,000 1,494,037,000 1,484,101,000
Return on Assets
(ROA)
1.14% 1.65% 1.15%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 0.84% 1.14% 0.90%
Percentage
Diference ROA-
Weighted Avg.
0.30% 0.51% 0.25%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 12. Wachovia: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 7,791,000 6,643,000 5,214,000
Total Assets 707,121,000 520,755,000 493,324,000
Return on
Assets (ROA)
1.10% 1.28% 1.06%
Per Capita
Growth Rate*
3.20% 3.70% 4.20%
Weighted
Average
0.82% 0.95% 0.84%
Percentage
Diference
ROA-Weighted
Avg.
0.28% 0.33% 0.21%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 13. Wells Fargo & Co.: December 31-December 2006
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 8,482,000 7,671,000 7,014,000
Total Assets 481,996,000 481,741,000 427,849,000
Return on Assets
(ROA)
1.76% 1.59% 1.64%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 1.14% 1.11% 1.18%
Percentage
Diference ROA-
Weighted Avg.
0.62% 0.48% 0.46%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 14. Comparison of Percentage Differences in ROA-
Weighted Averages.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Bank of America 0.45% 0.33% 0.30%
JP Morgan 0.27% 0.11% 0.03%
Citigroup 0.30% 0.51% 0.25%
Wachovia 0.28% 0.33% 0.21%
Wells Fargo 0.62% 0.48% 0.46%
0.00%
0.10%
0.20%
0.30%
0.40%
0.50%
0.60%
0.70%
2004 2005 2006
Bank of America JP Morgan Citigroup Wachovia Wells Fargo
2011 225
Summary and Conclusions
Summary
The purpose of this study was to develop an appropriate
econometric model whereby the primary determinants of
proftability of the top fve bank holding companies in the
United States could be examined and understood. To this
end, the research showed that proftability determinants for
the banking industry include positive relationship between
the return of equity and capital to asset ratio as well as the
annual percentage changes in the external per capita income.
There was also a virtual consensus identifed concerning
the effect that the internal factor of size as measured by
an organization’s total assets had on its ability to compete
more effectively, even in times of economic downturns.
This relationship was clearly identifed in the data
analysis, wherein all fve of the leading bank holding
companies in the United States (as of June 2007) enjoyed
increases in their weighted averages of return on assets
despite decreases in the GDP per capita rate. While there
were some fuctuations and variations identifed between
the respective organizations in spite of their relative sizes,
these differences could be accounted for by virtue of specifc
economic initiatives in place during these isolated points in
time that are not accounted for in the analytical model.
Some of the constraints identifed during the research
process included the fact that the majority of research to
date has investigated determinants of proftability at the
bank and/or industry level, with the choice of variables
used failing to provide internal consistency in some cases
and a paucity of timely research concerning the potential
infuence of the macroeconomic environment in which
fnancial services companies compete. There were also
a broad range of unique fscal activities that took place
during the time period examined that clearly infuenced a
given bank’s proftability, either positively or negatively,
over the short-term while failing to provide any indication
of its potential long-term impact on the bank’s proftability.
In the fnal analysis, it would appear that the industry
analysts and experts were absolutely correct in their
assertions that although it is possible to develop a model that
can provide researchers with an indication of the relative
importance of internal and external factors on a company’s
proftability, the analysis is rife with opportunities to miss
important yet unforeseeable infuences that may contribute
to changes in short-term proftability while leaving long-term
proftability unaffected. Likewise, there are unforeseeable
factors concerning innovations in technology that relate
both to a fnancial services company’s internal and external
environment that will depend on the effectiveness of the
bank’s management to add value or not. Responsiveness
to the external economic environment and potential threats
to far-fung operations based on terrorist activities abroad
represent yet more factors that may be highly signifcant,
but are diffcult to model accurately.
Conclusions
The studies on the determinants of bank’s performance
in the United States in recent years have shown some
mixed results, with some researchers fnding that little
cost saving can be achieved by increasing the size of the
banking frm and others report signifcant scale economies
for banks whose asset size extends well into the billion
range such as those investigated herein. In addition to being
extremely diffcult to measure on an international level,
determinants of proftability may be skewed by an inability
to obtain accurate and timely data and by an unbalanced
competitive environment in which a given bank may be
forced to compete with smaller players that can market
high proft products more effciently. In fact, it would be
reasonable to conclude that the determinants of proftability
can be discerned at a given point in time as they relate to
a specifc set of factors, but there is much more involved
in the analysis than a straight-forward modeling approach
can address. Nevertheless, the research was absolutely
consistent in emphasizing the need to gauge proftability
and to identify what revenue sources provide fnancial
services companies with the biggest “bang for the buck”
so they can focus their energies where they will do the
most good both in the short term and for the long haul. In
conclusion, analyzing proftability in the fnancial services
sector requires constructing a comprehensive picture of a
wide range of relevant performance metrics. This analysis
also requires a signifcant amount of judiciousness and
recognition of their limitations when interpreting them.
References
Abreu, M. and V. Mendes. (2002). “Commercial bank
interest margins and proftability: evidence from
E.U countries.” Porto Working paper series in Ben
Naceur, 2003.
Angbazo, L. (1997). “Commercial bank net interest
margins, default risk, interest-rate risk, and
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
226 Business Intelligence Journal July
off-balance sheet banking.” Journal of Banking and
Finance, 21, 55-87 in Ben Naceur, 2003.
Athanasoglou, Panayiotis P., Sophocles N. Brissimis
and Matthaios D. Delis. (2005, June). “ Ba nk-
Specifc, Industry-Specifc and Macroeconomic
Determinants of Bank Proftability.” MPRA
Paper 153, University Library of Munich,
Germany, revised 2006. [Online]. Available:http://www.bankofgreece.gr/publications/pdf/
Paper200525.pdf.
Bank of American Form 10-Q. (2007, August 7). Edgar
Online. [Online]. Available: ht t p: / / yahoo.
brand.edgar-online.com/ fetchFilingFrameset.
aspx?dcn=0001193125-07-
175583&Type=HTML.
Bank of America. (2007). Yahoo! Finance Company
Profiles. [Online]. Available: http:// finance.
yahoo.com/q/pr?s=BAC.
Barajas, A., R. Steiner, and N. Salazar. 1999. “Interest
spreads in banking in Colombia 1974- 1 9 9 6 . ”
IMF Staff Papers, 46, 196-224.
Ben Naceur, Samy. (2003, October). “The
Determinants of the Tunisian Banking Industry
Proftability: Panel Evidence.” Université Libre de
Tunis: Department of Finance.
Ben Naceur Samy and M. Goaied. (2001). “The
determinants of the Tunisian deposit banks’
performance.” Applied Financial Economics, 11,
317-19 in Ben Naceur, 2003.
Berger, Allen N. (1995). “The relationship between
capital and earnings in banking.” Journal o f
Money, Credit and Banking, 27, 404-31 in Ben
Naceur, 2003.
Berger, Allen N. and R. Deyoung. (2006). Technological
progress and the geographic expansion of the
banking industry. Journal of Money, Credit &
Banking, 38(6), 1483.
Berger, Allen N., Astrid A. Dick, Lawrence G.
Goldberg and Lawrence J. White. (2007).
“Competition from Large, Multimarket Firms and
the Performance of Small, Single- Market Firms:
Evidence from the Banking Industry.” Journal of
Money, Credit & Banking, 39(2-3), 331.
Bourke, P. (1989). “Concentration and other
determinants of bank proftability in Europe, North
America and Australia.” Journal of Banking and
Finance 13, 65-79 in Athanasoglou et a l . ,
2005.
Brealey, Richard A., Alastair Clark, Charles Goodhart,
Juliette Healey, Glenn Hoggarth et al. ( 2001) .
Financial Stability and Central Banks: A Global
Perspective. London: Routledge.
Charemza, W. W. and D. F. Deadman. (1997). New
Directions in Econometric Practice: General
to Specifc Modelling, Cointegration, and Vector
Autoregression. Lyme, NH: Edward Elgar.
Citigroup, Inc. (2007). Yahoo! Finance Company
Profles. [Online]. Available:http://fnance.
yahoo.com/q/pr?s=C.
Cover, Jerry. (1999). “Proftability Analysis
- A Necessary Tool for Success in the 21st
Century.” ABA Banking Journal, 91(2), 78.
Demirguc-Kunt, A. and H. Huizinga. (2000). Financial
Structure and Bank Profitability. World B a n k
Mimeo in Athanasoglou et al., 2005
Fraenkel, J. R. and N. E. Wallen. (2001). Educational
Research: A Guide to the Process.
Mahwah, NJ: Lawrence Erlbaum Associates.
Frieder, Larry A. (1991). “Determinants of Bank
Acquisition Premiums: Issues and Evidence.”
Contemporary Policy Issues, 9(2), 14.
Goddard, J., P. Molyneux and J. O. S. Wilson. (2004a).
“Dynamics of Growth and Profitability i n
Banking.” Journal of Money, Credit & Banking,
36(6), 1069.
--- (2004b). “The proftability of European banks: a
cross-sectional and dynamic panel
analysis.” Manchester School, 72(3), 363-381.
Gratton, C. and I. Jones. (2003). Research Methods for
Sport Studies. New York: Routledge.
Grigorian, David A. and Vlad Manole. (2006).
“Determinants of Commercial Bank Performance
in Transition: An Application of Data Envelopment
Analysis.” Comparative Economic Studies,
48(3), 497.
Gup, B. E. (2003). The Future of Banking. Westport,
CT: Quorum Books.
Guru B., J. Staunton and Balashanmugam. 2002.
“Determinants of commercial bank proftability
in Malaysia.” University Multimedia working
papers in Ben Naceur, 2003.
JP Morgan. (2007). Yahoo! Finance Company Profles.
[Online]. Available: http:// finance.yahoo.
com/q/pr?s=JPM.
Kaushik, Surendra K. and Raymond H. Lopez. (1996)
“Profitability of Credit Unions, Co mme r c i a l
2011 227
Banks and Savings Banks: A Comparative
Analysis.” American Economist, 40(1), 66.
Li, Shaomin, Yigang Pan and David K. Tse. (1999).
“The Impact of Order and Mode of Market
Entry on Proftability and Market Share.” Journal
of International Business Studies, 30(1), 81.
Mogel, L. (2002). Making It in Public Relations:
An Insider’s Guide to Career Opportunities.
Mahwah, NJ: Lawrence Erlbaum Associates.
Molyneux, P. and J. Thornton. (1992). “Determinants
of European bank profitability: A note. Jour nal
of Banking and Finance, 16, 1173-1178 in
Athanasoglou et al., 2005.
“Weighted Average.” (2007). MoneyTerms.
Co. [Online]. Available: moneyterms.co.uk/
glossary-p/.
Neeley, M. C. and D. C. Wheelock. (1997, March/
April). “Why Does Bank Performance Vary
Across States?” Federal Reserve Bank of St. Louis
Review, 79(2), 27-40.
Neuman, W. Lawrence. (2003). Social Research
Methods: Qualitative and Quantitative Approaches,
5th ed. New York: Allyn & Bacon.
Prendergast, C. (1993, March). “The Provision of
Incentives in Firms.” Journal of Economic
Literature, 7, 63 in Brealey et al., 2001.
Saunders, Anthony and Ingo Walter. (1998). Universal
Banking in the United States: What C o u l d
We Gain? What Could We Lose? New York:
Oxford University Press.
Short, B. K. (1979). “The relation between commercial
bank profit rates and banking concent rat i on
in Canada, Western Europe and Japan.” Journal of
Banking and Finance 3, 209-219.
Silverman, D. (2005). Doing Qualitative Research: A
Practical Handbook (2nd ed.). London: Sage.
Spindler, J. Andrew, Jonathan T. B. Howe, and David
F. Dedyo. (1990, May). “The Performance
of Internationally Active Banks and Securities Firms
Based on Conventional C o mp e t i t i v e n e s s
Measures”. Federal Reserve Bank of New York,
Mimeo in Saunders and Walter, 1998.
Stiroh, Kevin J. (2004). “Diversifcation in Banking:
Is Noninterest Income the Answer?” Jour nal
of Money, Credit & Banking, 36(5), 853.
Toyne, Michael F. and James D. Tripp. (1998).
“Interstate Bank Mergers and Their Impact on
Shareholder Returns: Evidence from the 1990s.”
Quarterly Journal of Business and Economics,
37(4), 48.
Wachovia Corp. (2007). Yahoo! Finance Company
Profles. [Online]. Available:http://fnance.
yahoo.com/q/pr?s=WB.
Wells Fargo & Co. (2007). Yahoo! Finance Company
Profiles. [Online]. Available: http:// finance.
yahoo.com/q/pr?s=WFC.
Wood, G. D. and R. C. Ellis. (2003). “Risk management
practices of leading UK cost cons ul t ant s . ”
Engineering, Construction and Architectural
Management, 10(4), 254-62.
Yahoo! Finance. (2007). [Online]. Available: http://
fnance.yahoo.com/.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
228 Business Intelligence Journal July
United States' Largest Bank Holding Companies (as of June 2007)
Appendix A
Rank Institution Name (RSSD ID) Location Total Assets
1 CITIGROUP INC. (1951350) NEW YORK, NY $2,220,866,000
2 BANK OF AMERICA CORPORATION (1073757) CHARLOTTE, NC $1,535,684,280
3 JPMORGAN CHASE & CO. (1039502) NEW YORK, NY $1,458,042,000
4 WACHOVIA CORPORATION (1073551) CHARLOTTE, NC $719,922,000
5 TAUNUS CORPORATION (2816906) NEW YORK, NY $579,062,000
6 WELLS FARGO & COMPANY (1120754) SAN FRANCISCO, CA $539,865,000
7 HSBC NORTH AMERICA HOLDINGS INC. (3232316) PROSPECT HEIGHTS, IL $483,630,057
8 U.S. BANCORP (1119794) MINNEAPOLIS, MN $222,530,000
9 SUNTRUST BANKS, INC. (1131787) ATLANTA, GA $180,314,372
10 ABN AMRO NORTH AMERICA HOLDING COMPANY (1379552) CHICAGO, IL $160,341,966
11 CITIZENS FINANCIAL GROUP, INC. (1132449) PROVIDENCE, RI $159,392,731
12 CAPITAL ONE FINANCIAL CORPORATION (2277860) MCLEAN, VA $145,937,957
13 NATIONAL CITY CORPORATION (1069125) CLEVELAND, OH $140,648,168
14 REGIONS FINANCIAL CORPORATION (3242838) BIRMINGHAM, AL $137,624,205
15 BB&T CORPORATION (1074156) WINSTON-SALEM, NC $127,577,050
16 BANK OF NEW YORK COMPANY, INC., THE (1033470) NEW YORK, NY $126,457,000
17 PNC FINANCIAL SERVICES GROUP, INC., THE (1069778) PITTSBURGH, PA $125,736,711
18 STATE STREET CORPORATION (1111435) BOSTON, MA $112,345,777
19 FIFTH THIRD BANCORP (1070345) CINCINNATI, OH $101,389,721
20 KEYCORP (1068025) CLEVELAND, OH $93,490,903
21 BANCWEST CORPORATION (1025608) HONOLULU, HI $70,661,335
22 HARRIS FINANCIAL CORP. (1245415) WILMINGTON, DE $64,475,903
23 NORTHERN TRUST CORPORATION (1199611) CHICAGO, IL $59,609,734
24 COMERICA INCORPORATED (1199844) DETROIT, MI $58,945,727
25 MARSHALL & ILSLEY CORPORATION (1199497) MILWAUKEE, WI $58,327,527
26 M&T BANK CORPORATION (1037003) BUFFALO, NY $57,869,069
27 UNIONBANCAL CORPORATION (1378434) SAN FRANCISCO, CA $53,173,833
28 CHARLES SCHWAB CORPORATION, THE (1026632) SAN FRANCISCO, CA $49,003,812
29 ZIONS BANCORPORATION (1027004) SALT LAKE CITY, UT $48,703,130
30 COMMERCE BANCORP, INC. (1117679) CHERRY HILL, NJ $48,231,325
31 POPULAR, INC. (1129382) SAN JUAN, PR $46,985,000
32 MELLON FINANCIAL CORPORATION (1068762) PITTSBURGH, PA $43,389,057
33 TD BANKNORTH INC. (1249196) PORTLAND, ME $42,981,084
34 FIRST HORIZON NATIONAL CORPORATION (1094640) MEMPHIS, TN $38,395,825
35 HUNTINGTON BANCSHARES INCORPORATED (1068191) COLUMBUS, OH $36,422,081
36 COMPASS BANCSHARES, INC. (1078529) BIRMINGHAM, AL $34,938,942
37 SYNOVUS FINANCIAL CORP. (1078846) COLUMBUS, GA $33,295,823
38 NEW YORK COMMUNITY BANCORP, INC. (2132932) WESTBURY, NY $29,638,404
39 RBC CENTURA BANKS, INC. (1826056) RALEIGH, NC $25,374,678
40 COLONIAL BANCGROUP, INC., THE (1080465) MONTGOMERY, AL $23,823,484
41 ASSOCIATED BANC-CORP (1199563) GREEN BAY, WI $20,849,531
2011 229
Rank Institution Name (RSSD ID) Location Total Assets
42 BOK FINANCIAL CORPORATION (1883693) TULSA, OK $19,363,601
43 W HOLDING COMPANY, INC. (2801546) MAYAGUEZ, PR $17,894,049
44 FIRST BANCORP (2744894) SAN JUAN, PR $17,596,317
45 WEBSTER FINANCIAL CORPORATION (1145476) WATERBURY, CT $16,964,451
46 SKY FINANCIAL GROUP, INC. (1071203) BOWLING GREEN, OH $16,807,287
47 FIRST CITIZENS BANCSHARES, INC. (1075612) RALEIGH, NC $16,012,041
48 COMMERCE BANCSHARES, INC. (1049341) KANSAS CITY, MO $15,531,107
49 NEW YORK PRIVATE BANK & TRUST CORPORATION (3212091) NEW YORK, NY $15,095,466
50 FULTON FINANCIAL CORPORATION (1117129) LANCASTER, PA $15,078,415
Appendix B
Excel Spreadsheet Results of Data Analysis
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Bank of America
Net Income 21,133,000 16,465,000 14,143,000
Total Assets 1,459,737,000 1,291,803,000 1,110,457,000
Return on Assets (ROA) 1.45% 1.27% 1.27%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 1.00% 0.95% 0.98%
Percentage Diference ROA-Weighted Avg. 0.45% 0.33% 0.30%
JPMorgan
Net Income 14,444,000 8,483,000 4,466,000
Total Assets 1,351,520,000 1,198,942,000 1,157,248,000
Return on Assets (ROA) 1.07% 0.71% 0.39%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 0.80% 0.59% 0.35%
Percentage Diference ROA-Weighted Avg. 0.27% 0.11% 0.03%
Citigroup
Net Income 21,538,000 24,589,000 17,046,000
Total Assets 1,884,318,000 1,494,037,000 1,484,101,000
Return on Assets (ROA) 1.14% 1.65% 1.15%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 0.84% 1.14% 0.90%
Percentage Diference ROA-Weighted Avg. 0.30% 0.51% 0.25%
Wachovia
Net Income 7,791,000 6,643,000 5,214,000
Total Assets 707,121,000 520,755,000 493,324,000
Return on Assets (ROA) 1.10% 1.28% 1.06%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 0.82% 0.95% 0.84%
Percentage Diference ROA-Weighted Avg. 0.28% 0.33% 0.21%
Wells Fargo & Co
Net Income 8,482,000 7,671,000 7,014,000
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
230 Business Intelligence Journal July
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Total Assets 481,996,000 481,741,000 427,849,000
Return on Assets (ROA) 1.76% 1.59% 1.64%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 1.14% 1.11% 1.18%
Percentage Diference ROA-Weighted Avg. 0.62% 0.48% 0.46%
*Based on 2006 estimates from CIA World Factbook and bar graph data from International Monetary Fund
2011 231
THE STRATEGIC IMPACT OF THE BUSINESS DYNAMICS
IN EMERGING COUNTRIES ON CONTEMPORARY
PERSPECTIVES
Walter Gerard Amedzro St-Hilaire, Ph.D
Researcher on Strategy and Governance of Public Organizations in HEC Montréal
Researcher for the Research Center on the Governance of Natural Resources
Email: [email protected]
Abstract
The current competitive environment requires countries to make strategic choices regarding international trade which are not without
impact on their value regards to positioning. Therefore, it becomes important for business owners and for analysts to assess the impact
of policy decisions on global competitiveness and wonder if becoming a future “Giant” does not pass through a choice of sector for
export growth. Also how is the comparative advantage of the current Asian Dragon evaluated in the various sectors that underpin its
economy? In this paper, we propose a both strategic and economic analysis case of the new global competitive order based on the concept
of comparative advantage to fnd that China, by taking advantage of all its assets, has built a model which is typical while integrating
the new realities of contemporary global trade, occupying a prominent place in the global economic landscape. The present article goes
beyond this Asian dragon’s rather negative image to highlight its success in adapting to the reconceptualization of economic exchanges.
Key words: Strategy, competitiveness, added value, economic analysis, economic development.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Marking the 2000s by its ascension into world trade
1
,
and after surpassing, in 2008, the German level in
manufactured exports, China saw, in 2009, its share in total
world trade of goods exceeding the Germany
2
. From 1998
to 2008, the total Chinese share in world exports grew at
23% per year, twice the growth rate of total world exports.
If the trend continues, exports from China could represent
a quarter of the global total in 2018, surpassing the U.S.
in the 1950s that didn’t reach this fgure since totalizing
at the time, only 18% of world exports (as against 8% in
2009). However, the possibility of a decline of the Chinese
exports growth rate remains. According to IMF forecasts,
these exports are expected to represent 12% of world total
in 2010. Worth 9.6% in 2009, this share corresponded to
that of Japan at its peak, is 1986
3
. In addition, the IMF, in
a study published in 2009, indicated that a Chinese GDP
growth rate remaining at or above 8% and a growth still
heavily dependent on exports would mechanically raise this
dragon’s share in world exports to 17% in 2020. However,
the crisis of 2008-9 caused the share of exports in China’s
GDP to drop from 36% in 2007 to 24% in 2009.
In this regard, it is worth asking whether the future
growth of Chinese exports wouldn’t come from its exports
of high added value products (computers and vehicles)
rather than textile. How comparative advantage in this
country is evaluated the various sectors that underpin its
economy?
To address this issue of International Affairs, it will be
conducted a comparative analysis of this dragon with the
rest of the world by sector, and with the assumption that
it will, as regards the export growth, take the same way as
Japan that gradually exported products of increasingly high
added value. It will, in this article, be traced the concept
of “comparative advantage”, to then bring out a stark and
pertinent analysis of the results recorded by the Chinese
trade during the last decade.
Conceptual Framework
For this paper purposes, we are led to identify one
key concept necessary in understanding our theme: the
concept of comparative advantage. It is the ability of an
economic agent (individual, company, country) to create
a product or service for a lower opportunity cost than
1
Figures on China's place in global trade given below are from the article "China's
export prospects, Fear of the Dragon", The Economist, January 7, 2010.
2
In 1997, China was only ..... the 10th largest exporter! See the table of the main
exporters in 1997.
3
Since then, Japan’s share fell to 5%, partly because of the rising yen.
Business Intelligence Journal - July, 2011 Vol.4 No.2
232 Business Intelligence Journal July
that borne by other economic agents. For Paul Krugman,
it’s especially the ability to grasp the difference between
comparative advantage and absolute advantage that defnes
an economist
4
. If comparative advantage should not be
confused with absolute advantage, which is a special case of
comparative advantage, it should not be with the concepts
of free trade (international trade without tariffs and other
artifcial restrictions to trade), the gains from trade, or the
international division of labor which are distinct, though
related to that of comparative advantage. The latter is,
with returns to scale, the main determinant of international
trade
5
.
Although the idea of comparative advantage was exposed
by Robert TORRENS in 1815
6
, its authorship is attributed
to David Ricardo (1817), whose best known proof could be
found in his Principles of political economy and taxation.
Taking the example of England and Portugal, both of able
to produce wine and cloth, he shows that the comparison
of the production costs of wine to cloth (or cloth to wine)
in each country simultaneously reveals the comparative
advantage of each country. The ratio of production costs
is also known as relative cost, opportunity cost, or even
relative productivity. The two basic models of comparative
advantage are Heckscher-Ohlin’s and Ricardo’s.
The Heckscher-Ohlin-Samuelson model
Named after its three principal architects, the Swedish
economists Eli Heckscher Filip (1879-1952) and Bertil
Ohlin (1899-1979) and American economist Paul Samuelson
(1915-2009), the Heckscher-Ohlin-Samuelson model is
frequently referred to as the “HOS model”. But the name
4
" If there were an Economist’s Creed, it would surely contain the affrmations «
I understand the Principle of Comparative Advantage » and « I advocate Free
Trade ». For one hundred seventy years, the appreciation that international trade
benefts a country whether it is “fair” or not has been one of the touchstones of
professionalism in economics. Comparative advantage is not just an idea both
simple and profound: it is an idea that conficts directly with both stubborn
popular prejudices and powerful interests. This combination makes the defence
of free trade as close to a sacred tenet as any idea in economics” - Paul KRUG-
MAN, "Is Free Trade Passé?" by Paul R. KRUGMAN, Journal of Economic
Perspectives, vol.1, no.2, 1987, pp.131-44.
5
I like to begin classes on international trade by telling students that there are
two basic explanations of international trade. The frst is comparative advan-
tage, which says that countries trade to take advantage of their differences – a
concept that lay at the heart of Alan Deardorff’s beautiful, classic paper “The
general validity of the law of comparative advantage” (1980). The second is in-
creasing returns, which says that countries trade to take advantage of the inher-
ent advantages of specialization, which allows large-scale production – which
is what the “new trade theory” was all about" - Paul KRUGMAN, “Increasing
Returns in a Comparative Advantage World”, novembre 2009, 15 pages.
6
But the issue is controversial [see, for example, the article by Roy RUFFIN,
Debunking a Myth: Torrens on Comparative advantage]
of Wassily Leontief (1909-1999) should also be associated
with this model because the empirical test that he has done.
The study of this model requires, initially, understanding the
functioning of a “Heckscher-Ohlinian” economy in a closed
economy or “autarky”, and then comparing it to another
economy that is identical in all respects except for relative
factor endowments. This comparison aims to highlight the
direct dependence of the concept of comparative advantage
to both countries’ factor endowments: it is the heart of the
Heckscher-Ohlin “theorem”.
Indeed, stating that comparative advantage is determined
by differences in relative endowments between countries,
the HOS theorem is a direct consequence of the Rybczynski
theorem which allowed us to establish, under the HOS
model with two goods and two factors, that the higher the
capital-labor ratio of a country, the higher the supply curve
is shifted to the right. The proof of this theorem involves
the comparison of the autarkic equilibrium of two countries
with identical technologies and different production relative
factor endowments.
Let two countries, h, home country, and f, foreign
country when the country h has a capital-labor ratio higher
than the country f’s (k > k*)
7
. The h’s relative supply curve
is more right than f’s while their relative demand curves
merge, the two countries having the same utility function. It
follows that in absence of international trade, at equilibrium
of supply and demand of each country, the relative price of
good 1 is lower in country f than h while good 2 has lower
relative price in h than in f. In other words, the country h
has a comparative advantage in the production of good 2
while f harbors a comparative advantage in the production
of good 1.
It should be noted that, in this model, comparative
advantage is explained neither by differences in technology
(since, by defnition, production technologies are identical
from one country to another) nor by differences in taste
(preferences for goods are identical). The only reason why
relative prices before exchange are different lays in the fact
that h has a capital-labor ratio greater than f, which gives it
a comparative advantage in producing good 2. Conversely,
f has a higher labor-capital ratio than h, which gives it a
comparative advantage in producing good 1.
In sum, in our example, where k > k* and where, by
hypothesis
8
, production of good 2 requires relatively more
7
To distinguish the country f from the country h, an asterisk is added to the sym-
bols on the country f.
8
This hypothesis, like other HOS model assumptions is discussed in more details
athttp://www.mazerolle.fr/Economie-internationale/Glossaire-economie-inter-
nationale/Modele-2X2-standard-Hypotheses-centrales.pdf
2011 233
capital than labor, the Heckscher-Ohlin theorem can be
stated as follows: the country h, relatively abundant capital
(k > k*) has a comparative advantage in production of good
2 (since the production of good 2 uses capital relatively
intensively). As for the country f, relatively abundant
in labor, it has a comparative advantage in production
of good 1 (because the production of good 1 uses the
work of relatively intensively). It follows that at autarkic
equilibrium, the country h produces relatively more of
good 2 than f (conversely, the country f produces relatively
more of good 1 than h). Openness to trade will cause h
to specialize in production and export of good 2 and f’s
specialization in production and export of good 1. As this is
a model of just two countries, exports of h will go to f and
f’s exports will go to h.
Optimality of Free Trade in the Ricardo Model
On another hand, the RICARDO model bases itself
on usual concepts of microeconomic theory to show that
the specialization of each country in the production and
export of goods for which it has a comparative advantage,
improves well-being in each country. For this, it shows
that the international exchange based on the exploitation of
comparative advantage improves the purchasing power of
wages in both countries. Indeed, the comparison between
autarky and free trade shows the overall superiority of
the purchasing power of wages in conditions of free trade
relative regards to the purchasing power of wages in the
autarkic economy.
For example, RICARDO considers two countries,
namely England and Portugal, and two goods, wine and
cloth, from which he evaluates the purchasing power of the
hourly wage for these two products in autarkic economy,
then where free trade occurs. In England, the hourly wage
yields 0.01 hl of wine for free trade instead of 0.008 hl
in autarky. Regarding cloth, the purchasing power of
wages is the same in autarky and free trade (0.01 m2). In
Portugal, the hourly wage yields 0.0125 m2 of cloth in
free trade instead of 0.011 m2 in autarky. As regards wine,
the purchasing power of wages is the same in autarky and
free trade (0.0125 hl). We can therefore conclude that free
trade improves both the English and Portuguese overall
purchasing power of wages.
The Theory of International Trade
In regard to the model that was previously shown, it is
important to understand the concept of exchange, including
international exchange, distinct from that of comparative
advantage, though linked to it. Based on the concept of
exchange among economic agents, the positive theory of
international trade of David Ricardo, developed mainly
in the microeconomics feld, is rooted in the ideas of
neoclassical economists, particularly in the theory of general
equilibrium including consideration of the phenomenon
of exchange without production, called “pure trade” and
the exchange with production. Insofar as the exchange
becomes international, it fts into the broader framework
of international economic relations. International Economy
could be seen as an observable reality that should be
described from a statistical point of view and facts: the
economic exchanges between the nations are steadily
growing and being an essential component of globalization,
while as trade policy of countries. In this latter sense, it
is often the terms “international trade” or “international
trade”, which are used.
Methodology: Content Analysis Linked
to the Index of Competitiveness of the
World Economic Forum
The study of countries’ growth factors will be based
on an analysis of the competitiveness index published
annually by the World Economic Forum in Davos. This
index is a score assigned to different countries varying, in
2009, between 2.58 (Burundi) and 5.6 (Switzerland), from
which countries are ranked. This classifcation is meant to
refect the competitiveness of countries, which is evaluated
by 12 criteria (the 12 pillars of competitivity
9
) defned by
the WEF and also by the interaction of these criteria allow
in fact distinguishing three categories of economies.
The Three Stages of Development According
to the Wef
It must indeed be noted that the 12 factors or pillars
of competitiveness defned by the WEF interact and even
mutually reinforce themselves. For example, innovation
(12th pillar) is inconceivable in a world where powerful
institutions (Pillar 1) are absent, because starting innovation
requires that a protection by copyright IP. Similarly, if the
workforce is not educated (5th pillar), or the different
9
See the notice "Les 12 piliers de la compétitivité selon le World Economic Fo-
rum",http://www.mazerolle.fr/Economie-internationale/Glossaire-economie-
internationale/12-piliers-de-la-competitivite-selon-le-World-Economic-Forum.
pdf
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
234 Business Intelligence Journal July
An Emerging Giant in Global
Competitiveness
According to the WEF latest report of (Global
Competitiveness Report 2009-10), this Asian giant’s Global
Competitiveness Index 2009-10 is 4.74 out of 7, giving it
the 29th position out of 133 countries, an honorable and
progressive position. Below is China’s ranking with more
details on each of the 12 pillars:
Represented by a so-called “spider” graph and
compared with the average performance of category 2
countries (economies whose competitiveness is determined
by effciency), this ranking shows that the dragons’ country
does as well or better, especially as regards the market size
and macroeconomic stability.
markets not effcient (pillars 6, 7 and 8) or there is no
infrastructure (pillar 2), innovation will not be possible
or will be very limited. From these interactions emerges a
classifcation of countries according to factors that infuence
their competitiveness. These economic classes defne the
countries’ development level, and by extension the role
of the 12 pillars. Thus, improving the competitiveness of
Chad does not go through the same channels as in Japan
or the United States, these countries not being at the same
stage of development. As the latter evolves, the level of
wage increases, and thus increases the income per head
of population, hence the need to improve productivity to
maintain the living standards. Three types of economies
can be distinguished, including economies whose
competitiveness is under the infuence of their resources in
production factors, those whose competitiveness is under
the infuence of effciency and those whose competitiveness
is infuenced by innovation.
• Economies whose competitiveness is determined by
their resource inputs. These are economies whose
competitiveness is sustained by their resources
in production factors, mainly unskilled labor and
natural resources. Companies in these countries
sell commodities, manufactured by unskilled labor
and therefore not well paid. The maintenance
and enhancement of competitiveness at this stage
of development essentially requires the proper
functioning of public and private institutions (pillar 1),
the effciency of infrastructure (pillar 2), the stability of
the macroeconomic environment (pillar 3) and healthy
workforces a basic education (pillar 4).
• Economies whose competitiveness is determined by
their effciency. As wages rise, as production methods
become more complex and product quality improves,
competitiveness is mainly governed by the workforce
qualifcation and vocational training (pillar 5), effcient
goods and services market (pillar 6), labor market
(Pillar 7) and capital markets (pillar 8), and by the
domestic and / or export market size (pillar 10) and
technological agility (pillar 9).
• Economies whose competitiveness is determined by
innovation. The economies that fall into this category
are those where wages are high, even very high, as well
as living standards, and where the main competitive
factors become innovation (pillar 12), which takes
the form of knowing how to make new and different
products, as well as a dense and advanced organization
intra and inter frms (pillar 11).
Result and Analysis
Still ranked by the WEF in 2008-9 in an intermediate
stage between stages 1 and 2, according to the GDP per
capita criterion (in current dollars not in PPP dollars), China
is now ranked among countries of group 2 with its GDP per
capita in current dollars over $ 3,000.
GDP threshold for membership in a stage of development
Development stage GDP per capita (in current dollars)
Stage 1: Economies whose
competitiveness is determined by
their resourse inputs
17000 dollars
2011 235
However, economic development does not take place
smoothly. In this regard, the chart below highlights the
most problematic aspects of business:
Competitive Advantage and Competitive
Disadvantage
The small square appearing next to the China’s rank
for each indicator shows a situation of either competitive
advantage (dark blue) or competitive disadvantage (light
gray). For determining the nature of competitiveness (the
color of the square), the WEF proceed in accordance to the
position of countries in the global CGI index. Indeed, those
classifed in the top 10 have a competitive advantage for an
indicator when it also has a rank lower than 10. Otherwise,
it is considered a competitive disadvantage. Furthermore,
countries at ranks higher than or equal to 11 (it’s the
case of the Asian dragon that is 29
th
) have a competitive
advantage for an indicator when its rank is lower than the
country’s global ranking, otherwise there is a competitive
disadvantage for the indicator. Indicators ranked below 29
have therefore a dark blue square to indicate the competitive
advantage of the dragons’ country.
Some Examples of Expanding Industries
If the Asian giant is an emerging economy, relatively
new to the world trading scene, it does however have
many assets to shine on it and make the longer-established
countries worry. Indeed, expanding in many sectors, the frst
country in the world in terms of market size is participating
in reshaping world trade, particularly by cutting itself a
special place.
The Aviation Industry
Until the late 2000s, with the U.S. McDONNELL
DOUGLAS that disappeared in 1997, and the Brazilian
EMBRAER, BOMBARDIER the Canadian, Russian and
Japanese Mitsubishi SUKHOI limited to regional aircraft
and having no real infuence on neither of the Toulouse and
Seattle giants’ strategies, airliners manufacturers competing
with Boeing and Airbus were rare. However, AVIC, the
newcomer from the land of dragons, supported by this
land’s 11th Five-Year Plan (2006-2010), could change the
deal, and make the two global giants worry, especially by
the planned launch of a medium-haul with 150 seats, thus
competing with the two giants much earlier than expected.
Next to AVIC, Bombardier, with its CSeries, also
becomes a potential competitor:
“Annoncé pour une capacité standard de 130 sièges, il
pourrait ainsi faire de l’ombre à l’A318, le plus petit des
Airbus, dès 2013. Sukhoï, avec son nouveau Superjet et
Embraer avec son E-190 affchent également leur ambition
sur ce segment de marché. Mais la principale menace
est le futur appareil chinois C919, dont le premier vol est
annoncé pour 2014. D’une capacité de 170 à 190 sièges, ce
futur rival de l’A320 et du B737 pourra lui aussi compter
sur le dernier cri en matière de réacteurs et d’équipements.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
236 Business Intelligence Journal July
Bien que très jeune, l’industrie aéronautique chinoise
a en effet tous les atouts. Outre ses centaines de milliers
d’ingénieurs, elle peut compter sur les moyens fnanciers
colossaux de l’Etat et sur un marché national en plein
boom, susceptible d’absorber quelque 2.800 avions neufs
d’ici à 2026. Aucun de ses concurrents n’a autant de cartes
en main. Bombardier l’a d’ailleurs bien compris, qui a
conclu un partenariat stratégique avec l’industrie chinoise,
lui offrant son expertise et son implantation internationale
pour le C919 en échange d’un coup de pouce fnancier
pour son Cseries”
10
. It should also be noted that with the
recent installation in Tianjin of the A320 frst assembly line
outside Europe, Airbus aims to take a part in the largest
market of world
11
.
Satellite Positioning System
In addition to the aviation industry, the Asian giant
cuts itself a place in the satellite space by launching the
building of its own system, called “Beidou”, whose release
is scheduled for 2015. Direct competitor of the European
Galileo project, it has more than 30 satellites, fve of
which have already been installed. Concerned, Europe,
which has fallen behind in the Galileo project, denounced
the fact that China is currently pre-empting the frequency
allocated to the European Union by the International
Telecommunication Union. Indeed, China has published
frequencies and signals that overwrap those of Galileo.
However, the strongest signal “crushing” the others, some
of Galileo signals could be jammed by the Chinese satellites
and made non-functional, especially the encrypted signals
reserved among others to Defense
12
.
High Speed Trains
In terms of land transport, land of dragons is setting
up travel at high speed. Indeed, although the French
Alstom denied it the transfer of its technology, the Chinese
TGV was nevertheless established, namely the Wuhan-
Guangzhou line (departure from the new Guangzhou
Railway Station), where circulate Nippon and German
13
See « La Chine s’affrme comme un acteur majeur de la grande vitesse ferrovi-
aire », Les Echos, June 21, 2010.
14
See “Deux oiseaux dans les télécoms chinoises”, by Philippe ESCANDE, Les
Echos, April 7, 2010.
designed TGV(derived from the Japanese Shinkansen and
the German Velaro) reaching peaks of 350 km / h (cons
320 maximum in France). He still benefted from the SNCF
expertise in station construction. Now, China intends to
sell its know-how on TGV to Saudi Arabia and also plans
to build, by 2020, 45,000 km of high-speed lines, widely
outdoing the French 1000km of high speed lines
13
.
Mobile Phone Industry
Regarding the mobile sector, it is in February 2010 that
the Asian giant passed the milestone of one billion users
with a mobile operator market controlled entirely by local
businesses, which do well on the OEM market too.
An article in “Les Echos” focuses on the two giants
China Telecom, ZTE and Huawei
14
that, like many others,
are typical of the rise of the Asian Dragon’s economy and,
more signifcantly, follows the curve of its evolution. In the
early 1980s, Ren Zhengfei, , a researcher in the laboratories
of the army in Beijing and founder of Huawei, and Hou
Weigui, ZTE’s founder, working in Xian in a research
center in electronics, adopted a similar strategy to win :
“Face à un marché balbutiant en Chine et des acteurs
européens comme Alcatel, Siemens, Nokia ou Ericsson,
qui trustent les relations avec les grandes sociétés d’État,
les deux nouveaux venus se sont lancés à l’assaut des
campagnes chinoises totalement sous-équipées, avec
une arme, l’engouement pour le téléphone mobile. Et un
argument massue, le prix. Quand le fournisseur européen
vendait son matériel de 200 à 300 dollars la ligne, Huawei
et ZTE séduisaient les PTT de troisième zone avec leurs
matériels trois fois moins chers”.
They then repeated this low cost strategy in Pakistan,
Burma and Africa, with this incredible feature, inspired by
the postwar Marshall Plan, i.e. fnance the purchase of their
own products with the help of the Chinese Government to
these countries, and now they are entering India, the old
Europe and the strong America. Indeed, something to give
cold sweat to Alcatel that, a decade ago, reigned supreme
but must now be content with the bare minimum:
“Ainsi, lors de son dernier appel d’offres en décembre
dernier pour son nouveau réseau au protocole Internet, le
premier opérateur du pays, China Mobile, a en quelque
sorte livré une photo du nouvel équilibre des forces. Selon
10
Bruno TREVIDIC, Le troisième avionneur mondial sera-t-il chinois ? Les
Echos, December 7, 2009.
11
Seehttp://www.mazerolle.fr/Economie-internationale/Glossaire-economie-in-
ternationale/IDE-francais-en-Chine-et- chinois-en-france.pdf
12
But Beijing was involved in the Galileo project to the tune of 65 million euros
and a portion of the work was entrusted to Chinese companies. Despite this
cooperation, the Chinese government has developed a competing system.
2011 237
15
The price-to-book ratio, or P/B ratio, is a fnancial ratio used to compare a com-
pany’s book value to its current market price. Book value is an accounting term
denoting the portion of the company held by the shareholders; in other words,
the company’s total tangible assets less its total liabilities.
16
Clive COOKSON, China scientists lead world in research growth, Financial
Times, January 25, 2010.
les analystes d’iSuppli, il aurait accordé 35% du marché
à ZTE, autant à Huawei, 10% à Alcatel- Lucent et 5 % à
Ericsson. On comprend mieux les problèmes d’Alcatel, qui
dix ans avant était le leader incontesté de ce marché”.
Banking
The fnancial sector does not fall behind, given the vast
expansion of banks in the country of dragons since the
2000s. As a proof, in 2010, four of the top fve banks in the
world are Chinese according to their book to market ratio
15
.
Scientific Research
Regarding the scientifc research, the Asian giant is
experiencing a continuous boom since the 1980s, 1990s and
2000s, ranking second in the world behind the United States,
according to the Financial Times
16
. Although the indicators
of this development are not necessarily very illuminating
and reliable, the newspaper bases itself on the research
articles published in scientifc journals criterion to make
this classifcation. Since 1981, the number of publications
has been multiplied by 64 reaching, in 2008, no fewer than
112,318 articles published by Chinese researchers (against
332,916 in the U.S.).
The Financial Times quoting James Wilsdon, director of
the Science Policy Centre of the Royal Society, shows three
factors thought to be behind this growth, namely the huge
public investment in education and research at all research
levels, in both schools and universities, the fast and well
organized circulation and dissemination of scientifc
information and the set up by the Chinese government of
very attractive opportunities for researchers in the Diaspora,
allowing them to teach in their country, while continuing to
spend part of their time abroad.
Steelmaking
Concerning heavy industry, steel production, considered
as an indicator of the power of a national industry, because
of the use of this material in the composition of a large
number of industrial products, brand the supremacy of the
Asian giant, by far the largest producer of steel. Moreover,
the Chinese Hebei Steel and Bao Steel are two of the fve
largest producers of steel in the world (see graphic).
Multiple Assets: Acquisitions, Counterfeiting,
Espionage and Authoritarian Planning ... But
Also Tourism Counterfeiting, the Cybersitter
Case Example
To reach the level of the greatest in areas where
technology is usually a hurdle, China does not hesitate
to use the counterfeit, fast and inexpensive, to build
its competitiveness in high technology industries. The
example of the Green Youth Escort Dam, complex
software tracking keywords to prevent any connections
to websites deemed politically sensitive
17
is typical of the
use of counterfeiting at the State’s highest level. In 2009,
imposed on all manufacturers of computers working on the
world’s largest market (since June 2009, the installation is
no longer mandatory but only recommended), this software
is actually derived from the piracy of a parental control
software belonging to American CYBERSITTER, based
in Santa Barbara, California. The company attacked, in
17
The Green Dam completes the censoring system already imposed on the Internet
in this country. It prevents access to websites reporting on the repression of the
Tiananmen place riots, the Tibet crisis or the Falungong organization activities.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
238 Business Intelligence Journal July
Taken not long ago as nasty polluters, as gravediggers
for the Copenhagen Summit of December 2009, the Asian
giant is now poised to overtake the objectives of the Summit.
In fact, while in France the carbon tax, is sacrifced for
economic growth, the land of dragons carved a very large
place on the wind turbine market. Furthermore, world’s
fourth solar panel manufacturer, the company Yingli Solar
produces good quality panels for a price 30% cheaper,
bought by the biggest companies in the world, including
GDF Suez, which has ordered 145,000 of them to build a
large solar power plant in southern France. We must note
that four of the top 10 solar panels makers are Chinese:
Suntech (2nd), Yingli (4th), Trina Solar (6th) and Solarium
(8th).
It should be noted, however, that the largest population
of the world remains one of the biggest polluters in the
world, although progressing in the right direction while
many others, such as France, although prone to criticism,
go down. And, indeed it was in June 2010 that Beijing
launched its subsidy plan for electric cars in fve test
cities
18
. However, it is likely that these subsidies are not
18
Dans un programme d’aides comparable à ceux expérimentés en Occident et au
Japon, le ministère des Finances a annoncé qu’il allait offrir aux habitants
des villes de Shanghai, Shenzhen, Hangzhou, Hefei et Changchun un rabais
allant jusqu’à 50000 yuans , soit 5800 euros lors de l’achat d’une automobile
hybride rechargeable et de60000 yuans (7200 euros) pour chaque inves-
tissement dans un véhicule tout électrique». This targeted strategy aims to help
local manufacturers, which are located differently in these fve cities: SAIC in
Shanghai, BYD is in Shenzhen, Geely is in Hangzhou, Chery is in Hefei and
FAW in Changchun.
suffcient for the demand of electric cars to take off if they
are not accompanied by massive public investment in
infrastructure adequate for their use (battery management
terminals or loading points, or battery exchange stands).
Goods and Services
Comprehensive and systematic data on world trade are
known about 18 months apart from the current date. In
2009, for example, fgures for 2008 are published around
the middle of the year in the World Trade Report 2009 of
the WTO. The frst data to be published are those of the
classifcation of countries according to their total exports
and imports. An early circulation of data is possible, but
it carries the risk of a wrong recycling of data from the
previous year exploiting the difference between the date of
publication and the time of data.
Goods International Trade In 2009
After studying the Chinese economy through its business
activities, it is important to consider it on a bigger feld:
International trade. It requires noting that the landscape of
global trade depends on how the EU-27 is processed in the
distribution of trade fows, because of the centrality of the
EU. Thus, should be considered the case in which the main
goods exporters and importers include the EU countries as
individual entities, then the case of the main goods exporters
and importers excluding intra-EU trade. In the table below,
each country of the EU-27 is treated separately. Several
interesting points emerge
19
:
• In 2009, China became the world’s largest exporter,
overtaking Germany.
• Three major powers are clearly distinguishable from
others, as regards both exports and imports: China,
Germany and the United States.
• Regarding China, it should be noted that a signifcant
portion of Chinese exports are actually produced by
Western and Japanese multinational frms operating in
the largest market in the world.
• The role of France in international trade, though
modest, is signifcant mainly as a consumer-importer
(6th exporter, importer 4th)
• The bulk of world exports and imports is the result of a
small number of countries.
19
Source: WTO, World Trade Report 2010
An extraordinary and unexpected development in the
environmental sector
Five keys to understand the Solar Sector
• Germany represented more than half of the world´s photovoltaic systems
in 2009.
• Down last year, the global market should have its growth back in 2010,
with an increase of the sales volume worth at least 40% according to
experts.
• The actual production capacities (around 17 Gig watts) are already able
to cover market´s needs till 2012-2013, according to PwC.
• U.S. and China are considered as the next growth relay of this sector.
• This sector employs 5500 people in France. According to the renewable
energies Union, it could have around 15000 employees the year 2012.
U.S. court, (the communist government of the country of
dragons totally controlling the local judiciary) seven PC
manufacturers and two local designers for hacking 3000
code lines. This has not been to slow the Asian giant, which
plans to export its product.
2011 239
• The total of world exports and imports do not coincide
due to signifcant statistical error margins.
Dealing instead with the EU as a whole and treats
the intra-European as intra-regional trade, such as trade
between California and Texas in the United States, or
between Guangdong and Sichuan in China, draws a quite
different picture
20
:
• The EU-25 then appears as the leading exporter (unless
you add China, Hong Kong and Taiwan).
• The U.S. remains the largest importer before the EU
and the Asian giant.
• The disappearance of trade between EU countries
“reduces” the total world trade and therefore increases
the share of major countries. For example, the dragons’
country’s share is 11.8% (excluding Hong Kong and
Taiwan).
• In both cases, we note that:
• Countries like India and Russia are also major emerging
economies; however, they remain far behind the world
three major goods exporters and importers.
• Japan remains a major exporting power, and it is more
apparent if we look at the chart excluding intra-EU trade,
but it was surpassed by China in a decade, refecting
the speed of the transformations characterizing global
economy.
The Dominance of Intra-Regional Over Inter-
Regional Trade
Like the EU, “the regional blocs” are very important in
the assessment of international trade, since, in general, the
largest share of world trade takes place within the blocks,
especially within the EU.
In addition to goods, there are, on the international
market, a sizeable fow of other types of properties, less
tangible than the frst, namely services. It is therefore
important to understand well the services international
trade, given their increasingly important role in an economy
more and more globalized. In the Balance of Payments
Statistics, the various parts of the current account are titled
respectively: Goods, Services (including services provided
or received by the government), Income (investment
income and wages) and Current transfers (operations
without consideration).
20
Idem
Not existing as such, the commercial services category,
defned by the WTO “as equal to services minus those
provided or received by public administrations”
21
, is
divided into three categories: transportation, travel and
other commercial services.
Transportation category. Covering all transport
services (sea, air and others, including land and inland water
transport, through space and pipeline) offered by residents
of an economy to those of another, it concerns transporting
passengers, goods carriage (freight), rental (charters) of
carriers with crew and other linked services.
Travel category. Covering goods and services purchased
for personal use by travelers - for health, educational or
other purposes - as well as those who travel for professional
reasons, it does not involve a particular kind of service,
but an assortment of goods and services “consumed” by
travelers. The most common entries in this category are
housing, food and beverages, entertainment, transportation
(within the economy visited), gifts and souvenirs.
Other Commercial Services
• Communication services: they include the
telecommunications, postal services and courier
services. Telecommunication services include the
transmission of sound, images or other information by
telephone, telex, telegram, radio and cable television
broadcasting, satellite, electronic mail, fax etc..,
including communications networks, teleconferencing
and support services, but not the value of the transmitted
information. Are also included mobile telephone
services, Internet services and basic online access
services, including the provision of Internet access;
• Building and public works services: they include work
performed on construction and installations projects by
employees of a company outside the economic territory
(the one-year rule used to determine the resident status
is to be applied with fexibility). Furthermore, the
goods used for these projects are included, involving a
tendency to overestimate these services.
• Insurance services related to various forms of insurance
to non-residents by resident insurance companies and
vice versa, as freight insurance, direct insurance (eg life
insurance) and Reinsurance;
• Computer and information services including data
services (services related to hardware and software and
21
Detailed classifcation source: WTO metadata,http://www.wto.org/french/res_f/
statis_f/its2007_f/its07_metadata_f.pdf pages 161-2.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
240 Business Intelligence Journal July
data processing services), the news agency services
(provision of information, pictures and articles to
the media) and other information services (services
database and web search);
• Royalties and license fees, including payment and
receipts pertaining to the export of non-fnancial
intangible assets and property rights such as patents,
copyrights, trademarks, industrial processes, and
franchises;
• Other services to enterprises, namely trade-related
services, operational leasing (leasing without
operators), and miscellaneous business, professional
and technical services such as legal, accounting and
management consulting, public relations services,
advertising, market research and opinion polling,
research and development services, architectural
services, engineering and other technical services,
agricultural services, mining and processing on site.
• Personal, cultural and recreational services subdivided
into two sub-categories:
• Audiovisual services covering services and commissions
relating to the production of flms, radio and television
broadcasts and musical recordings.
• Other cultural and recreational services which cover,
among other services associated with museums,
libraries, archives and other cultural activities, sports
and recreation.
When providing services to foreign countries is done
through the establishment of a subsidiary or a branch, it
is excluded from international trade fows measured by
the balance of payments, hence an under-estimation of the
importance of the commercial services provided by foreign
frms to a country. This is highlighted by data on foreign
direct investment in companies operating in the services
sector showing an absorption in this sector of at least 50-
55% of total foreign direct investment with, as a dominant
means of service providing, the establishment of foreign
subsidiaries.
In this regard, marking a major turning point in
international trade, the strategy initiated by Deng in
1978, generally considered to have borne fruit, despite
the signifcant imbalances and the major challenges
(environment, population, income disparities, and
geographic inequalities) to meet in the coming decades,
hard to explain. However, in an article in the Financial
Times, Gideon Rachman writes:
“The background to Deng Xiaoping’s liberalisation of
the Chinese economy in 1978 was a fscal and foreign
22
Bankruptcy could be good for America, Financial Times, January 11, 2010.
23
See Nicolas BAVEREZ, “La bataille des Normes”, Le Monde de l’économie,
April 20, 2010.
exchange crisis. Finding itself desperately short of cash,
the Chinese government was much more willing to embrace
heterodox economic ideas that promised to deliver faster
growth and higher revenues. The rest is history”
22
It would thus be the realization that, during a severe
crisis, the former communist regime was at its loss, which
made the land of dragons to change course in addition to
having a capable man in command.
In 2009, despite criticism of its recovery plan, this
country surprised the world by the effectiveness of its
response to the moderate decline in exports to the United
States and Europe, ensuring a record growth rate of 8.5%,
triggering a good recovery of its exports after a relatively
short period of very strong decrease.
Done in favor of domestic demand, but more towards
the infrastructure demand than in direction of consumer
demand, the recovery plan has given birth to a fear of
overcapacity. In particular, the generosity of banks in
granting loans to enterprises supported by the CPC (on
its request) has generated signifcant growth in the money
supply.
Moreover, the latter continues to grow with the purchase
of dollars from the products exported by the Bank of China
to maintain the fxed parity. However, there is doubt about
the “sterilization” of this currency (i.e. the withdrawal of an
equivalent amount of traffc to avoid excess liquidity) that
is not very likely, given the highly expansionary monetary
policy. The reserve requirement imposed on banks may
well have been raised (and is already among the highest in
the world, approximately 16%), this important liquidity, not
necessarily quite appropriately invested causes imbalances.
Since the entry of the Asian giant in the WTO in 2001
and especially since his claim as world’s top exporter
in 2009, thanks to the global crisis, some no longer
hesitate to speak of the “Beijing Consensus” to describe
the emergence of a new world economic order “fondé
sur la dissociation de la liberté politique et de la liberté
économique, la restriction de l’État de droit, le contrôle de
l’information et de la société, le pilotage du développement
et des investissements par la puissance publique, la gestion
discrétionnaire du crédit et du change”
23
. This term, coined
in 2004 by Joshua Cooper Ramo, is quoted in a recent book
by American scholar Stefan Halper, in its title: The Beijing
Consensus, How China’s Authoritarian Model Will
Dominate the Twenty-First Century.. However, in the land
2011 241
24
Title of an article published in « The Economist », May 6, 2010.
of dragons, where it appears that “The Beijing Consensus
is to keep quiet”
24
, one remains quite discreet about this
concept even hesitant to talk about “a Chinese development
model”, despite various publications such as”China Model:
A New Development Model from the Sixty Years of the
People’s Republic” (November 2009), or “China Model:
Experiences and Diffculties”(January 2010).
Conclusion
At the end of this article about the strategic impact of the
business dynamics in emerging countries on contemporary
perspectives and its questions about International Affairs,
it should be drawn some useful lessons. Indeed, China, the
largest market in the world, occupies a prominent place
in the global economic landscape. Booming, thanks to
the expansion of its various sectors, this Asian giant has
come to upset the world trade with a growing interest in
high technology. Taking advantage of all its assets (a
large population, a strong regulatory culture, etc...) and no
inhibitions (counterfeit) to have a place beside or in front
of the previously dominant economies, this has something
to be feared by those economies, as well in the aviation
industry as in Steelmaking and banking. Beyond its rather
negative image (communist, repressive and polluter), the
Asian dragon has built a model which is typical while
integrating the new realities of contemporary global trade.
While meeting the challenge of modernity, it succeeded in
adapting to the reconceptualization of economic exchanges
which are often defned today, so much broader than simple
barter of property against another, as envisaged in the theory
of exchange. Exchanges of this country with the world in fact
include not only goods and services but also international
capital fows, technology transfers, the movement of
ideas, economic migration, research and development as a
determinant of international competitiveness of enterprises,
etc...: from this point of view it is important not to isolate
this giant, frmly committed to shape its destiny, with too
unilateral visions, partial views but proved by the linear
reality of its winning evolution recursion, because this
country has specifc properties that exceed the global
business environment.
References
SMITH, A., 1776. Recherche sur la nature et les causes
de la richesse des nations. Paris : Gallimard, 1976.
BALASSA, B. 1963. « An Empirical Demonstration
of Classical Comparative Costs Theory», Review
of Economics and Statistics, volume 40, pages 231-
8.
Banque mondiale, Online Atlas of the Millenium
Development Goal,http://devdata.worldbank.org/
atlas-mdg/ CCAVES.
Richard E., FRANKEL, J. A. et JONES, R. W.,
Commerce et paiements internationaux, 9ème
édition.
Banque Mondiale, Rapport sur le développement
dans le monde. Publié chaque année vers le mois
d’octobre.
Banque Mondiale, site internet :http://www.
worldbank.org/
KAWECKI, B. et A. 2000. Économie Internationale.
Paris: Dunod, collection « Travaux dirigés », pages
25-65.
GRUBEL H. G. et LOYD, P.J., 1975, Intra-industry
Trade, the Theory and Measurement of International
Trade in Differentiated Products. Londres :
CIA World Factshttps://www.cia.gov/library/
publications/the-world-factbook/
Classement des 100 premières multinationales
et nombreuses informations utiles : http://
www. u n c t a d . o r g / Te mp l a t e s / St a r t Pa ge .
asp?intItemID=2068&lang=2 D
DEARDORFF, A.V., 1985. « Testing Trade Theories
and Predicting Trade Flows », in R. JONES et
P. KENEN, éditeurs, Handbook of Intemational
Economics, Amsterdam : North Holland. Volume
2, pages 467-518.
Conférence de Nations Unies pour le Commerce et
le Développement (CNUCED) : Publication d’un
rapport annuel sur l’Investissement International et
l’activité des frmes multinationales.
De MELO, J., et GRETHER, J-M., 1997. Commerce
International. Bruxelles : DeBoeck, collection
«Balises», pages 190-238.
Fond Monétaire International,http://www.imf.org/
external/french/index.htm FRIEDMAN,
Douglas A., Against the Tide: An Intellectual History
of Free Trade, 1996.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
242 Business Intelligence Journal July
KRUGMAN, P., 1980. « Scale Economies, Product
Differentiation, and the Pattern of Trade ».
American Economic Review , 70, pages 950-959.
KRUGMAN, P., 1995. « Increasing Returns,
Imperfect Competition and the Positive Theory of
International Trade », in Handbook of International
Economics, volume 3, edited by GROSSMAN
Gene, M. et Kenneth ROGOFF. North Holland.
LANCASTER, K., 1980. « Intra-Industry Trade under
Perfect Monopolistic Competition », Journal of
International Economics, mai 1980, pages 151-75.
LASSUDRIE-DUCHENE B., 1971. « La demande de
différence et l’échange international » , Economies
et sociétés.
LEONTIEF, W., 1956. « Factor Proportions and the
Structure of American Trade: Further Theoretical
and Empirical Analysis», The Review of Economic
and Statistics, novembre, pages 386-- 407.
LINDER, S., 1961. An Essay on Trade and
Transformation. New York, John WILEY and Sons.
LEONTIEF, W., 1954. « Domestic Production and
Foreign Trade the American Capital Position Re-
examined», Economia Internationale, n°1, février,
pages 3-32.
Mac DOUGALL, G.D.A. 1951. «British and American
Exports : A Study Suggested by the Theory of
Comparative Costs, part 1 ». Economic Journal.
Volume 61, décembre, pages 697-724.
Mac DOUGALL, G.D.A. 1952. « British and
American Exports : A Study Suggested by the
Theory of Comparative Costs, part 2». Economic
Journal . Volume 62, septembre, pages 487-521.
HELPMAN, E. et KRUGMAN, P., 1985. Market
Structure and Foreign Trade. MIT Press.
Mathematica Eurostat, page d’accueil du portail
statistique de l’UE Eurostat, page d’accueil des
statistiques de l’UE sur le commerce international.
Mc KENZIE R., «Comparative Advantage», Vidéo
You tube,http://fr.youtube.com/watch?v=aLe_
LF622JI , Mars 2007. O
Organisation de Coopération et de Développement
Economique (OCDE), Portail des échanges
internationaux:http://www.oecd.org/topic/0,3373,
fr_2649_37431_1_1_1_1_37431,00.html
Organisation Mondiale du Commerce, site internet :http://www.wto.org/index.htm
KRUGMAN, P. et OBSTFELD, M., Economie
internationale, 7ème édition.
KENEN, P. B., 1965. « Nature , Capital and Trade »,
Journal of Political Economy, 73, octobre, pages
437-60.
Programme des Nations-Unies pour le Développement,
site internet :http://www.undp.org/french.
RICARDO, D., 1817. Des principes de l’économie
politique et de l’impôt. Paris Flammarion, 1977.
RYBCZYNSKI, T. M., 1955. « Factor Endowments
and Relative Commodity Prices», Economica,
pages 336--41.
SAMUELSON, P. A., 1948, « International Trade and
the Equalization of Factor Prices », The Economic
Journal , juin, pages 181-97.
STOLPER, W. F. et SAMUELSON, P. A., 1941. «
Protection and Real Wages», Review of Economic
Studies, 9, novembre, pages 58-73.
2011 243
E-BANKING PATRONAGE IN NIGERIA: AN EXPLORATORY
STUDY OF GENDER DIFFERENCE
Dr. Asikhia Olalekan
Senior Lecturer, Department of Business Studies
College of Business and Social Sciences,
Covenant University, Ota, Canaan land,
Km.10, Idiroko, P.M.B.1023, Ota,
Ogun state, Nigeria.
Email: [email protected]
Abstract
Although some studies had already been done concerning electronic banking yet there has been no study on gender differences
in e-banking patronage. The aim of this study is to determine if gender difference exist in electronic banking patronage in Nigeria.
The major attributes of electronic banking are used as measuring items using likert scale, which is subjected to factor and descriptive
statistical analysis. The study includes a survey on gender patronage of e-banking by individual customers in Abuja and Lagos, Nigeria
conducted in December, 2008. The results of the study indicate gender differences in e-banking patronage and also customer satisfaction
is found to be low in this respect which calls for improvement in service delivery in terms of timeliness and consistency in standards.
Key words: Electronic banking, Patronage, Gender differences, Nigeria.
There has been transformation of the Nigerian banking
industry after the consolidation exercise in 2005 by the
Central Bank of Nigeria (CBN). Akpan (2009) asserts that
maximizing returns and optimizing proftability became
the focus of banks and these can only be achieved through
enhanced patronages; that is increased customer base with
attendant satisfaction suffcient to consolidate loyalty.
The banks are therefore confronted with delivering their
services in the most effcient ways, using electronic means
to deliver additional products and services. Thus, managing
their assets of service delivery to customers became a major
objective.
Electronic banking is the effective deployment of
information technology by banks. Jun and Cai (2001)
identifes three dimensions of service quality in E-banking,
they are banking service product quality, customer service
quality and online systems quality. Bank service product
quality is related to product variety and the diverse features
of the service products. Customer service quality is related
to the differences between customer expectations of service
providers’ performance and their evaluation of the services
they received. Online systems quality is the quality that
the customer perceived as the end-users of the information
system. They further identify the various dimensions of
electronic banking that may be perceived differently by
male and female customers, which may likely affect their
level of patronage; this involved the level of perceived
risks, user input factors, price factors, service product
characteristics, and individual factors.
Recent literature has established that communication,
information sharing, distant in electronic means, and
level of perceived usefulness and innovativeness are
major setbacks in E-banking acceptance and patronage by
customers (Mukherjee and Nath, 2003; Rotchanakitumnuai
and Speece, 2004; and Arne and Horst, 2006) and gender
has been noted in the literature as a factor that affects
customer patronage. Thus studying the gender differences
in e-banking patronage is a useful gap in literature that this
paper intends to fll, since the listed factors that are presently
seen as setbacks for e-banking may be perceived differently
by male and female customers alike. The research addressing
the issue of gender difference in e-banking patronage could
help bankers to fnd better ways of communicating with
both sexes which will guide electronic banking marketing
mix decisions.
This research will contribute to the body of consumer
behaviour literature by examining the interrelationships
among observed variables and subsequently, a model of
interrelationships will be created by means of exploratory
factor analysis.
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
244 Business Intelligence Journal July
Literature review
Numerous studies have provided considerable evidence
that gender relates to customers’ perceptions, attitudes,
preferences and purchase decisions (Slyke, Comunale
&Belanger,2002; Mitchell & Walsh, 2004; Fischer &
Arnold,2004; Bakewell & Mitchell,2006; Kwan,Yeung
and Au, 2008; Wan Omar, Ali, Hussin and Rahim,2009).
Sproles (1985) identifed some items that affect customers’
cognitive and affective orientation towards shopping
activities. These items were later refned by Sproles
and Kendall(1986) and a scale consisting of eight mental
customer style characteristics emerged. These include;
perfectionistic high- quality conscious consumer, brand
conscious “price equals quality” consumer, novelty-fashion
conscious consumer; recreational, hedonistic consumer;
Price conscious “value –for-money” consumer; impulsive,
careless consumer; confused by overchoice consumer, and
fnally habitual, brand-loyal consumer. These characteristics
differ from male to female (Mitchell and Walsh, 2004).
Bakewell and Mitchell (2003) later investigated the
decision-making methods of adult female generation Y
consumers in the UK and discovered fve meaningful and
distinct decision making groups; “recreational quality
seekers “, recreational discount seekers”, “trend setting
loyals”, “shopping and fashion uninterested” and confused
time/money conserving”. Similarly in studying the male
decision making method they found that the male consumers
exhibited all the eight traits earlier outlined by Sproles and
Kendall(1986) and four new traits were identifed namely;
store loyal/low price seeking, time-energy conserving,
confused time restricted and store-promiscuity. This is a
pointer to the point that apart from the fact that some of the
purchase characteristics differ on the basis of sex, they are
also product specifc.
Mitchell and Walsh (2004) compared the decision-
making method/styles of male and female shoppers in
Germany, they found that male individuals were slightly
less likely to be perfectionists, less novelty and fashion
conscious, and less likely to be confused when making
purchases than their female counterparts. In a similar study
by Bakewell and Mitchell (2006) they established that nine
decision-making styles were common to both genders,
and three new male traits; store-loyal/low-price seeking,
confused time-restricted and store promiscuity, and three
female traits; bargain seeking, imperfectionism and store
loyal were also discovered.
Recently Hanzaee and Aghasibeig (2008) established
10-factor style for males and 11-factor styles for females,
nine factor styles were common to both genders, they
concluded that male and female customers differ in their
decision-making styles.
It is obvious that all these studies aligned with the fact
that males and females differ in their decision making styles,
noting that e-banking has its inherent features as discussed
below, could it be possible for males and females disparity
in patronage? This is a gap to be flled by this study.
By accessing banking services from any place and at any
moment, end users can beneft from increased convenience,
simplicity and fastness. Besides banks can reduce their
transaction cost as e-banking is fve times cheaper than
traditional banking ways. They can strengthen their core
business and broaden their customer scope by reaching
valuable customers, selling new fnancial e-services as an
attractive differentiating tool, (Ezio, 2008).
Jun and Cai (2001) identifed bank customers’
perceptions of service quality dimensions using quantitative
techniques. The authors’ conceptualized internet banking
service quality based on three quality perspectives; banking
service product quality, customer service quality and online
systems quality.
Perceived risk is considered an important risk attribute
that impacts on the consumer decision-making process when
buying a product or consuming some services (Mitchell,
1998). Electronic banking is a technology-enabled channel
and consumers’ perceive the use of electronic banking as a
risky decision because technology-enabled services exhibit
invasive technological, unfamiliar and indefnite stimuli
(Davidow, 1986). Therefore, when consumers decide to use
electronic banking, they are exposed to uncertainties such
as the availability, the compatibility, and the performance
of the complementary electronic banking channels (Sarin,
Sego and Chanvarasuth, 2003). The degree to which
individuals accommodate these uncertainties may have
gender implications.
Consumers perceive greater risks when buying
services than tangible goods (Zeithaml, 1981). Services
are perceived as riskier than products because services are
intangible, non-standardized, and regularly sold without
guarantees or warrantees. Consumers can hardly ever
return a service to the service provider since they have
already consumed it, and some services are so technical or
specialized that consumers possess neither the knowledge
nor the experience to evaluate whether they are satisfed,
even after they have consumed the service (Zeithaml,
1981).
Consumer Perceived risks identifed by literature of
electronic banking include;
2011 245
• fnancial risk
• performance risk
• physical risk
• social risk and
• Psychological risk.
Financial risk represents the fnancial loss in using
electronic banking, as consumers may perceive that
reversing a transaction, stopping a payment after discovering
an error, or a refund may not be possible. Performance risk
in electronic banking is less satisfying than non-electronic
banking, as consumers may perceive that electronic banking
cannot be used to complete a transaction when needed due
to the denial of access to their account. Physical risk in
electronic banking refers to possible injury when personal
information is accessed by a third party. Social risk refers to
the older generation who may object to the use of electronic
banking due to their perception that non-electronic banking
is personal and friendly. Psychological risk represents
consumer perceptions that the use of electronic banking
would reduce the self-image of them, or have a negative
effect on their perceived image from other consumers. Time
risk in electronic banking implies that it takes more time
to complete a banking transaction than a non-electronic
banking transaction. Sathye (1999) and Polatoglu &
Ekin(2001) found that the reliability dimension was an
important determinant for consumers who used electronic
banking. Furthermore, Sathye (1999) and Liao & Cheung,
(2002) found that reliability was positively related to the
use of electronic banking. They concluded that the more
secure the consumer perceived electronic banking to be; the
more likely they were to use electronic banking, this can
also vary with sex.
Previous studies have identifed that user input factors
are a function of control, enjoyment and intention to use
(Ng and Palmer, 1999). Control could be described as the
amount of effort and involvement required by consumers
in electronic banking. Enjoyment is the perceived
playfulness and intrinsic value consumers experience from
the utilization of electronic banking. The intention to use
is described as the level of resistance to change, which is
associated with consumers’ intention to change from non-
electronic banking to electronic banking. This may differ
with gender. Gerrard & Cunningham (2003) identifed
that consumers who were more fnancially innovative had
a higher probability of adopting electronic banking than
less fnancially innovative consumers. Similarly, Sathye
(1999) found that even when consumers were aware of the
availability of electronic banking, some consumers might
still not operate this type of banking due to consumers’ low
intention to use electronic banking.
Price factors suggest that perceived relative economic
advantages will motivate consumers to use electronic
banking (Sathye, 1999). For example, consumers using
electronic banking could lower the fxed and variable
costs that are associated with the banking process, due to
reductions in personal error and labour cost savings.
The Report(1997) indicated that for consumers to use
technologies, the price to use technologies needed to be
reasonable when compared to alternatives. Sathye (1999)
argued that, in the context of internet banking, two kinds of
price were accounted for; the normal costs associated with
internet activities, and the bank costs and charges. Polatoglu
and Ekin’s (2001) study further identifed that users of
electronic banking were considerably satisfed with the
cost saving factor through electronic banking. Contrarily,
Sathye (1999) identifed that the costs associated with
electronic banking, such as the cost of electronic banking
activities and bank charges, had a negative effect on
electronic banking adoption. Bakewell and Mitchell (2003)
had earlier found that females tend to be money conserving
than males: this may also have an impact on the level of
patronage of e-banking by males and females.
In general, additional defnite service features, service
specifcations, targets of a service, and the core service
comprised the service product characteristics. The service
product characteristics of electronic banking including:
consumers’ perception of a standard and consistence
service, the time saving feature of electronic banking, and
the absence of personal interactions, have been empirically
found to infuence consumers’ use of electronic banking
which may have gender implications, (Polatoglu & Ekin,
2001 and Karjaluoto, Mattila & Pento, 2002).
The electronic banking literature supports individual
factors such as knowledge (Sathye, 1999), consumer
resources, such as money and information reception and
processing capabilities (Karjaluoto, Mattila and Pento,
2002; Gerrard and Cunningham, 2003), and lifestyle
(Polatoglu and Ekin, 2001) as having impacts on consumers’
adoption of electronic banking. Knowledge refers to the
consumers’ responsiveness of each type of electronic
banking channel in the marketplace, their awareness of
the benefts associated with electronic banking, and their
knowledge of how to utilize electronic banking. The
consumer resource money refers to the ease of access of a
Personal Computer (PC) and the internet. The information
processing and processing capabilities resource is
concerned with consumers’ computer expertise, aptitude of
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
246 Business Intelligence Journal July
internet, and the comprehensibility of electronic banking.
Lifestyle refers to the social life in consumers’ banking
patterns, such as the consumers’ value, the independence
of the electronic banking process, or values relating to the
personal interactions associated with the non-electronic
banking process which may have gender alignment.
Consumers’ knowledge of electronic banking plays
an important role in their use of electronic banking.
Sathye (1999) and Polatoglu and Ekin (2001) empirically
supported the idea that consumer knowledge had an effect
on electronic banking adoption while Sathye (1999) found
that the lack of awareness about electronic banking and
its benefts, including the perception of it (being non-
user friendly) contribute to the non-adoption of electronic
banking. Furthermore, Polatoglu and Ekin (2001) stated
that the more knowledge and skills a consumer possessed
about electronic banking, the easier it was for the consumer
to utilize electronic banking. Colgate, Nguyen and Lee
(2003) coroborated the fact that when consumers made
decisions for different alternatives in the marketplace, the
awareness of the existing alternatives was a determinant
for consumers to stay with their current banking provider
and this is a function of the amount of knowledge they
already possess. The knowledge possessed may vary from
sex to sex depending on the level of commitment to such
endeavour.
Consumer resources also infuence the use of electronic
banking. Mols (1998), Sathye (1999) and Karjaluoto,
Mattila, and Pento’s (2002)’s studies showed that some
consumers lacked access to a personal computer (PC)
and this prohibited the adoption of electronic banking.
Studies have also shown that consumer resources including
computer profciency infuence the consumers’ employment
of electronic banking. Sathye (1999) demonstrated that
consumers described incomprehensibility as a reason for
not using electronic banking. Similarly, Karjaluoto, Mattila,
and Pento’s (2002) literature suggests’ that non-electronic
banking users considered electronic banking as diffcult
to use because they found computers diffcult to operate.
Gerrard and Cunningham (2003) found that consumers
who were non-adopters of electronic banking could be
differentiated by their lower computation profciency and
computer skills. Similarly, Karjaluoto, Mattila, and Pento’s
(2002) empirical results suggested that non-electronic
banking users considered electronic banking as diffcult
to use because they found computers diffcult to operate.
Gerrard and Cunningham (2003) found that consumers
who were non-adopters of electronic banking could be
differentiated by their lower computation profciency and
computer skills. Acquisition of skills may vary with interest
which differs in gender.
Al-Ashban & Burney (2001) identifed white-collar
consumers as being most likely to use electronic banking.
It can be postulated that occupation status (namely white-
collar) is positively related to the choice of electronic
banking. They further showed that as consumers increased
their educational qualifcation level, their adoption of
electronic banking would increase as well. Chan (1997)
established that income was the single most important
variable that infuenced a consumer’s use of a credit
card. Empirical fndings of income positively infuencing
adoption of electronic banking can be found in Al-Ashban
and Burney (2001) and Karjaluoto (2002) studies.
Bank management need to continuously assess the
customers’ decision- making process as well as the
formation of attitudes, preferences and satisfaction of
automated services. It is of little use for banks to attempt
to position an offering to the female gender (for example)
by emphasizing particular attributes that do not constitute
signifcant choice criteria (Delvin,2002).
The Nigerian banking industry has tried to take
advantage of the productivity and customer service gains
that e-banking tend to offer. In spite of the advantages of
these new technologies, many customers are unwilling
to adopt them (Asikhia, 2007). It is therefore important
to understand the customer preferences attitudes and
adaptations of these services particularly in relation with
adoption and such information could be used as marketing
tools to attract new clients and retain the existing clients.
Kolodinsky et al (2000) and Kolodinsky and Hogarth
(2001) studied the adoption of new innovations and services
in banking in US. They concluded that gender differences
existed for phone banking, electronic funds transfer and PC
banking.
It is pertinent to say that evidence abound that male and
female consumers or customers differ in decision making
styles and obvious perceptions about a product or service,
it is believed that e-banking customers in Nigeria may
also have certain distinctive gender characteristics and
alignment in their patronage. This study flls this gap by
studying the differences in decision-making of e-banking
patronage based on gender in the Nigerian context.
E – banking in Nigeria
The competition train in the banking industry has
shifted to technology and intensive delivery of services
which has created paradigm shift in banking services in
2011 247
Nigeria. Banks like Zenith Bank, and Guaranty Trust Bank
used technology as a competitive weapon and successfully
became one of the largest banks in Nigeria within four
years post-consolidation era. Banking service delivery was
re-engineered with the introduction of technology. Several
new products were introduced, for example we have
internet banking, Automated Teller Machines (ATMs),
phone banking/ cell banking, debit cards, credit cards etc.
Integrated Banking Solutions (IBS) was also introduced
in most banks. A central server replaced all branch level
and controlling offce level servers. Every transaction that
is taking place across any service point like a branch or
ATM is recorded in the server. This new operational method
helps customers to conduct normal banking transactions
like deposit/withdrawal of cash from account, account
transfers etc. from any branch of the twenty - fve banks in
Nigeria without the account being domiciled in that branch.
Automated Teller Machines (ATM) are equipped to
facilitate several transactions, customers can withdraw
from the ATM of a bank that they don’t have an account.
Other roles like account transfers, payment of bills, balance
enquiry are all possible through ATM. Over 25,000 ATMs
are currently in use in the country connected to Inter Switch
Network.
Electronic banking is gaining patronage where
information from central server is made accessible to the
account holder using a PIN or PASSWORD. Account
statements, account transfer facilities, bill payment
facilities, list of cheques to be cleared, loan facilities in
form of assets, vehicle and mortgage, international bank
transfer are some of the facilities given by the banks to their
customers. However, recent security threat to customers
savings have made banks to be more careful in their online
transactions, customers are alerted from time to time on the
secrecy of their password. All banks often communicate
relevant information through electronic mail (e-mail) or
short messages on phone to their customers. Banks also
have their websites to promote their products and services
as well as publicize information that is relevant to the public
about the bank.
Credit and debit cards are gaining patronage; many
Nigerian banks have started providing these services.
Master and VISA are the most preferred companies with
which banks make arrangements for providing credit/debit
cards to clients. Many banks are also having their own
debit cards on which the Nigerian currency could be loaded
for transactions depending on the nature of the account.
If the account is a domiciliary account, then the foreign
currency could be loaded on it. And it is essentially used
to operate the account that the customer is having with the
bank, this has become important in the light of the fact that
most Nigerian banks now have branches in other parts of
the world particularly in Britain and most West African
countries. Account balances are now being sent through
phone and alerts are also sent when any money is deposited,
transferred or withdrawn from an account. Virtual banks
are on the horizon in Nigeria.
Method
The research instrument was self-administered, it consist
of 29- items e-banking characteristics features of service
quality (5 items), perceived risk factors (5items), user input
factors(5 items), price and service product characteristics
factor(5 items), individual factors(5 items) and customer
satisfaction factors (4 items). All scales are measured on
a 7-point Likert –type scales ranging from 1(very strongly
disagree) to 7 (very strongly agree). The reliabilities of
the items ranged from 0.79 to 0.97 as shown in Table
1.0. Order effects were reduced by random ordering of
items, and the research instrument asked questions to also
draw information on the demographic background of the
respondents. The questionnaire was self-administered to a
non-probability sample of 750 male and 750 female making
a total of 1500 questionnaires that were distributed randomly
amongst bank customers in Lagos and Abuja because these
cities are believed to have a near representation of all tribes
and professional spheres in Nigeria (Nigeria Bureau of
Statistics, 2007). The concentration in these two cities is
likely to reduce error within the measurement model being
infated by situational factors inherent in diverse samples
from the thirty six states of the federation. The banks used
in each of the location were selected through convenience
sampling method; customers of banks close to researcher’s
location within these cities were used. A total of 375 male
and 375 female customers were sampled randomly from the
two locations (Abuja and Lagos) and the eventual spread
of the customers according to banks and geographical
location is presented in appendix 1. Of 1500 questionnaires
distributed, one thousand and sixty questionnaires (1060)
were deemed usable for data analysis, which is a 70.7
percent response rate. The sample comprised of 54.7
percent males and 45.3 percent females. The demographic
data of the respondents is represented in table 1.0.
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
248 Business Intelligence Journal July
Table 1: Demographic Data of Respondents
/N Items Male Female
1 Age: 18-25 150 220
26-35 290 140
36-45 110 80
46 and above 30 40
Total 580 480
2 Educational qualifcation:
WASCE 100 60
OND 30 60
B.sc/BA 330 340
M.sc/PhD 110 20
Others specify 10 -
3 Duration of Patronage:
0-2 years 210 100
3-5 years 250 280
6-8 years 70 80
9 and above 50 20
4 Type of account:
Current 340 200
Savings 240 280
Others specify - -
Source: Fieldwork by Author.
Analysis
Exploratory principal component analysis with a
varimax rotation was used to summarize the items into
an underlying set of male and female decision-making
factors because they are constructs of a number of directly
observable variables. Exploratory factor analysis helps in
clustering these variables into factors with unique underlying
variables and it also helps to validate that respondents are
able to distinguish between the various variables despite
similarity of items questioned (De Vaus,2002).
Kaiser-Meyer – Olkin measure of sampling adequacy
for both samples were higher than the acceptable limit of
0.5 (Hair et al, 2005) The measure for male is 0.875 and
for females is 0.675 and Barlett’s tests were signifcant
refecting the suitability of data for factor analysis.
To assess the internal consistency of each factor
group obtained, a reliability analysis was conducted by
calculating the Cronbach’s alpha for each factor (Table 3).
For consistency, it was decided that reliability should not be
lower than 0.4, the same level used by Sproles and Kendall
(1986).
The descriptive statistics of the male and female samples
are shown in table 2.
Results and Discussion
The descriptive statistics of the samples (male and
female) revealed a highest average score of 5.62 in Q7
(i.e. In-branch banking involves too much queuing time) in
male samples and female samples has the highest average
score in Q18 (i.e. Electronic banking is convenient). These
two results tend to imply the same thing, while the male
customers denounced in-branch banking for its obvious
disadvantage of inconvenience, the female customer
upheld e-banking for being more convenient. Convenience
banking tend to be embraced by both male and female
customers (Q18 for male is 0.543). A further scrutiny of
the responses revealed that Q2, Q3, Q4, Q5, Q6, Q7, Q9,
Q10, Q11, Q12, Q13, Q14, Q15, Q16, Q17, Q18, Q20, Q21
had scores that ranged between 2.00 to 5.62 among males,
while Q1, Q2, Q5, Q6, Q7, Q8, Q9, Q15, Q16, Q17, Q18,
Q21 had scores that ranged between 5.00 to 5.62 among
females. The common factors where the male and female
customers had scores above 5.00 are Q2, Q5, Q6, Q7, Q9,
Q15, Q16, Q17, Q18 and Q21. These common factors are:
• Q2 : Transactions through electronic banking are
reliable
• Q5 : Electronic banking services are faster than in-
branch banking
• Q6 : Travelling to a bank involves too much queuing
time.
• Q7 : In-branch banking involves too much queuing
time.
Table 2: Descriptive Statistics of the Samples
Variables
Male Female
Mean
Std
Deviation
N Mean
Std
Deviation
N
Q1 4.92 1.742 580 5.36 1.953 480
Q2 5.29 1.643 580 5.18 1.652 480
Q3 5.12 1.618 580 4.77 1.828 480
Q4 5.18 1.600 580 4.69 1.067 480
Q5 5.34 1.430 580 5.23 1.597 480
Q6 5.17 1.478 580 5.33 1.644 480
Q7 5.62 1.434 580 5.05 1.716 480
Q8 4.95 1.501 580 5.05 1.654 480
Q9 5.16 1.664 580 5.10 1.465 480
2011 249
Variables
Male Female
Mean
Std
Deviation
N Mean
Std
Deviation
N
Q10 5.09 1.450 580 4.82 1.760 480
Q11 5.17 1.756 580 4.97 1.993 480
Q12 5.24 1.727 580 4.87 1.765 480
Q13 5.11 1.647 580 4.69 1.976 480
Q14 5.09 1.779 580 4.74 1.970 480
Q15 5.11 1.628 580 5.00 1.777 480
Q16 5.03 1.619 580 5.05 1.820 480
Q17 5.43 1.430 580 5.46 1.699 480
Q18 5.43 1.513 580 5.64 1.616 480
Q19 4.99 1.669 580 4.44 1.818 480
Q20 5.13 1.639 580 4.46 1.668 480
Q21 5.17 1.471 580 5.28 1.589 480
Q22 4.82 1.680 580 4.38 1.955 480
Q23 4.83 1.720 580 4.41 1.956 480
Q24 3.40 2.000 580 3.08 1.952 480
Q25 3.74 1.957 580 4.05 1.946 480
Q26 4.06 0.818 580 3.82 1.211 480
Q27 4.17 0.783 580 4.13 1.151 480
Q28 4.02 0.969 580 3.62 1.138 480
Q29 3.84 1.218 580 3.64 1.135 480
Source: Field work by Author
• Q9: Switching from electronic banking to in-branch
banking could be inconvenient to me.
• Q15: I like to use new methods to conduct transactions
e.g. ATM, telephone banking and internet banking.
• Q16: Electronic banking charges are expensive.
• Q17: Electronic banking is time saving.
Table 3: Results of Factor Analysis for Males and Females
Items Factor Leadings
All factors Male Female
Factor 1 – Service Quality (.568) (.551)
Q1 Transactions through electronic banking
are accurate.
.7056 .7868
Q2 Transactions through electronic banking
are reliable
.6799 .7758
Q3 I am familiar with e-banking .7913 .7593
Q4 I am comfortable with e-banking .7442 .7069
Q5 Electronic banking services are faster than
in-branch banking
.6068 .6855
Factor 2 – Perceived Risk Factors (.411) (.658)
Q6 Travelling to a bank involves too much
queuing time.
.6360 .7858
Items Factor Leadings
All factors Male Female
Q7 In-branch banking involves too much
queuing time.
.6924 .8545
Q8 Electronic banking services are faster than
in-branch banking.
.6847 .8378
Q9 Switching from electronic banking to
in-branch banking could be inconvenient
to me.
.6556 .6754
Q10 Going to a bank branch involves travel
costs.
.6335 .5622
Factor 3 – User Input Factors (.519) (.557)
Q11 I like to use electronic banking because it
ofers independence.
.8966 .8151
Q12 Electronic banking enables me to be fully
involved in my transactions.
.8032 .7408
Q13 E-banking is enjoyable to use. .9313 .6106
Q14 E-banking is user friendly. .8080 .7256
Q15 I like to use new methods to conduct
transactions e.g. ATM, telephone banking
and internet banking.
.7891 .6886
Factor 4 – Price Factor (.397) (.496)
Q16 Electronic banking charges are expensive. .5126 .4149
Factor 5 – Service Product Characteristics (.466) (.341)
Q17 Electronic banking is time saving. .7036 .7323
Q18 Electronic banking is convenient. .7267 .7261
Q19 Customer service in e-banking has
consistent standard.
.7836 .5852
Q20 Electronic banking has a wide variety of
services available.
.
Factor 6 – Individual Factors (.405) (.622)
Q21 I have a regular access to a computer. .6017 .5526
Q22 I have regular access to the internet. .6894 .6173
Q23 E-banking is easy to use. .6829 .7319
Q24 I use e-banking because my friends use it. .1042 .1622
Q25 The use of e-banking refects my social
status.
.3299 .2711
Factor 7 – Customers Satisfaction (.271) (.104)
Q26 How are you satisfed with time saving? .3560 .2978
Q27 How are you satisfed with convenience? .2439 .6529
Q28 How are you satisfes with the consistency
standards?
.3904 .4453
Q29 How are you satisfed with the wide
variety of services available?
.4463 .4216
Source: Field work by Author
• Q18: Electronic banking is convenient.
• Q21: I have a regular access to a computer.
The common reasons why both male and female
customers patronize e-banking in Nigeria is that it is
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
250 Business Intelligence Journal July
reliable, faster, does not involve travelling and queuing
time, it is time saving, and convenient. It is worthy to note
that an average customer sampled has access to computer.
And electronic banking is both seen to be expensive by
male and female customers.
The factors that are rated high by the male and not in
female, which can be called male factors, are: Q3, Q4, Q10,
Q11, Q12, Q13, Q14, and Q20. They are represented thus:
• Q3: I am familiar with e-banking.
• Q4: I am comfortable with e-banking.
• Q10: Going to a bank branch involves travel costs.
• Q11: I like to use electronic banking because it offers
independence.
• Q12: Electronic banking enables me to be fully involved
in my transactions.
• Q13: E-banking is enjoyable to use.
• Q14: E-banking is user friendly.
• Q20: Electronic banking has a wide variety of services
available.
The male customers patronizes electronic banking
because of its familiarity, comfort, cost effectiveness,
independence, high level of involvement, easy to use
feature, friendliness, and wide variety of services.
The factors that are peculiar to the female customers are
Q1 and Q8. They are represented thus:
• Q1: Transactions through electronic banking are
accurate.
• Q8: Electronic banking services are faster than in-
branch banking.
The female customer particularly patronizes the
electronic banking because of its accuracy and fastness.
Table 3 shows that the variables of service quality loaded
high for both male and female customers, they both patronize
electronic banking for the quality of service rendered, which
ranges from accuracy to speed. It is however important to
note that male customers loaded higher than the female
customers on all male factors (i.e. Q3, Q4, Q10, Q11, Q12,
Q13, Q14, and Q20) while female customers loaded higher
than male customers on factors peculiar to the female
customers as interpreted in table 2 (i.e. Q1 and Q8). Most
of the factors are important patronage of both male and
female customers except two of the individual factors for
both male and female customers; these are: Q24 – “ I use
e-banking because my friends use it”, Q25 – “The use of
e-banking refects my social status”.
This implies that both customers are not using
e-banking because of their friends and social status. These
factors should not be used as a unique selling proposition
for e-banking. In addition, strangely, male and female
customers loaded low on the group factor of customer
satisfaction, which means there is room for improvement
by banks to satisfy the customers better; presently the
expectations of the customers whether male or female are
higher than the service received particularly as it concerns
time saving (Q26 : .3560 for male and .2978 for female)
as well as consistency in standards (Q28 : .3904 for male).
The alignment of the male and female customer to some
of the factors of electronic banking is a signal to gender
implied impact on e-banking patronage, this result concur
with the fndings of Hanzaee and Aghasibieg (2008) who
established 10 – factors styles for males and 11 – factors
style for females, nine factors were common to both genders.
However both studies differ in the number of factors to
male and female, as well as common factors. This is so
because different product/service are being investigated
apart from using scale items peculiar to each study. For this
paper, 8 factors were common to male customers, 2 factors
to female customers and 10 factors were common to both.
Furthermore, the alignment of different factors by sex
confrms the fndings Kolodinsky et al (2000), Kolodinsky
and Hogarth (2001) who found that gender differences
existed in the adoption of new innovations and services
in banking in the United States. Special appeals could be
communicated to male and female customers based on the
identifed factors for further patronage.
The banks in Nigeria scored low in customer
satisfaction as it pertains to electronic banking, this calls
for improvement in service delivery in terms of timeliness
and consistency in standards.
Conclusion
It is evident that the male and female customers perceived
e-banking services differently and thus they patronize this
service for different reasons. Hence for the different banks
to sustain their interest there is dire need to emphasize the
common factors to the particular gender whose patronage
is essential depending on the product. The result also help
to identify the particular unique selling proposition to be
emphasized by the banks in the course of marketing any
gender inclined product, as their particular area of interest
is important for emphasis.
The issue of low satisfaction level of both genders is an
obvious area of improvement to be pursued by the banks.
2011 251
Some factors that are external to the operation of the banks
may be responsible for this, for example incessant power
failure. This means that banks may need to be proactive
to consistently keep their customers’ interest in a turbulent
business environment like Nigeria by ensuring that nothing
interrupts the fow of service delivery. Investments
should be made in this area to keep abreast of the level of
competition in the industry.
References
Akpan, I. (2009). Cross channel integration and
optimization in Nigerian banks. Telnet Press
Release, 20(1): 1-4.
Al-Ashban, A. & Burney, M. (2001) Customer
adoption of tele-banking technology: The case
of Saudi-Arabia. International Journal of Bank
Marketing 19(5): 191-200
Bakewell, C. & Mitchell, V.W. (2006). Male and
female decision making, Journal of Business
Research, 59: 1297- 1300
Bakewell, C. and Mitchell, V. W. (2003). Generation
Y female consumer decision-making styles.
International Journal of Retail & Distribution
Management, 31(2): 95-106
Chan, R.Y. (1997). Demographic and attitudinal
differences between active and inactive credit card
holders – The Case of Hong-Kong. International
Journal of Bank Marketing 15(4): 117 – 1200.
Cheung, M.I. and Liao, Z. (2002). Internet-based
e-banking and consumer attitudes: an empirical
study. Information and Management 39(4): 283 –
295.
Colgate, M; Nguyen, V. and Lee, C. (2003). Switching
barriers in New Zealand. University of Auckland
Business Review 5(1): 23 – 32.
Davidow, W. H. (1986). Marketing high technology:
An insider’s review. New York: Free Press.
De Vaus, D.(2002). Analyzing social science data.
London: Sage Publication
Delving, J. F (2002) Customer knowledge and choice
criteria in retail banking, Journal of Strategic
Marketing, 10:273-290.
Ezio, G. (2008) Authentication solution for e-banking.
New York: Prentice-Hall.
Fischer, E. & Arnold, S.J.(2004). Sex, gender identity,
gender role attitudes and consumer behavior.
Psychology & Marketing, 11(2):163-182
Gerrard, P. & Cunningham, J. (2003). The diffusion
of internet banking among Singapore consumers.
International Journal of Bank Marketing 21(1): 16-
28.
Hanzaee, K. H. & Aghasibieg, S. (2008) Generation
Y female and male decision-making styles in Iran:
Are they different? International Review of Retail,
Distribution and Consumer Research, 18(5):521-
537.
Horst, T. Andreas, P. & Arne, F. (2008) Success factors
of internet payment systems. International Journal
of Electronic Business 6(4): 365-385.
Jun,M. and Cai, S. (2001) The key determinants of
internet banking service quality: a content analysis.
International Journal of Banking Marketing
19(7):276-291.
Karjaluoto, H. (2002), Selection criteria for a mode
of bill payment: empirical investigation among
Finnish bank customers, International Journal of
Retail & Distribution Management, 30 (6):331-339.
Karjaluoto, H. Mattila, M. and Pento, T. (2002) Factors
underlying attitude formation towards online
banking in Finland. International Journal of Bank
Marketing, 20(6): 261-272.
Kolodinsky, Hogarth, J. M and Shue,J.(2001). Bricks
or Chicks? Consumers’ Adoption of Electronic
Banking technologies. Consumers Interest Annual
(46): 180-189.
Kolodinsky, J. and Hogarth, J. M (2001) The Adoption
of Electronic Banking Technologies by American
consumers. Consumer Interest Annuals, 47:1-9.
Kwan,C. W.,Yeung, K.W. & and Au, K.F.(2008)
Relationship between consumer decision making
styles and lifestyle characteristics: Young fashion
consumers in China. Journal of the Textiles Institute,
99(3): 193- 209.
Liao, Z., Cheung, M.T. (2002), Internet-based
e-banking and consumer attitudes: an empirical
study, Information and Management, 39:283-295.
Mitchell, V. & Walsh, G.(2004). Gender difference in
German consumer decision making styles. Journal
of Consumer Behaviour, 3(4):331-346.
Mitchell,V.W. & Bates, L. (1998). UK consumer
decision –making styles. Journal of Marketing
Management, 14: 199-225.
Mols, N.P. (1998). The internet and the banks strategic
distribution channel decisions. Journal of Internet
Research 8(4): 331 – 337.
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
252 Business Intelligence Journal July
Mukherjee, A. and Nath, P. (2003). A Model of Trust in
Online Relationship banking. International Journal
of Banking 21(1): 28-35.
Ng, T., & Palmer, E. (1999), Customer satisfaction
attributes for technology-interface services, MSIS
Department, School of Business and Economics,
University of Auckland, Auckland, unpublished
paper.
Nigeria Bureau of Statistics (2007). www.afdevinfo.
com/htm/reports/org/org_41583.html.
Polatoglu, V.N., & Ekin, S. (2001), An empirical
investigation of the Turkish consumers’ acceptance
of internet banking services, International Journal
of Bank Marketing,19(4):156-165.
Potchanakitummiai, S. & Speece, M. (2004). Barriers
to internet banking adoption. Journal of Electronic
Commerce Research 5(4): 270 – 286.
Report, W. (1997). The Financial System Inquiry Final
Report. New York: Free Press.
Sarin, S., Sego, T., Chanvarasuth, N. (2003). Strategic
use of bundling for reducing consumers’ perceived
risk associated with the purchase of new high-
tech products. Journal of Marketing Theory and
Practice, 11(3):71-83.
Sathye, M. (1999). Adoption of internet banking by
Australian consumers: an empirical investigation,
International Journal of Bank Marketing, 17(7):324-
334.
Slyke, C.V; Comunale, C.L; Belanger, F. (2002).
Gender differences in perceptions of web-based
shopping. Communications of the ACM 45(8): 82
– 86.
Sproles, G.B. & Kendall, E.L. (1986). A methodology
for profling consumers’ Decision-making
Styles, Journal of Consumer Affairs, 20(2): 267 – 279.
Sproles, G.B. (1985). From Perfectionism to Faddism.
Measuring Consumers’ Decision making Styles
proceedings. American Council on Consumer
Interest, 79 – 85.
Wan Omar, M; Ali, M.N; Hussin, Z; and Rahim, H.A.
(2009). Decision Orientations towards Shopping
and Buying among Young – Adult Malays in the
Universities, International Journal of Business and
Management 4(7): 26 – 30.
Zeithaml, V.A. (1981), How consumer evaluation
processes differ between goods and services, in
Donnelly, J.H., George, W.R. (Eds),Marketing
of Service, Proceedings Series of the American
Marketing Association.
2011 253
FOREIGN DIRECT INVESTMENT AND ITS EFFECT ON THE
NIGERIAN ECONOMY
Omankhanlen Alex Ehimare
Lecturer at the Department of Banking and Finance, Covenant University, Ota,
Ogun State, Nigeria.
Email: [email protected]
Abstract
This research study deals with the effect of Foreign Direct Investment on the Nigerian economy over the period 1980-2009.It helped
examined empirically if the following growth determining variables in the economy-Balance on current account (Balance of payment),
Infation and Exchange rate have any effect on Foreign Direct Investment. Also if Foreign Direct Investment have any effect on Gross
Domestic Product (GDP). Econometric models was developed to investigate the relationships between the aforementioned variables
and foreign direct investment. Based on the data analysis it was discovered that foreign direct investments have positive and signifcant
impact on current account balance in Balance of payment. While infation was seen not to have signifcant impact on foreign direct
investment infows. The exchange rate has positive effect on foreign direct investment. Therefore it is recommended that for Nigeria to
attract the desired level of FDI, it must introduce sound economic policies and make the country investor friendly. There must be political
stability, sound economic management and well developed infrastructure. Key words: Foreign direct investment, growth, human capital,
OLS, Nigeria, infrastructure, balance of payment, infation, exchange rate.
Governments have been trying to lift the country out
of the economic doldrums without achieving success as
desired. Each of these governments has not focused much
attention on investment especially foreign direct investment
which will not only guarantee employment but will also
impact positively on economic growth and development.
FDI is needed to reduce the difference between the desired
gross domestic investment and domestic savings.
Jenkin and Thomas (2002) assert that FDI is expected
to contribute to economic growth not only by providing
foreign capital but also by crowding in additional
domestic investment. By promoting both forward and
backward linkages with the domestic economy, additional
employment is indirectly created and further economic
activity stimulated.
According to Adegbite and Ayadi (2010) FDI helps fll the
domestic revenue-generation gap in a developing economy,
given that most developing countries’ governments do
not seem to be able to generate suffcient revenue to meet
their expenditure needs. Other benefts are in the form
of externalities and the adoption of foreign technology.
Externalities here can be in the form of licencing, imitation,
employee training and the introduction of new processes
by the foreign frms (Alfaro,Chanda,Kalemli- Ozean and
Sayek 2006).
Foreign direct investment consists of external resources
including technology, managerial and marketing expertise
and capital. All these generate a considerable impact
on host nation’s productive capabilities. The success of
government policies of stimulating the productive base
of the economy depend largely on her ability to control
adequate amount of FDI comprising of managerial,
capital and technological resources to boast the existing
production capacity. Although the Nigerian government
has being trying to provide conducive investment climate
for foreign investment, the infow of foreign investments
into the country have not been encouraging.
Given the Nigerian economy resource base, the country’s
foreign investment policy should move towards attracting
and encouraging more infow of foreign capital. The need
for foreign direct investment (FDI) is born out of the under
developed nature of the country’s economy that essentially
hindered the pace of her economic development. Generally,
policy strategies of the Nigerian government towards
foreign investments are shaped by two principal objectives
of the desire for economic independence and the demand
for economic development.
An analysis of foreign fow into the country so far have
revealed that only a limited number of multinationals or
their subsidiaries have made Foreign Direct Investment
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
254 Business Intelligence Journal July
in the country. Added to this problem of insuffcient
infow of FDI is the inability to retain the Foreign Direct
Investment which has already come into the country. Also
what effect have foreign direct investment have on such
variables as- Gross Domestic Product (GDP) and Balance
of Payment(BOP).Moreover, what effect does infation and
exchange rate have on Foreign Direct Investment? Carkovic
and Levine (2002) in their study concluded that exogenous
component of FDI does not exert a robust positive infuence
on economic growth.
According to Ayanwale (2007). The relationship
between FDI and economic growth in Nigeria is yet unclear,
and that recent evidence shows that the relationship may be
country and period specifc. Therefore there is the need to
carry out more study on their relationship.
Developing countries economic diffculties do not
originate in their isolation from advance countries. The
most powerful obstacle to their development comes from
the way they are joined to the international system. Added
to this problem is the poor external image Nigeria have
and the concept of European Economic Community that
include Eastern Europe.
This translates to the fact that investment fows that
would normally come from western countries now go to
poor European Economic Communities which include
Eastern Europe.
Foreign direct investment (FDI) is a major component
of capital fow for developing countries, its contribution
towards economic growth is widely argued, but most
researchers concur that the benefts outweigh its cost on the
economy. (Musila and Sigue, 2006).
Mc Aleese (2004) states that “FDI embodies a package
of potential growth enhancing attributes such as technology
and access to international market” but the host country
must satisfy certain preconditions in order to absorb and
retain these benefts and not all emerging markets possess
such qualities. (Boransztain De Gregorio and Lee 1998, and
Collier and Dollar, 2001). This paper is divided into fve
parts. Part one above is the introduction. Part two reviews
the relevant literature, part three discusses the methodology
employed in this study, and part four is data presentation
and analysis while part fve discusses the fndings and
recommendation.
This study will evaluate the fow of FDI in Nigeria and
its Effect on the Nigerian Economy. The period 1980-2009
will be investigated in the study. Only FDI, Government
Expenditure and Gross Capital formation will be used
as the explanatory variables. While GDP and balance on
current account of Balance of payment will be used as the
dependent variables.
Literature Review
The Changing international economic and political
environment has led to a renewed interest in the benefts
foreign direct investment (FDI) can offer to developing
countries in achieving economic growth. The growing
interest in foreign direct investment (FDI), stand from
the perceived opportunities derivable from utilizing this
form of foreign capital injection into the economy, to
augment domestic savings and further promote economic
development in most developing economies (Aremu 2005).
FDI is believed to be stable and easier to service than
bank credit. FDI are usually on long term economic
activities in which repatriation of proft only occur when
the project earn proft. As stated by Dunning and Rugman
(1985) Foreign Direct Investment (FDI) contributes to
the host country’s gross capital formation, higher growth,
industrial productivity and competitiveness and other spin-
off benefts such as transfer of technology, managerial
expertise, improvement in the quality of human resources
and increased investment.
According to Riedel (1987) as cited by Tsai (1994) while
the potential importance of FDI in less developed countries
(LDCs) development process is getting appreciated, two
fundamental issues concerning FDI remains unresolved. In
the frst place what are the determinants of FDI? Specifcally
from LDCs point of view are there factors in the control of
the host country that can be manipulated to attract FDI?
Or as some researchers claim that by and large LDCs
play a relatively passive role in determining the direction
and volume of FDI. This is the question about the demand
side determinants (or host country factors) of FDI which
are widely discussed in the literature.
There are also the supply side determinants or frm
specifc factors of FDI (Ragazzi 1973). The supply
side factors are beyond the control of LDCs. A body of
theoretical and empirical literature has investigated the
importance of FDI on economic growth and development
in less developed countries. For example see (Dauda 2007)
(Akinlo 2004) (Deepak, Mody and Murshid 2001) (Aremu
2005) e.t.c.
Modern growth theory rest on the view that economic
growth is the result of capital accumulation which leads to
investment. Given the overriding importance of an enabling
environment for investment to thrive, it is important to
2011 255
examine necessary conditions that facilitate FDI infow.
These are classifed into economic, political, social and
legal factors. The economic factors include infrastructural
facilities, favourable fscal, monetary, trade and exchange
rate policies. The degree of openness of the domestic
economy, tariff policy, credit provision by a country’s
banking system, indigenization policy, the economy’s
growth potentials, market size and macroeconomic stability.
Other factors like higher proft from investment, low
labour and production cost, political stability, enduring
investment climate, functional infrastructure facilities and
favourable regulatory environment also help to attract and
retain FDI in the host country. (Ekpo 1997).
According to the International Monetary Fund (1985)
foreign Direct Investment is an investment made to acquire
a lasting interest in a foreign enterprise with the purpose
of having effective voice in management. While Dunning
(1993) describe it as an investment made by an investor
based in a country to acquire assets in another country
with the intention to manage the assets. Mwillima (2003)
describe foreign direct investment as investment made so
as to acquire a lasting management interest (for instance
10% of voting stocks) and at least 10% of equity shares in
an enterprise operating in another country other than that of
the investor’s country.
Foreign Direct investment can also be describe as an
investment made by an investor or enterprises in another
enterprises or equivalent in voting power or other means
of control in another country with the aim to manage the
investment and maximize proft. This investment involves
not only the transfer of fund but also the transfer of physical
capital, technique of production, managerial and marketing
expertise, product advertising and business practice with
the aim to make proft.
In recent years due to the rapid growth and changes in
global investment patterns, the defnition of Foreign Direct
investment have been broadened to include the acquisition
of a lasting management interest in a company or enterprise
outside the investor’s home country.
Generally, the theory that explains the nexus between
FDI and growth in terms of output and productivity is
signifcantly positive. However, empirical literature
yields varying results. Some research studies fnd positive
outcome from outward FDI for the investing country (Van,
Poffelsberghe, De La Potterie & Lichtenberg, 2001), but
suggest a potential negative impact from inward FDI on
the host country. This results from a possible decrease in
indigenous innovative capacity or crowding out of domestic
frms. Other studies report more fndings that are positive.
For example, Nadiri (1993) fnds positive and signifcant
effects from US sourced FDI on productivity growth of
manufacturing industries in France, Germany, Japan and
United Kingdom. Borensztein, Gregorio and Lee (1998)
also fnd a positive infuence of FDI fows from industrial
countries on developing countries’ growth. However, they
also report a minimum threshold level of human capital for
the productivity enhancing impact of FDI, emphasizing the
role of absorptive capacity.
In the neo-classical production function approach, output
is generated by using capital and labour in the production
process. With this framework in mind, FDI can exert an
infuence on each argument on the production function.
FDI increases capital, and may qualitatively improve the
factor labour and by transferring new technologies, it also
has the potential to raise total factor productivity. Therefore,
in addition to the direct, capital augmenting effect, FDI
also have additional indirect and thus permanent effects
on output growth rate. Further, by raising the number of
varieties for intermediate goods or capital equipments, FDI
can also increase productivity (Borensztein, Gregorio &
Lee, 1998).
Therefore, though FDI could produce a signifcant effect
on output growth through speeding up capital formation
process, the effect tends to diminish in the long run because
of the principle of diminishing return.
As opposed to the limited contribution that the neo-
classical theory accredits to FDI, the endogenous growth
literature points out that FDI can not only contribute to
economic growth through capital formation and technology
transfers (Blomstrom, Lipsey & Zejan, 1996) but also do
so through the augmentation of the level of knowledge via
labour training and skill acquisition (De Mello, 1997).
Research Methodology Model
Specification
This study is based on the assumption that the infow
of FDI affects economic growth in Nigeria (GDP) and
Nigeria’s Balance of Payment (BOP). And again, that
infation and exchange rate in turn affect the infow of
Foreign Direct Investment (FDI). In other-words, GDP and
BOP are dependent on FDI, hence the model:
GDP = f (FDI)
(1)
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
256 Business Intelligence Journal July
Estimated Results
Model 1
t 0.842 1.568 10.237 3.063
GDP = ?
o
+ ?
I
FDI + ?
2
GOV + ?
3
GCF + e
GDP= 1.6709 + 4.0912FDI + 6.2835GOV. + 1.5457GFC
S.E.= (1.9847) (2.6086) (0.61381) (0.50454)
R
2
= 0.989607 F-Statistic= 825.24 D.W.= 2.74
Model 2
BCA = ?
o
+ ?
I
FDI - ?
2
GOV + ?
3
GCF + e
BCA= -1.3500 + 7.0662FDI + -0.49248GOV. + 0.42403GFC
S.E.= (1.1447) (21.5046) (0.35404) (0.29101)
t -1.179 4.696 -1.391 1.457
R
2
= 0.919443 F-Statistic = 98.917 D.W.= 1.72
BCA = f (FDI)
(2)
FDI = f (INFL., EXR.)
(3)
Where:
GDP = Gross Domestic Product
BCA = Current Account Balance
FDI = infow of Foreign Direct Investment
INFL. = Infation rate
EXR. = Exchange rate
Considering the fact that the GDP and BOP of an
economy are not determined by FDI alone, the inclusion
of two more growth determining variables is made so as to
get a more realistic model: Hence, equation (1) is extended
thus:
GDP = f (FDI, GOV, GCF)
(4)
BCA = f (FDI, GOV, GCF)
(5)
FDI = f (INFL, EXR.)
(6)
Where:
GOV = Government expenditure
GCF = Gross fxed capital formation
Equations (4) and (5) show that GDP and BCA are
dependent on FDI, GOV and GCF.
The statistical forms of the models are thus:
(7)
GDP = ?
o
+ ?
I
FDI + ?
2
GOV + ?
3
GCF + e
BCA = ?
o
+ ?
I
FDI - ?
2
GOV + ?
3
GCF + e
(8)
(9)
FDI = ?
o
- ?
I
INFL. - ?
2
EXR. + e
Where:
?0 = the intercept for equations (1) and (2)
?0 = the intercept for equation (3)
?I = the parameter estimate of FDI.
?2 = the parameter estimate of GOV.
?3 = the parameter estimate of GCF.
? I = the parameter estimate of INFL.
? 2 = the parameter estimate of EXR.
e = the random variable or error term.
Annual time-series data on the variables under study
covering thirty year period 1980-2009 are used in this study
for estimation of functions. Foreign Direct Investment
infow (FDI), Government Expenditure (GE) and Gross
fxed Capital Formation (GCF) are the relevant explanatory
variables. Equally, the Gross Domestic Product and Balance
on Current Account are the dependent variables. The Gross
Domestic Product is the quantitative variable that measures
economic performance of a country and the Balance on
Current Account measures BOP.
Presentation of Results
The regression analysis and tests of hypotheses are
conducted at 5% signifcance level. After running the
relevant regressions, the following results were obtained
and are presented below:
2011 257
Model 3
FDI = ?
o
- ?
I
INFL. - ?
2
EXR. + e
FDI= -14108. + -310.46 INFL. + 3731.5 EXR.
S.E.= (58549) (1678.9) (538.18)
t -0.241 -0.185 6.934
R
2
= 0.666903 F-Statistic= 27.029 D.W.= 0.453
Model 1
From the regressions result, the R-squared (R²) value of
0.989607 shows that at 98.96% the explanatory variables
explain changes in the dependent variable. This means that
at 98.96% the independent variables explain changes on
the Gross Domestic Product (GDP). This simply means
that the explanatory variables explain the behaviour of the
dependent variable at 98.96%. The calculated F-statistics
of 825.24 which is greater than the value in the F-table
(2.9751) implies that all the variables’ coeffcients in the
regression result are all statistically signifcant to GDP.
The Durbin-Watson (DW) as shown in the regression
analysis is 2.74 which shows that there is the presence of
autocorrelation.
The above model tested the effect of three different
variables namely – Foreign Direct Investment (FDI),
Government Expenditure (GOV) and Gross fxed Capital
Formation (GCF) on Gross Domestic Product (GDP). In
order to obtain the regression result, the OLS technique
with the help of the PC Give software was used.
The result obtained from the regression shows that there
is positive impact of Foreign Direct Investment (FDI) on
Gross Domestic Product (GDP) with a coeffcient of 4.0912.
However, this coeffcient is not statistically signifcant as
revealed by its corresponding standard error and t-values.
Hence, FDI is inelastic to GDP. This positivity in the
coeffcient of Foreign Direct Investment is in conformity
to the economic a priori expectation of a positive impact of
Foreign Direct Investment on the economic growth of the
economy (GDP).
Also, the regression result shows that the Government
Expenditure has a positive impact on GDP with a
coeffcient of 6.2835. The standard error and t-values
showed that this parameter is statistically signifcant. Thus,
the Government Expenditure is elastic to Gross Domestic
product. This positivity of the coeffcient of GOV conforms
to the economic a priori expectation of a positive impact of
Government Expenditure on GDP.
Furthermore, the result obtained from the regression
shows that Gross Fixed Capital Formation has a positive
impact on GDP. This is indicated in its positive coeffcient
of 1.5457. This coeffcient is revealed to be statistically
signifcant by the standard error and t-values. Thus, from
this it implies that Gross fxed Capital Formation is elastic
to GDP. The coeffcient of Gross fxed Capital Formation
being positive conforms to the economic a priori expectation
of a positive impact of GCF on the growth of the economy
vis-à-vis GDP.
Model 2
From the regressions result of model 2, the R-squared
(R²) value of 0.919443 shows that at 91.94% the explanatory
variables explain changes in the dependent variable. This
means that at 91.94% the independent variables explain
changes on Current Account Balance (BCA). This simply
means that the explanatory variables explain the behaviour
of the dependent variable at 91.94%. The calculated
F-statistics of 98.917 which is greater than the value in the
F-table (2.9751) implies that all the variables’ coeffcients
in the regression result are all statistically signifcant to
GDP.
The Durbin-Watson (DW) as shown in the regression
analysis is 1.72 which shows that there is the presence of
autocorrelation.
The above model tested the effect of three different
variables namely – Foreign Direct Investment (FDI),
Government Expenditure (GOV) and Gross fxed Capital
Formation (GCF) on Current account Balance (BCA). In
order to obtain the regression result, the OLS technique
with the help of the PC Give software was used.
The result obtained from the regression shows that
there is positive and signifcant impact of Foreign Direct
Investment (FDI) on Current Account Balance (BCA)
with a coeffcient of 7.0662. This coeffcient is statistically
signifcant as revealed by its corresponding standard
error and t-values. Hence, FDI is elastic to BCA. This
positivity in the coeffcient of Foreign Direct Investment
is in conformity to the economic a priori expectation of a
positive impact of Foreign Direct Investment on Current
Account Balance of the nation.
Also, the regression result shows that the Government
Expenditure has a negative impact on BCA with a
coeffcient of -0.49248. The standard error and t-values
showed that this parameter is not statistically signifcant.
Thus, the Government Expenditure is inelastic to Current
Account Balance. This negativity of the coeffcient of GOV
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
258 Business Intelligence Journal July
Discussion of Findings
The OLS regression analysis is carried out to determine
the impact of FDI, Government expenditure and Gross
fxed Capital Formation on GDP (proxy for economic
performance) and Balance of Payment through Balance
on Current Account (BCA), . Hence, GDP and BCA were
regressed on FDI, GOV and GCF. Though the impact of
FDI is of primary concern here, the other two economic
variables were included to serve as “control variables” to
check the overstating of the estimated coeffcient of FDI.
In model 3, the effects of two macroeconomic indicators,
infation and exchange rates were also examined. Hence,
FDI was regressed on infation and foreign exchange rates.
The results of the fndings show that FDI has positive
effect, though not statistically signifcant on GDP. In other
words, the infow of FDI into the Nigerian economy for the
stipulated period this research was carried out (1980-2009),
showed that FDI was not a major contributor to economic
growth of the nation. However, the fndings show that FDI
has positive and signifcant impact on BOP through current
account balance during the same period of analysis.
The effect of infation and foreign exchange rates on
FDI, brought under scrutiny, also showed that whereas
infation rate did not have major effect on the infow of
FDI into the Nigerian economy, foreign exchange rate had
great effect on the infow of FDI into the Nigerian economy
within the same period (1980-2009).
From the foregoing discussion, it should be pointed
out that although the government have made reasonable
efforts in attracting FDI, certain economic and political
circumstances prevalent in the country have hindered its
infow and its overall performance.
The primary objective of this study was to determine the
impacts and signifcance of FDI on the Nigerian economy
and the nation’s Balance of Payment (BOP). This was
achieved through the use of the OLS regression analysis of
data on the GDP, BCA, FDI, Government Expenditure and
Gross fxed Capital Formation sourced from the Central
Bank of Nigeria Statistical Bulletin.
The study also gave an opportunity for the examination
of the effects of infation rate and exchange rate on FDI. The
impact of Government expenditure on economic growth
was also examined. Therefore the following fndings were
revealed:
First, it was discovered that, FDI have not contributed
signifcantly to the economic growth of Nigeria in the
period under consideration.
conforms to the economic a priori expectation of a negative
impact of Government Expenditure on BCA.
Furthermore, the result obtained from the regression
shows that Gross Fixed Capital Formation has a positive
impact on BCA. This is indicated in its positive coeffcient
of 0.42403. However, this coeffcient is revealed not to be
statistically signifcant by the standard error and t-values.
Thus, from this it implies that Gross fxed Capital Formation
is inelastic to BCA. The coeffcient of Gross fxed Capital
Formation being positive conforms to the economic a priori
expectation of a positive impact of GCF on Balance of
Payment vis-à-vis BCA.
Model 3
From the regressions result of model 3, the R-squared
(R²) value of 0.666903 shows that at 66.69% the explanatory
variables explain changes in the dependent variable. This
means that at 66.69% the independent variables explain
changes on Foreign Direct Investment (FDI). This simply
means that the explanatory variables explain the behaviour
of the dependent variable at 66.69%. The calculated
F-statistics of 27.029 which is greater than the value in the
F-table (3.3541) implies that all the variables’ coeffcients
in the regression result are all statistically signifcant to FDI.
The Durbin-Watson (DW) as shown in the regression
analysis is 0.453 which shows that there is the presence of
autocorrelation.
The above model tested the effect of two different
variables namely –infation rate (INFL.) and Foreign
Exchange Rate (EXR.) on Foreign Direct Investment (FDI).
In order to obtain the regression result, the OLS technique
with the help of the PC Give software was used.
The result obtained from the regression shows that there
is negative and non-signifcant impact of infation on Foreign
Direct Investment (FDI) with a coeffcient of -310.46.
Hence, infation is inelastic to FDI. This negativity in the
coeffcient of infation is in conformity to the economic a
priori expectation of a negative impact of infation on FDI.
Again, the regression result shows that foreign exchange
has a positive effect on FDI with a coeffcient of 3731.5.
The standard error and t-values showed that this parameter
is statistically signifcant. Thus, the foreign exchange rate is
elastic to FDI. This negativity of the coeffcient of foreign
exchange rate does not conform to the economic a priori
expectation of a negative impact of foreign exchange rate
on FDI.
2011 259
Also, infation was found not to have any major effect on
the infow of FDI into the country. But exchange rate was
found to have major effect on FDI infow into the country.
In addition FDI contributed to Balance of payment
position through Current account balance. While Gross
fxed capital formation is inelastic to Balance on current
account.
In conclusion after the OLS regression analysis had been
carried out and with the study about the various factors
affecting FDI within the country, it is seen that:
1. There is no empirical strong evidence to support the
notion that Foreign Direct Investment has been pivotal
to economic growth in Nigeria; which could have
justify the effort of successive governments in the
country at using FDI as a tool for economic growth.
2. Governments direct involvement in the provision of
goods and services by establishing and controlling
corporations, for example, has contributed little
to economic growth in Nigeria. This justifes the
privatization policy of the various administrations in
our government to allow for the possible takeover by
investors (both foreign and domestic) of the government
corporations.
3. Though FDI has contributed signifcantly to Balance
of Payment (BOP) through the nation’s current account
balance. This is thus an effective measure of correcting
balance of payment disequilibrium in our economy.
Recommendations
The most signifcant factors that make Nigeria a good
host for FDI are her abundance in natural resources and
large population, indicating a large market.
The outcome of this study shows that though FDI was
not found to have signifcantly contributed to the nation’s
economic growth, if well harnessed it can contribute to
economic growth in Nigeria. To increase the infow of FDI
and its performance, the following recommendations from
this study are enunciated:
i. Balasubramanyam et al (1996) showed that most
economies beneft best from FDI when they are open to
foreign trade. Hence, the Nigerian government should
reduce the bureaucratic bottlenecks in foreign trade
especially the one constituted by the customs and port
authorities.
ii. Broensztein et al (1998) proved that there is a
high positive relationship between FDI and the level of
educational standard in the host economy. Based on this,
the country’s education should be in favour of science and
technology which would provide the economy with the
required skills that FDI require.
iii. Competitiveness should be encouraged, and as a
result, the existing and ‘yet-to-exist’ export processing and
free trade zones should be equipped with state-of-the-art
infrastructures and technologies.
iv. The infrastructures in the country need to be
enhanced to meet the needs/requirements of foreign
investors. For example, electricity should be provided at an
uninterrupted level to reduce the extra cost that investors
incur in the procurement of power generating sets coupled
with their maintenance. Also, good network roads and
adequate water supply should be provided so as to cut the
cost of investors doing business.
v. Appropriate measures should be placed to check
economic and fnancial crimes.
vi. The nation’s monetary authorities should develop
and implement measures that will ensure that both infation
and foreign exchange rates are sustained at levels that will
ensure increasing level of infow of FDI.
vii. The government and the private sector stakeholders
of our economy should consider harnessing infow of
FDI as a measure of improving the nation’s BOP through
current account balance, thereby ensuring the country’s
international competitiveness.
viii. Policy consistency should be emphasized.
References
Adegbite E.O and F.S. Ayadi (2010) “The Role of FDI
in Economic Development: A Study of Nigeria.”
World Journal of Entrepreneurship, Management
and Sustainable Development.Vol.6 No 1/2[Internet]
Available from www.worldsustainable.org
Akinlo, A.E. 2004. “Foreign direct investment and
growth in Nigeria: An empirical investigation”.
Journal of Policy Modeling, 26: 627–39
Alfaro, L., Chanda, A., Kalemli-Ozcan, S. & Sayek,
S. (2006). “How Does Foreign Direct Investment
Promote Economic Growth? Exploring the Effects
of Financial Markets on Linkages”. NBER Working
Paper no. 12522, National Bureau of Economic
Research, Cambridge, MA.
Aremu, J.A. 2005. “Foreign direct investment and
performance”. Paper delivered at a workshop on
Foreign Investment Policy and Practice organized
by the Nigerian Institute of Advanced Legal Studies,
Lagos on 24 March
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
260 Business Intelligence Journal July
IMF (1985) “Foreign Private Investment in
Developing Countries”, International Monetary
Fund, Washington .D.C.
Jenkins, C. and Thomas, L. (2002) “Foreign direct
investment in Southern Africa:determinants,
characteristics and implications for economic
growth andpoverty alleviation”, University of
Oxford [internet.] Available fromhttp://www.csae.
ox.ac.uk/reports/pdfs/rep2002-02.pdf
McAleese, D. (2004) Economics for business:
competition, macro-stability and globalisation. 3rd
ed. Edinburgh Gate: Pearson Education Limited
Musila, J.W. and Sigué, S.P. (2006) “Accelerating
foreign direct investment fow to Africa: from policy
statements to successful strategies”. Managerial
Finance, 32(7), 577-593.
Nadiri, M.I., 1993. “Innovations and Technological
Spillovers,” Working Papers 93-31, C.V. Starr
Center for Applied Economics, New York
University.
Nwillima N. (2008) “Characteristics, Extent and
Impact of Foreign Direct Investment on African
Local Economic Development”. Social Science
Research Network Electronic Paper Collection.http://ssrn.com Ragazzi, G. (1973) “Theories of the
Determinants of Direct Foreign Investment,” IMF
staff papers,20
Tsai, P.L. (1991) “Determinants of Foreign Direct
Investment in Taiwan: An Alternative Approach
with Time Series Data” World Development.
Van Pottelsberghe De La Potterie, B. and F. Lichtenberg
(2001), “Does Foreign Direct Investment Transfer
Technology Across Borders?” The Review of
Economics and Statistics Vol. 3 (3), 490-497.
Ayanwale A.B. (2007) “FDI and Economic Growth:
Evidence from Nigeria”. Africa Economic Research
Consortium Paper 165 Nairobi
Balasubramanyan, V., N. M.A. Salisu and D. Sapsford.
(1996). “Foreign Direct Investment and Growth
in EP and IS Countries”, Economic Journal, 106:
92–105
Borensztein, E., J. De Gregoria and J. Lee. 1998. “How
does foreign investment affect economic growth?”
Journal of International Economics, 45(1): 115–35.
Carkovic, M and Levine, R. (2002) “Does Foreign
Direct Investment Accelerate Economic Growth”
University of Minnesota Working Paper,
Minneapolis.
Collier, P. and Dollar, D. (2001) “Development
effectiveness: What have we learnt”? Development
Research Group, the World Bank[internet].
Availablefromhttp://www.oecd.org.
Dauda, R. O.S. (2007) “The Impact of FDI on Nigeria’s
Economic Growth: Trade Policy Matters”. Journal
of Business and Policy Research Vol. 3, No.2, Nov
– Dec
Deepak M., Mody A. and Murshid A.P. (2001)
“Private Capital Flows and Growth”, Finance and
Development. Vol. 38, June ,No.2.
De Mello L.R. (1997) “Foreign Direct Investment-
Led Growth: Evidence from Time Series and Panel
Data”, Oxford Economic Papers, 51: 133- 151.
Dunning J. (1993) Multinational Enterprises and the
Global Economy. England:Workingham Addison
Wesley Publishing.
Ekpo, A.H. 1995. “Foreign direct investment in
Nigeria: Evidence from time series data”. CBN
Economic and Financial Review, 35(1): 59–78
2011 261
Appendix I
REGRESSION RESULT
PcGive 8.00, copy for meuller session started at 13:39:56 on 24th October 2010
Data loaded from: alexpr~1.wks
EQ( 1) Modelling GDP by OLS - The present sample is: 1 to 30
Variable Coefcient Std.Error t-value t-prob PartR2
Constant 1.6709e+005 1.9847e+005 0.842 0.4075 0.0265
FDI 4.0912 2.6086 1.568 0.1289 0.0864
GOV._EXP. 6.2835 0.61381 10.237 0.0000 0.8012
GFC 1.5457 0.50454 3.063 0.0050 0.2652
R2 = 0.989607 F(3, 26) = 825.24 [0.0000] s = 820013 DW = 2.74
RSS = 1.748293983e+013 for 4 variables and 30 observations
EQ( 2) Modelling BCA by OLS - The present sample is: 1 to 30
Variable Coefcient Std.Error t-value t-prob PartR2
Constant -1.3500e+005 1.1447e+005 -1.179 0.2489 0.0508
FDI 7.0662 1.5046 4.696 0.0001 0.4590
GOV._EXP. -0.49248 0.35404 -1.391 0.1760 0.0693
GFC 0.42403 0.29101 1.457 0.1571 0.0755
R2 = 0.919443 F(3, 26) = 98.917 [0.0000] s = 472972 DW = 1.72
RSS = 5.816258697e+012 for 4 variables and 30 observations
EQ( 3) Modelling FDI by OLS - The present sample is: 1 to 30
Variable Coefcient Std.Error t-value t-prob PartR2
Constant -14108. 58549. -0.241 0.8114 0.0021
INFL. -310.46 1678.9 -0.185 0.8547 0.0013
EXR 3731.5 538.18 6.934 0.0000 0.6404
R2 = 0.666903 F(2, 27) = 27.029 [0.0000] s = 155120 DW = 0.453
RSS = 6.496770871e+011 for 3 variables and 30 observations
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
262 Business Intelligence Journal July
ARTIFICIAL LEARNING AND SUPPORT VECTOR MACHINES:
DEFAULT RISK PREDICTION
1
FABIO FRANCH, Ph.D.
2
University of Trento (Italy)
471 4th Ave Apt F – Chula Vista, CA 91910
Email: [email protected]
Abstract
The fnancial crisis that started in 2008 has shown how much work has still to be done in order to precisely predict the bankruptcy
of those actors who ask for credit to their banks. In this paper I focus my attention on large companies, using a database kindly
provided by Unicredit, one of the most important European banking groups. The size and the complexity of the problem has required
the simplifcation of the database and the use of Principal Component Analysis (PCA) in order to reduce the problem to a dimension that
is manageable by the Support Vector Machine (SVM) software chosen for this study. The best confguration found allowed the correct
classifcation of 84% of all companies and such results are found to be higher than many other reported in the literature. Key words:
Bankruptcy prediction, Basle 2, Default Risk, Principal Component Analysis, Support Vector Machines.
1
Master thesis presented at the University of Trento, under the valuable supervi-
sion of Professor Alessandro Zorat of the University of Trento. Final grade:
110/110 cum laude.
2
Fabio Franch earned his Ph.D. from West Virginia University and studied at Uni-
versity of Trento, where he successfully completed his studies (BA and MA) in
2007. He was also an exchange student at the Abo Akademi in Turku, Finland.
His interests are machine learning and the dynamic analysis of economic phe-
nomena.
3
For applications of artifcial learning techniques to fnancial and economic
problems, see Zhang, G.Peter 2004, Deforche, Koen 2004, Fauci F. 2004,
Fernandez-Rodriguez, F.,Gonzalez-Martel, C., & Sosvilla-Rivero, S. 2000, Fu-
kushima, K. 1975, Gencay, R. 1996, Kuoa R.J., P. Wub, C.P. Wangc 2002, Ro-
driguez, Claudia P. 2004, Rosenblatt, F. 1962, Scarselli, Franco, Sweah, Liang
Yong, Markus, Hagenbuchner, Ah Chung Tsoi 2005, Tewoldeberhan, Tamrat
W., Alexander, Verbraeck, Simon S. Msanjila 2005, Tortorella, Francesco 2002,
Viswanathan, Murlikrishna, Young-kyu Yang, e Taeg Keun Whangbo 2005,
Wan, H. A.,Hunter, A., & Dunne, P. 2002. For applications of artifcial learning
for default risk prediction, see Yongqiao Wang, Shouyang Wang, and K. K. Lai
(…) it doesn’t matter whether the forecast of the future is
true or not. If a mathematician should predict a long and
happy reign for me, a time of peace and prosperity for the
Empire -- Eh, would that not be well? (Asimov, 1988)
Bankruptcy prediction has been studied by many
researchers from different felds
3
and it is now getting
more and more important due to the Basel 2 Agreement
4
and the ongoing fnancial and economic crisis. Despite
the relatively high precision achieved by more traditional
statistical techniques, much more is needed to reach
perfection at predicting bankruptcy.
I decided to apply Support Vector Machines (SVMs) on
the prediction of corporate defaults for several reasons
5
:
1. the corporate sector is the set of companies
6
that is
affected by a relatively higher default rate
7
.
2. the prediction of the risk of default of large companies
using data from the balance sheet is far from perfection.
In order to accomplish such objective, I decided to turn
to a big fnancial group which already has at its disposal a
large amount of information about its customers as required
by the Basel 2 Agreement. According to this fnancial
agreement, banks commit themselves to store an amount
of resources for each loan that refects as accurately as
possible the risk connected to that particular operation.
Unicredit collects a large amount of information in
order to decide whether they deserve a loan or not. These
data enable them not only to describe potential customers
2005, Martin-del-Brio, Udo, G. 1993, Odom, M. & Sharda, R. 1990, Barniv
R., A. Agarwal, and R. Leach 1997, Bel et al. 1990, Rudorfer, Gottfried 1995,
Ohlson, J. 1980, Kaski Samuel, Janne Sinkkonen, and Jaakko Peltonen 2001.
4
The Basel 2 Agreement has been in force since 2007 for most banks in Europe.
Countries members of the Basel 2 Agreement are Belgium, Canada, France,
Germany, Italy, Japan, Luxemburg, Netherlands, Spain, Sweden, Switzerland,
the United Kingdom and the United States of America. To know more on the
Basel 2 Agreement, see International Financial Risk Institute (IFRI) 2000,
Moody’s 2006, Panizzolo Davide 2006, Ronchino Beatrice 2005.
5
The Probability of Default (PD) is usually estimated for a homogeneous group
of borrowers (based on the type of loan and features of the borrowers) and rep-
resents an assessment of the percentage of borrowers that are going to bankrupt
within a year. It entails the assessment of an expected value and an unexpected
value, which is used as safety margin.
6
These companies are all Unicredit customers.
7
These information is related to the business activity earlier than year 2007.
2011 263
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
in quantitative terms, but also to track the change of the
fnancial status of the borrowers over time.
However, the protection of such information is vital
for the survival of the fnancial group in the competitive
arena of today’s markets. In other words, Unicredit has to
protect the identity of its customers from its competitors
(for obvious reasons of maintaining its customer base), but
it has also to do it in order to avoid that its competitors
discover its market strategies and its policies and nullifes
the advantage that comes with them.
Moreover, Unicredit protects the identity of the
variables it has selected as most relevant to predict a
correct value of the risk of default, in order to preserve its
competitive advantage that comes from a better estimate
of the risk, which leads to a lower amount of resources to
be stored and to an increase of available resources which
can be invested in a more proftably way. From this it
follows that the database I was presented does not contain
any label that provides a clear description of what those
variables represent. Thus, it is not possible to state whether
that variable is about the net income, or gross sales, or any
other variable that Unicredit deemed necessary to obtain a
better assessment of the probability that a particular record
(company) will fulfll his obligations.
Despite of the lack of information about the meaning of
each column, the database contains a wide large amount of
information for each client. In fact, it is possible to identify
registry, quantitative, and qualitative pieces of information,
all of which refer to year 2003
8
.
8
Unicredit kindly provided the same data also for year 2004. However, I decided
to focus only on year 2003 in order to see how much SVMs can be useful to
predict a bankruptcy event using the fnancial, economic data in a precise time
of the year (the end of the year), without considering their evolving over time.
Please note that a longer time span is required to predict the Probability of De-
fault (PD), as regulated by the Basel 2 Agreement.
It is possible to fnd data relative to:
1. the specifc industry of each borrower;
2. place and date of foundation;
3. current year;
4. bank account information of the registered offce and of
the local offces;
5. borrower’s county of residence;
6. county to which belong the Chamber of Commerce in
charge of taking care of each borrower;
7. amount of employees (as of the end of the year);
8. total amount of employees;
9. total amount of seasonal labor;
The database presents also 40 columns that monitored
the conformity to some unknown criteria).
Another section follows (labeled as EL_FOND) which
contains data about budget funds fows and has numerical
values
9
. The section labeled as “EL_VOCERIC” is 300
columns long and provides information about budget items
and reclassifed budget items. Another section follows,
which defnes the type of reclassifcation used in the balance
sheet and then 6 columns, labeled as “GPL_SEM” and an
other set of columns called “GPL_SEM”
10
.
At the end of the database it is possible to fnd the
columns labeled as “SEZ_12”, the ones that describe the
typology of the borrowers (even though it is unclear how
borrowers are classifed) and then a column that states
whether that particular borrower was able to comply with
his/her fnancial obligations
11
.
SVM: Implementation, Pre-Processing
And First Results
SVMs are artifcial learning techniques based on the
linearity assumption in a multidimensional space and on
the optimization theory. This theory implements a learning
method which fnds its origins in the statistical learning
theory. This learning method was frst introduced by Vapnik
and others (Burges 1998; Cristianini and Shawe-Taylor
2000) and has enabled the resolution of different kinds
of problems since its earlier adoption, offering a superior
performance to other statistical techniques (Abbasi and
Chen 2005; Boyacioglu, et al. 2009; Drucker et al. 1999).
Once all observations (or examples) have been collected
for a specifc problem (including the outputs or labels, the
values to be correctly predicted), a researcher can make a
decision concerning the hypothesis regarding the specifc
problem. Among these, linear hypotheses are the ones that
can be more easily understood and used in order to explain
how SVMs work.
9
In this case (zero means that there are no funds, thus it is not a missing value.
10
It is not possible to understand the meaning of the data reported in these col-
umns.
11
Default or no default. It is a dummy variable.
10. total amount of managers;
11. legal expiry date of the bank loan;
12. operational expiry date;
13. legal structure;
14. goods price/value (Euros/Italian Liras);
15. budget type/form;
Business Intelligence Journal - July, 2011 Vol.4 No.2
264 Business Intelligence Journal July
( ) ( ) ( )
y
x
y
x
l
l
S , ,..., ,
1
1
=
where
?
*
is the solution of the following quadratic
optimization problem
( )
l i
to subject
W
l
i i
i
j i j
l
j i
i
j i
l
i
i
y
x x
y y
,..., 1 , 0
0
2
1
max
0
1 , 1
= ?
=
? ? =
?
? ?
=
= =
?
?
?
? ? ?
the vector
?
=
=
l
i i i
i
x
y
w
1
*
?
is the maximum margin
hyperplane with a geometric margin equal to:
.
1
2
w
= ?
According to the Karush-Kuhn-Tucker condition, the
optimal solution needs to satisfy the condition
( ) | | . ,..., 1 , 0 1
* * *
l i
b x w
y
i i
i
i
= = ? + ?
?
\which implies that only to input
xi
for which the
functional margin is equal to 1 and which are in the
proximity of the hyperplane, correspond values of
?
*
i
which are different from zero.
All the other parameters
?
*
i
are equal to zero (Burges
1998; Cristianini and Shawe-Taylor 2000).
It follows that in the expression of the weights vector
only the points close to the hyperplane matter (see Figure
1), and it is for this reason that these points are termed
‘support vectors’.
Figure 1: Example of Support Vectors
WINSVM
12
was chosen as the tool to be used in this
study thanks to its Windows interface and its optimization
function, which enables the generation of a random set of
parameters and the testing of the SVM
13
with the generated
parameters.
The database presented several issues:
1. missing values: this issue was solved by deleting
records or columns that contained one or more of such
values;
2. redundancy of information: issue solved by deleting all
the columns that had values which were equal for all
50,000 records;
3. redundant information: Unicredit itself reported the
uselessness of using some pieces of information to
predict the risk of default;
4. mix of qualitative and quantitative values: issue
solved through the binarization of qualitative variables
(columns).
Given the size of the database and the insuffcient
computing power to delete/flter records or columns of the
entire database, I sorted all remaining records according to
the outcome of the bank loan operation (default/no default).
12
WINSVM is a Windows-based software which allows regression and classif-
cation. Seehttp://www.cs.ucl.ac.uk/staff/M.Sewell/winsvm/ for instructions on
how to use it. Other useful pieces of software are SVM and Kernel Methods
Matlab Toolbox (http://asi.insa-rouen.fr/~arakotom/toolbox/index.html), Near-
est Point Algorithm (http://guppy.mpe.nus.edu.sg/~mpessk/npa.shtml), SVM-
QP (https://projects.coin-or.org/SVM-QP), 2D Svm Interactive Demo (http://
www.learningwithkernels.org/software.html), and others (seehttp://www.ker-
nel-machines.org/software.html,http://www.support-vector.net/software.html,http://www.support-vector-machines.org/SVM_soft.html).
13
See Cristianini, Nello, John, Shawe-Taylor 2000 for a detailed description of
the theoretical assumptions on which is based the learning of a Support Vector
Machine.
In a linearly separable sample S
2011 265
Given this positive results, I then extended the dataset
to 1353 records, of which 60% for training and 40% for
testing. However, this time the optimization process was
more intensive than before; as a result, the system crashed.
Despite such an event, several confgurations have been
saved
18
: I decided then to run WINSVM on two computers
(similar to the computer described above), aiming at
obtaining as many confgurations as possible. This is
possible thanks to the random generation of the parameters
and their accuracy testing process. This operation required
The Unicredit database was further simplifed by:
1. deleting columns that the Unicredit Credit Risk
manager himself suggested as useless for our purpose;
2. keeping only columns whose values were expressed in
Euros;
3. keeping only the records of borrowers that reside on the
Italian territory;
4. deleting records which contain one or more missing
values;
5. deleting records that had more than 4 marks in the
section preceding the one labeled as “EL_FOND”
17
.
6. deleting of columns which presented constant values
for all records.
The Unicredit database, at the moment of its handing,
contained around 50,000 records, but only a small
percentage described companies which turned out to be in
fnancial troubles. After the pre-processing phase and the
simplifcation mentioned above the database included 2254
records and 607 columns. I then labeled “0” all records
which refer to bankrupted enterprises and “1” to all the
remaining ones.
I went on testing the effectiveness of the SVM by
removing all registry data, which required the binarization
and therefore an explosion of the amount of columns. The
goal at this stage was to test the computing power needed
to process the entire database. I randomly chose 79 records
for the training phase and 11 records for the testing. On
an ASUS laptop, 2.8Ghz processor, 512Mb RAM the
optimization phase lasted between 2 and 7 minutes, using
14
50% of the database constituted companies that paid back the loans after one
year, the remaining 50% were not able to do it.
15
A much larger amount of borrowers were able to fulfll their obligations due to
the fact that customers that received the loans were those that Unicredit deemed
deserving and enough wealthy to borrow money.
16
This provides a better assessment of which company will default and which one
is not going to.
17
Some records had more than 4, which was against the rules that were set by Uni-
credit before creating the database. This is probably due to wrong data entries
or misunderstandings within the organization itself.
both the regression and the classifcation. Less than a second
was the time required to train and test the SVM using the
selected parameters.
The SVM performed really well on this small dataset
(see Table I).
Average loss : 0.12007405
Avg. loss pos : 0.22013576 (6 occurences)
Avg. loss neg : 0 (5 occurences)
Mean absolute error : 0.5098412
Mean squared error : 0.39600989
Accuracy : 0.90909091
Precision : 0.83333333
Recall : 1
Predicted values:
| + | -
---+-----+-----
+ | 5 | 0 (true pos)
- | 1 | 5 (true neg)
Table I: output of the testing phase with 79 training samples and
11 testing samples.
I then selected all the records regarding those borrowers
that were not able to fulfll their obligations and an equal
amount of borrowers that were able to repay the loan
14
. This
was done to reduce the size of the database and to balance
the amount of records of the borrowers that did not pay
back their loans
15
. The equal size of the two classes (default
and non default) is supposed to enhance the capability of
the Support Vector Machine to understand the difference
between the two classes
16
.
18
Some of the confgurations lead to high accuracy predictions: accuracy is the
percentage of positive values and negatively values that were respectively pre-
dicted by the SVM as positive and negative out of the total of values.
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
266 Business Intelligence Journal July
19
200 confgurations were deemed suffcient to proceed with the analysis.
20
After this simplifcation, the database had 1799 records and 582 columns. We
assumed that the SVM would have probably performed badly with such a high
number of columns; thus, we deemed Principal Component Analysis necessary
to signifcantly reduce the amount of columns.
21
Unicredit provided data that was mostly about a particular feld it chose as the
most troublesome given the relatively higher percentages of bankruptcy.
I then
20
:
1. deleted every record (company) which operated in a
different industry
21
;
2. deleted the information about the industry to which
belongs each borrower;
3. deleted the data about the Chamber of Commerce and
the statistical ID of each borrower: this information
was deemed to be useless to predict the Probability of
Default;
4. deleted all the columns that presented data that was
used only to identify in some way each record;
5. deleted all the columns about employees: too many
missing values are in fact included in these columns;
6. added qualitative information that was removed in the
previous step;
Principal Component Analysis (PCA)
The original database was subject to a simplifcation
process that only marginally reduced the amount of columns
due to the assumption that a lower amount of them will
enhance the performance of the SVM when it processes
new samples. However, the steps described in the previous
section results to be insuffcient to signifcantly reduce the
complexity of the database.
I therefore decided to use Principal Component Analysis.
This statistical technique generates new factors that are
then chosen as axis of bi-dimensional maps which are then
interpreted in order to understand the degree to which these
factors can actually explain the underlying phenomenon.
Factors are generated by a correlation matrix; this
matrix is then processed in order to have as many factors
as the number of variables included in the process. Factors
have the property of being independent and are linearly
constructed using the original variables (Bowler 2002;
Bonifazzi 2006; Easton 2009).
Factors are then sorted according to their ability to
provide the highest amount of information, or, in other
words, to better explain the phenomenon. According to the
literature, a specifc amount of factors can be selected on
the basis of criteria like the following:
1. keep all factors which have an auto-value greater than
1;
2. keep all factors that can explain at least 66% of the
variability;
3. keep all factors that can explain at least 75% of the
variability;
4. keep those factors that are located on the left part of
the graph generated after choosing the variables to be
processed to generate the new factors.
5. keep all factors that are deemed to hold important
information (Mortara 2006).
It is important to remember that the software used for
this purpose, INSTAMAP
22
:
7. deleted all the columns that included the same value in
all records.
6 to 7 hours, because the execution of SVM had to be
repeated several times in order to obtain a large amount of
confgurations
19
.
I then tried several confgurations of parameters like C,
epsilon, nu, Kernel that led to a testing MSE that is less
than 0.01 and I located the one that returns the highest
accuracy value in the training phase. Such a confguration
has a radial Kernel, gamma equal to 0.3, C equal to 10 and
epsilon equal to 0.0001. Although this one was the best
confguration, it performed poorly in the testing phase:
accuracy is in fact rather low, 69%. I concluded from this
analysis that using fewer training samples and even less
testing samples made it easier for the SVM to recognize
the hyper-plane that correctly classifed 10 out of 11 testing
items. Raising the number of training samples apparently
led the SVM to defne a hyper-plane that did not separate
correctly the two classes of companies. It was plausible to
assume that such an issue was due to the higher complexity
of defning a multi-dimensional hyper-plane, using a much
larger amount of variables. In other words, the SVM seemed
to get confused by the excessive amount of data.
Thus, I proceeded further simplifying the database,
looking for a meaningful criterion which would allow
the removal of a considerable amount of columns while
keeping the number of training and testing samples as high
as possible.
22
See Mortara V. (2006) for more information.
1. cannot read column labels that are longer than 5
characters;
2011 267
Average loss : 0.55649621
Avg. loss pos : 0.18958699 (181 occurences)
Avg. loss neg : 0.92340542 (181 occurences)
Mean squared error : 0.99198794
Accuracy : 0.47513812
Precision : 0.40425532
Recall : 0.10497238
Predicted values:
| + | -
---+------+------
+ | 19 | 162 (true pos)
- | 28 | 153 (true neg)
Accuracy is at this point very low, less than 48%, with
mostly negative value only predicted. Thus, I decided to test
the SVM performance using all generated confgurations.
Only one confguration performed really well, with an
accuracy of 81.49%. I tried then to improve its performance
Table III: output of the classifcation procedure with the best set
of parameters
*** mySVMversion 2.1.4 ***
----------------------------------------
Predicting
Average loss : 0.40571761
Avg. loss pos : 0.75934415 (181 occurences)
Avg. loss neg: 0.052091077 (181 occurences)
Mean absolute error : 0.98534041
Mean squared error : 0.99136138
Accuracy : 0.62430939
Precision : 0.57525084
Recall : 0.95027624
Predicted values:
| + | -
---+------+------
+ | 172 | 9 (true pos)
- | 127 | 54 (true neg)
Table II: testing with the optimal confguration
Accuracy was at this point even lower than the one
previously estimated. Thus, I decided to test all the other
confgurations, but with no improvement in the testing
performance.
At this stage, the average loss was rather high in any
experiment conducted. Having an average loss equal to
0.5 means that predictions are greater than or less than the
correct value to be predicted by an average of 0.5. However,
the best confguration appeared to be the frst one, to which
corresponds an average loss equal to 0.35784271.
Given the poor performance of regression, I decided
to use classifcation. I chose to keep the confgurations
generated with regression, but I had to change the values
of the last column of the database
23
from “0” (no default
occurred) to “-1” (a default event occurred in the short),
while keeping all remaining labels the same.
I started again from the confguration of parameters that
performs the best in the training phase. The output of the
classifcation procedure with the set of parameters with the
lowest MSE is presented in Table 4.
23
Default “yes” or default “no”.
2. cannot process a large amount columns at a time:
I therefore chose to split the database in several
subsets containing 19 columns each and a last one
which included also the fnal column containing the
information about the occurring of the default event for
each record.
In this study I decided to keep as many factors as were
needed in order to explain 66% of the variability of the
data. Doing so reduced the size of the columns to 180 to
which I had to add the 75 columns of qualitative data that
were spared from the factor generation process given their
sparseness.
All factors were then copied into a new spreadsheet,
saved later in the TXT format which is needed in order
for WINSVM to read the data. I chose to use 80% of the
records for the training phase and the remaining 20% for
testing.
WINSVM crashed again due to the large amount of
information presented, but only after having generated
more than 300 possible optimal confgurations. I decided
that such an amount of options was enough to have an idea
of which Kernel can perform the best.
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
268 Business Intelligence Journal July
24
This confguration uses a very large Soft Margin, which is used to cope with
the outliers and the noise, all problems that probably affected the Unicredit
database.
by marginally decreasing or increasing one or more of its
parameters (see below). Values of epsilon equal to 0.01,
0.2, 0.15, 0.17, 0.18, 0.19 generated an accuracy that never
exceeded 81%.
I observed that with an epsilon value of 0.18 it was
possible to get the least average loss. I kept this value
of epsilon and changed C and gamma to optimize the
performance of the SVM. However, it was not possible to
refne the predictions using as starting confguration the
one whose MSE is equal to 0.2637. The best confguration
using the Radial Kernel had therefore a C value that is
equal to 1000000, an epsilon value of 0.1 and gamma equal
to 0.01
24
. In short, the performance of Radial Kernels was
usually very high in the training phase, but vey low in the
testing phase, which led us to think that Radial Kernels
defne a hyper-plane that is overly based on the specifc
training data it processed during the training. I needed to
fnd a confguration that is able to perform better in the
testing phase, which would support my claim of robustness
of the fndings of this work.
Thus, I went on testing other Kernels, choosing again
the confgurations to which corresponded a low error
(MSE). I chose to analyze the performance of Polynomial
Kernels. The best confguration seemed to be the one that
has a MSE equal to 0.85472, C equal to 1, epsilon equal to
0.0001, third degree. MSE was rather high, still it managed
to correctly classify all 1437 records in the training phase.
However, results in the testing phase were rather bad, with
an accuracy of almost 64%. Once again, I tried to optimize
the confguration of parameters, but only reducing the
degree from 3 to 2 led to an improvement of results. Not
even raising the C value positively affected the performance
which was stationary around an accuracy level equal to
71% (see below).
Table IV: accuracy in the testing phase with polynomial Kernel
I decided then to switch to an Anova Kernel, starting once
again from the best confguration and trying to optimize its
parameters in order to improve the accuracy in the testing
phase. The best confguration seemed to be the one that had
a MSE equal to 0.0691369, C equal to 100, epsilon equal to
0.0001, gamma equal to 0.9 and degree equal to 4. To such a
confguration of the SVM corresponded an accuracy equal
to 77%. Once again, I optimized the parameters, aiming at
fnding a better solution. It turned out that a high value of
gamma (gamma=30) improved the prediction’s accuracy of
6 percentage points
25
(see below).
25
Accuracy is thus equal to 83%.
Change in parameter Testing results
Degree=2
Average loss : 1.6596905
Mean squared error : 126.33603
Accuracy : 0.70718232
Precision : 0.69230769
Recall : 0.74585635
Change in parameter Testing results
Degree=2, C=0.0001
Average loss : 1.6596905
Mean squared error : 126.33603
Accuracy : 0.70718232
Precision : 0.69230769
Recall : 0.74585635
Degree=2
Average loss : 1.6596905
Mean squared error : 126.33603
Accuracy : 0.70718232
Precision : 0.69230769
Recall : 0.74585635
Degree=4
Average loss : 0.68685004
Mean squared error : 46.331544
Accuracy : 0.70718232
Precision : 0.68115942
Recall : 0.77900552
Change in parameter Testing results
Gamma=1
Average loss : 0.37226828
Avg. loss pos : 0.42555524 (238 occurences)
Avg. loss neg : 0.26999171 (124 occurences)
Mean squared error : 1.396154
Accuracy : 0.77071823
Precision : 0.71875
Recall : 0.88950276
Table V: testing with Kernel Anova
2011 269
I observed at this point the superior performance of
SVMs with a Neural Kernel. Almost all randomly generated
confgurations led to very accurate predictions. Moreover,
the last prediction has an accuracy that is almost equal to
84%, the highest observed so far. To such a high accuracy
corresponds a low number of misclassifed records. This
high-performance confguration was also characterized
by a very low average loss and especially by the highest
precision level
26
. Such a high value of precision is due
to the low amount of negative values that were predicted
as positive. Only the recall
27
value was not the highest
among those I observed until that moment. This was due
to the relatively higher amount of positive values that
were predicted as negative. Thus, I concluded that such
a confguration was the optimal one, given the highest
accuracy and precision achieved. The sharp decline of the
accuracy level when testing for robustness of the results
confrms the complexity of the problem under investigation
26
Precision is defned as the positive values that have been predicted as such out
of all values that have been predicted as positive even if they were negative.
27
Recall is defned as the amount of positive values predicted as such out of all
“real” positive values.
At this point I was not still satisfed with the results
obtained. I decided to see what the use of a Neural
Kernel entailed. I started from the confguration to which
corresponded a MSE equal to 0.314255, C equal to 0.01,
epsilon equal to 0.00001, a equal to 0.4 and b equal to 0.9
and a prediction accuracy equal to 81%.
Such accuracy was already interesting in itself; however,
I wanted to see if there was any way to further increase its
accuracy (see below). After having tried several options
by modifying only epsilon, I was unable to improve the
accuracy and only in one case the accuracy was as high
as in the former confguration. Thus I decided to increase/
decrease a and b and to keep all remaining parameters at
the same level (see below).
Change in parameter Testing results
Gamma=2
Average loss : 0.32014074
Avg. loss pos : 0.39044826 (234 occurences)
Avg. loss neg : 0.1916098 (128 occurences)
Mean squared error : 1.043599
Accuracy : 0.79281768
Precision : 0.73451327
Recall : 0.91712707
Gamma=30
Average loss : 0.2352293
Avg. loss pos : 0.34048536 (203 occurences)
Avg. loss neg : 0.10084578 (159 occurences)
Mean squared error : 0.60141935
Accuracy : 0.82596685
Precision : 0.76818182
Recall : 0.93370166
Variation in a e b
values
Risultati del testing
a=1
b 0.1
Average loss : 0.24368945
Avg. loss pos : 0.34219957 (198 occurences)
Avg. loss neg : 0.12475651 (164 occurences)
Mean absolute error : 0.65285786
Mean squared error : 0.60010934
Accuracy : 0.8121547
Precision : 0.7627907
Recall : 0.90607735
Table VI: results in the testing phase modifying the reported
parameters. C and epsilon are held constant.
Variation in a e b
values
Risultati del testing
a 0.1
b 0.1
Average loss : 0.20976328
Avg. loss pos : 0.31135803 (201 occurences)
Avg. loss neg : 0.08292759 (161 occurences)
Mean absolute error : 0.62960401
Mean squared error : 0.5516383
Accuracy : 0.83149171
Precision : 0.77272727
Recall : 0.93922652
a 0.1
b 1
Average loss : 0.19665233
Avg. loss pos : 0.20571275 (186 occurences)
Avg. loss neg : 0.18707711 (176 occurences)
Mean absolute error : 0.64513177
Mean squared error : 0.55654285
Accuracy : 0.83977901
Precision : 0.83606557
Recall : 0.84530387
Predicted values:
| + | -
---+------+------
+ | 153 | 28 (true pos)
- | 30 | 151 (true neg)
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
270 Business Intelligence Journal July
in this work. In fact, when using 30% of the selected data
for testing (up from 20%), the accuracy slightly drops to
79%. An additional ten percentage points in accuracy are
lost when using 40% of the data for testing purposes.
Discussion
It is important to understand the meaning of the
predicted values in order to measure the impact of using
such a model. I shall recall that, using classifcation,
predicted values range from “-1” to “+1”, where “+1” is
given to records to which corresponds an event of default
and “-1” in all other cases (no default). That means that
those 28 misclassifed records represented companies that
appeared to Unicredit as solid and deserving credit, but that
were at the end unable to fulfll their obligations, which
is exactly what Unicredit aimed at minimizing
28
. I should
therefore conclude that, after all, I should not choose such a
confguration, since I am only interested in having the least
number of companies possible that borrow money, even if
they cannot pay it back (a confguration that misclassifes
28 companies is therefore not the best one).
In order to draw even better conclusions, I have to
understand what the fact of having only 30 misclassifed
companies implies. The 30 companies predicted as positive,
but negative in reality, represent in fact companies that
the model classifes as weak and therefore not deserving
credit, but which are actually solid enough to fulfll their
obligations.
A large banking group like Unicredit should minimize
also such an occurrence, since it represents a loss of
customers. However, in a real world situation an important
banking group like Unicredit has a rather wide customer
base, which enables it choose to lend money only to those
companies that are clearly able to pay their loans back. In
such a situation, the confguration to which corresponded
an accuracy value of 84%
29
appears to be the optimal in
order to minimize losses and maximize revenues.
Conclusion
The purpose of this paper is to determine whether SVMs
can be helpful to predict companies’ default risk. This study
had to deal with several issues, among which the most
critical one was the lack of knowledge of the meaning of
28
This strategy is currently being pursued.
29
The confguration which uses a Neural Kernel.
the variables included in the database. Such an issue turned
out to be irrelevant, if I consider that PCA and the SVM
processed the data in such a way that the most important
variables were not rejected thanks to their informative
content.
Given an accuracy level equal to 84%, I can state that
SVMs can successfully be used to improve the ability to
predict their ability to fulfll their obligations.
The results achieved here appear to be even superior to
those found in the literature (see below).
Table VII: prediction accuracy in Yongquiao Wang, Shouyang
Wang, and K. K. Lai 2005
30
Source: Yongquiao Wang, Shouyang Wang, and K. K. Lai 2005
In conclusion, I believe that it is possible to further
enhance the performance of the classifcation process in
several ways, by:
1. performing again PCA, after having removed the most
infuential outliers. This would marginally reduce
the amount of records, but it would also increase the
variance of the data, making the task of fnd a better
hyperplane easier for the SVM. I chose not to go this
way because I believed that the qualitative information
would have, in some way, justifed the values
corresponding to the outliers.
2. performing again the PCA and reducing the informative
content to a lower level (less than 66%). I did not go
this way because the literature advises against such an
option
31
.
30
These results are the best ones obtained by Yongquiao Wang, Shouyang Wang,
and K. K. Lai 2005. They used a dataset containing 653 records and 15 features.
31
As reported in Mortara V. 2006.
The rating class is the class a borrower belongs to given its ability to fulfll its
obligations. The more solid a company is, the higher its ranking will be –AAA,
BB and so on –.
2011 271
3. processing as many records as possible, therefore
including a larger amount of records of companies that
were actually able to pay their loans back. I believe
that such an option could lead to some interesting
results, after having observed that most of the times the
amount of calculated Support Vectors was almost equal
to the amount of training samples. This means that the
SVM “considers” each training sample as “special”
and calculates a Support Vector for (almost) each of
them. In other words the hyper-plane is very close to
most of the training samples, which led us to think that
increasing the amount of training samples would help
the SVM defning a more accurate hyper-plane.
In conclusion, I have to point out that the predicted
values cannot directly be interpreted as the Probability of
Default (according to Basel 2 regulation, PD has also to
be calculation on a 5-year base at least: I only used data of
year 2003). These values have rather to be interpreted as a
measure of reliability of stating that a particular company
will bankrupt or not. In particular, the closer the predicted
value to “-1” or “+1” is (or even greater than 1), the more
credible/reliable the statement about the possible default
of that specifc company. The closer the predicted value to
zero, the more unreliable such a statement.
It would have been interesting to try to predict the rating
class of each borrower
32
, but to do so I would have needed
to:
1. have the information of the rating class of each
borrower, needed to train the SVM on them;
2. deal with multi-class classifcation problem, which
involves the use of as many SVMs as the number of the
rating classes.
Given the high accuracy I obtained by using a Neural
Kernel, I can state that SVM can proftably be used to
predict the fnancial solidity of potential borrowers. SVMs,
in other words, offer themselves as an additional tool with
which to fght and avoid dangerous fnancial crises like the
one that been plaguing the international community since
the second half of year 2008.
References
Abbasi, A. and H. Chen (2005) "Applying authorship
analysis to extremist-group web forum messages."
IEEE Intelligent Systems 20(5): 67-75.
Asimov, Isaac (1988) Prelude to Foundation, New
York: Bantam Books, p. 2.
Barniv R., A. Agarwal, and R. Leach (1997) Predicting
the outcome following bankruptcy fling: A
three-state classifcation using neural networks,
IntelligentSyst. in Accounting, Finance, and
Management, vol. 6.
Bell, T., G. Ribar, and J. Verchio (1990) Neural nets vs.
logistic regression: A comparison of each model’s
ability to predict commercial bank failures, in Proc.
(1990 Deloitte Touche/Univ. Kansas Symp. on
AuditingProblems, 29–53.
Belli Giorgio (2004) Valutazione delle imprese e
merito creditizio alla luce del nuovo accordo di
Basilea, thesis written at the University of Rome
“La Sapienza”, Academic Year 2003/2004.
Bonifazzi, C., M. Cinti, et al (2006) "Principal
Component Analysis." Physica Medica 21: 1.
Bowler, S. and T. Donovan (2002) "Democracy,
institutions and attitudes about citizen infuence on
government." British Journal of Political Science.
32(02): 371-390.
Boyacioglu, M., Y. Kara, et al (2009) "Predicting
bank fnancial failures using neural networks,
support vector machines and multivariate statistical
methods: A comparative analysis in the sample of
savings deposit insurance fund (SDIF) transferred
banks in Turkey." Expert Systems with Applications
36(2): 3355-3366.
Burges, C (1998) "A tutorial on support vector
machines for pattern recognition." Data mining and
knowledge discovery 2(2): 121-167.
Cristianini, Nello, John, Shawe-Taylor (2000) An
introduction to support vector machines and
other kernel-based learning methods, Cambridge:
Cambridge University Press.
Deforche, Koen (2004) Bayesian Network Learning
Università di Leuven per workshop The HIV Data
Management and Data Mining, South Africa.
Drucker, H., D. Wu, et al (1999) "Support vector
machines for spam categorization." IEEE
Transactions on Neural networks 10(5): 1048-1054.
Easton, D (2009) "A re-assessment of the concept
of political support." British Journal of Political
Science. 5(04): 435-457.
Fauci F (2004) Studio e sperimentazione di un
nuovo metodo per la classifcazione di immagini
mammografche University of Palermo.
Fernandez-Rodriguez, F.,Gonzalez-Martel, C., &
Sosvilla-Rivero, S (2000) On the proftability of
technical trading rules based on artifcial neural
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
272 Business Intelligence Journal July
networks: Evidence from the Madrid stock market,
Economic Letters, 69, 89–94.
Fukushima, K. (1975) Cognitron: A Self-Organizing
Multilayered Neural Network, Biological
Cybernetics (20: 121–136.
Gencay, R. (1996) Nonlinear prediction of security
returns with moving average rules, Journal of
Forecasting, 15, 165–174.
Kaski Samuel, Janne Sinkkonen, and Jaakko Peltonen
(2001) Bankruptcy Analysis with Self-Organizing
Maps in Learning Metrics, IEEE Transactions on
neural networks, Vol. 12, n.4, July.
Kuoa R.J., P. Wub, C.P. Wangc (2002) An intelligent
sales forecasting system through integration of
artifcial neural networks and fuzzy neural networks
with fuzzy weight elimination.
Martin-del-Brio, B. & Serrano-Cinca, C. (1995) Self-
organizing neural networks: The fnancial state of
Spanish companies.
Mortara V (2006) INSTAMAP: un programma per
la produzione “istantanea” di mappe fattoriali.
University of Trento.
Odom, M. & Sharda, R. (1990) A neural network
model for bankruptcy prediction, Proceedings
of the IEEE International Conference on Neural
Networks, 163–168.
Ohlson, J. (1980) Financial ratios and the probabilistic
prediction of bankruptcy, Journal of Accounting
Research, 18(1), 109–131.
Rodriguez, Claudia P (2004) Energy Price Forecasting
in the Ontario Competitive Power System Market,
IEEE Transactions on power systems,. Vol. 19, n.1
February.
Rosenblatt, F (1962) Principles of Neurodynamics,
Spartan Books.
Rudorfer, Gottfried (1995) Early Bankruptcy Detection
Using Neural Networks.
Scarselli, Franco, Sweah, Liang Yong, Markus,
Hagenbuchner, Ah Chung Tsoi (2005) Adaptive
Page Ranking with Neural Networks, 936-937.
Tewoldeberhan, Tamrat W., Alexander, Verbraeck,
Simon S. Msanjila (2005) Simulating Process
Orchestrations in Business Networks: A Case Using
BPEL4WS.
Tortorella, Francesco (2002) Approccio statistico alla
classifcazione course slides of ‘Teoria e Tecniche
di Interpretazione delle Immagini’, University of
Cassino.
Viswanathan, Murlikrishna, Young-kyu Yang, e Taeg
Keun Whangbo (2005) Minimum Message Length
and Classical Methods for Model Selection in
Univariate Polynomial Regression ETRI Journal,
27, Number 6, December, 750.
Yongqiao Wang, Shouyang Wang, and K. K. Lai.
(2005) A New Fuzzy Support Vector Machine to
Evaluate Credit Risk IEEE Transactions on Fuzzy
Systems, 13, n.6.
Zhang, G.Peter (2004) Neural Networks in Business
Forecasting, Idea Group Publishing.
2011 273
REVAAM MODEL APPLIED TO MULTIPLE VALUATION
COMPARISON AMONG DIFFERENT WORLD REGIONS
Carlos Acosta-Calzado (MBA)
Business Intelligence Service
IBKAN Consultants, Investment Banking
Email: [email protected]
Abstract
This paper uses the REVAAM Model used to calculate adjusted multiples, through multiple linear regression, used commonly in
valuating companies such as the Price Earnings (PE) and/or the Enterprise Value to Earnings Before Interest, Taxes, Depreciation and
Amortization (EV/EBITDA). For this paper, we used REVAAM to calculate historical adjusted multiples for eight regions in the world:
Africa, Asia, Western Europe, Eastern Europe, Latin-America and the Caribbean, Middle East, Oceania and United States & Canada;
all the World; and the BRIC group (Brazil, Russia, India and China). We show the yearly obtained multiples, as well as the long term
multiples for each region and compared them. During our analysis, we found that although emerging markets do show higher multiples,
these are converging towards mature market multiples. When we compare long term multiples to those for each year we can distinguish
differences that could be consider bubbles due to speculation and to fundamentals (proftability factors). Key words: REVAAM, multiple
valuation, PE multiple, EBITDA multiple, EV/EBITDA
REVAAM
1
model was created as a way to make
commonly used multiple valuation a true "comparable"
valuation tool, rather than just a mathematical averaging
method that could be applied to the company we are
valuing. This model implies considering other fundamental
factors that truly affect valuation of a specifc company
and that are different from other companies even within
the same industry. These parameters refer to proftability
factors and are company specifc, but the methodology
of the model lets incorporate other parameters that could
include macroeconomic or industry specifc factors.
Although there are several multiples used in relative
valuation: Price Earnings
2
(PE), Enterprise Value to Earnings
Before Interest, Taxes, Depreciation and Amortization
3
(EV/EBITDA), Price to Sales
4
(PS), Price to Book Value
of Equity
5
(PBV), and variations to these, REVAAM model
concentrates in P/E and EV/EBITDA multiples as they are
the most used multiples and refer to the proftability of the
company.
For purposes of our work, we constructed a database
with historic information from thousands of companies
around the world and grouped them in 10 proposed regions,
from years ending 2000 to 2010. We wanted frstly to use
REVAAM model to calculate adjusted multiples for these
regions and compare them to see if there were differences
among regions, and secondly obtain and compare short term
multiples (calculated each year) and long term multiples
(without differentiating each year).
Hypotheses
As stated before, multiples vary among markets
according to region specifc risks, economic conditions
and risk-return parameters, as opposed to only fundamental
factors such as proftability at company level. Additionally,
we believe that markets are not rational in the short term
and tend to overreact to region specifc or world events,
thus making fair market value deviate from the theoretical
long-run value. Taking these arguments into consideration
we will try to test the following hypotheses:
Hypothesis 1: Adjusted long term and short term
multiples for emerging or underdeveloped regions
will be higher than those in mature markets.
1
Relative Valuation Adjusted Model
2
Price Earnings ratio is calculated as the market price per share divided by net
earnings per share
3
Enterprise Value to Earnings Before Interest, Taxes, Depreciation and Amortiza-
tion ratio is calculated as the value of operating assets divided by operating
earnings before interests and taxes plus depreciation and amortization
4
Price to Sales ratio is calculated as the market price per share divided by sales
per share
5
Price to Book Value of Equity is calculated as the market price per share divided
by the book value per share
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
274 Business Intelligence Journal July
Hypothesis 2: Adjusted yearly multiples will
signifcantly differ from the adjusted long term
multiples in each market.
Methodology
Model
REVAAM Model adjusts PE multiple by using the
Return on Equity
6
(ROE) and the Net Margin
7
as the
indicators of proftability for equity shareholders, and
adjusts EV/EBITDA multiple by using the Return on
Capital
8
(ROC) and the Pre-tax Operating Margin
9
. These
adjustments are made using multiple regression analysis in
order to minimize errors, where the independent variables
are the multiples and the other factors are the dependent
variables. These relationships are shown in the equations
below:
6
Return on Equity measures the rate of return on shareholders’ equity and is cal-
culated as the Net Income divided by the book value of equity
7
Net Margin measures the proftability and is calculated as the net income divided
by revenues
8
Return on Capital measures the rate of return on the company’s capital and is
calculated as operating earnings after taxes divided by the book value of equity
plus the book value of debt
9
Pre-tax Operating Margin or EBIT measures proftability before interests and
taxes and is calculated as operating revenues less operating expenses plus non-
operating income
(1)
Where,
P/E = the price earnings ratio calculated as the price per
share divided by net earnings per share
?
PE
= the value of the intercept calculated by the
regression analysis
?
ROE
= the value of the coeffcient for ROE
ROE = the return on equity, calculated as the net income
divided by the company’s beginning equity value
?
NM
= the value of the coeffcient for Net Margin
Net Margin = calculated as the net income of year t
divided by the sales of year
? = the sum of errors derived from the regression analysis
(2)
? Margin Operating tax - Pre * ?
ROC * ? ? EV/EBITDA
OM
ROC EVEBITDA
+
+ + =
Where,
EV/EBITDA = the EBITDA multiple calculated
as the Enterprise Value or Value of the Net Operating
Assets divided by the Earnings before Interests, Taxes,
Depreciation and Amortization
?
EVEBITDA
= is the value of the intercept calculated by the
regression analysis
?
ROC
= the value of the coeffcient for ROC
ROC = the return on capital, calculated as the after-tax
operating income (EBIT*(1-tax)) divided by the company’s
beginning equity value plus the debt balance
?
OM
= the value of the coeffcient for Net Margin
Pre-tax Operating Margin = calculated as the pre-tax
operating income of year t divided by the sales of year
? = the sum of errors derived from the regression analysis
? Margin Net * ROE * P/E
NM ROE PE
+ + + = ? ? ?
As stated in the REVAAM Model, the basic linear
regression model assumes that the contributions of the
different independent variables to the prediction of the
dependent variable are additive and they tend to follow
normal distributions. In our case, the relationships between
our variables may be multiplicative and also they have
highly skewed distributions (positive values). Hence it may
be possible to make their distributions more normal-looking
by applying the logarithm transformation, as shown in the
following equations:
From equations (1) and (2), we apply natural logarithms
to obtain:
? Margin) Ln(Net *
Ln(ROE) * Ln(P/E)
NM
ROE PE
+
+ + =
?
? ?
(3)
? Margin) Operating tax - Ln(Pre *
Ln(ROC) * DA) Ln(EV/EBIT
OM
ROC EVEBITDA
+
+ + =
?
? ?
(4)
Equations (3) and (4) will be used to obtain long-term
adjusted multiples for each region without considering
differences among years.
For short-term or yearly adjusted multiples we need to
differentiate samples for each year and, in order to simplify
the number of equations, we decided to consider dummy
variables for each year, so that the value will equal to 1 for
the corresponding year and zero for all other years.
2011 275
Hence, from equations (2) and (3), we introduce dummy
variables as follows:
(5)
(6)
Ln(EV/EBITDA) = ?
EVEBITDA
+ ?
ROC
*
Ln(ROC) + ?
OM
* Ln(Pre-tax Operating
Margin) + ?
2000
*2000 +?
2001
*2001 +
?
2002
*2002 + ?
2003
*2003 + ?
2004
*2004 +
?
2005
*2005 + ?
2006
*2006 + ?
2007
*2007 +
?
2008
*2008 + ?
2009
*2009 + ?
2010
*2010 + ?
Ln(P/E) = ?
PE
+ ?
ROE
*Ln(ROE) + ?
NM
*Ln(Net Margin) + ?
2000
*2000 +?
2001
*2001
+ ?
2002
*2002 + ?
2003
*2003 + ?
2004
*2004 +
?
2005
*2005 + ?
2006
*2006 + ?
2007
*2007 +
?
2008
*2008 + ?
2009
*2009 + ?
2010
*2010 + ?
Where,
20xx equals to 1 if it corresponds to the year and zero
for any other year
Sample
These regressions should be performed at company
level due to the incorporation of proftability factors and
results could be grouped by industry, country, region,
etcetera. REVAAM Model could be applied to any sample
of companies where proftability factors are available.
Since information of private companies is limited, we need
to use public information. For purpose of our analysis,
we obtained information of public companies from 113
countries with one or more stock exchanges, from years
2000 to 2010, where available. The following table
summarizes the number of companies we were able to
obtain for each region.
Table 1. Number of companies by sector per year
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total
Africa 0 0 307 357 380 440 512 622 378 813 857 4,666
Asia 0 0 3,075 6,087 7,353 9,081 9,675 10,621 11,949 20,245 22,083 100,169
Europe_east 0 0 3 613 737 886 1,269 1,719 569 1,213 1,462 8,471
Europe_west 0 0 3,992 3,343 3,695 3,821 4,297 4,384 4,301 4,840 4,871 37,544
Latam&Carib 0 0 628 501 578 532 589 694 776 882 923 6,103
Middle East 0 0 320 399 463 580 721 818 549 1,123 1,154 6,127
Oceania 0 0 0 495 531 654 681 722 881 1,571 1,428 6,963
US&Canada 5,752 7,154 7,440 7,575 6,072 6,162 8,579 8,246 8,059 9,303 7,863 82,205
World 5,752 7,154 15,765 19,370 19,809 22,156 26,323 27,826 27,462 39,990 40,641 252,248
BRIC 0 0 1,353 2,118 2,339 3,079 3,643 4,447 3,434 8,615 10,119 39,147
From the total 252,248 companies, some could be
considered in several years but with data corresponding to
each particular year.
When applying the samples to the model, we found that
the number of companies that can be used was reduced
because:
a. Negative fgures and values equal to zero cannot be
transformed to natural logarithms,
b. we took out "outlier" companies with ROE or ROC
higher or equal to 200% and those with Net margin or
Pre-tax Operating Margin higher or equal to 100%, and
c. some companies had missing information for some
proftability factors.
Thus, the following tables summarize the number of
companies we used for each multiple after the adjustments
on the samples.
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
276 Business Intelligence Journal July
Table 2. Final sample for P/E REVAAM calculation
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total
Africa 0 0 237 269 301 361 390 475 231 277 339 2,880
Asia 0 0 2,516 4,843 6,340 7,819 8,029 8,764 8,154 6,789 8,939 62,193
Europe_east 0 0 0 281 284 343 865 1,355 316 369 462 4,275
Europe_west 0 0 2,707 2,197 2,471 2,580 3,032 3,178 2,853 2,127 2,385 23,530
Latam&Carib 0 0 411 364 442 445 519 574 562 515 604 4,436
Middle East 0 0 212 261 342 437 549 604 309 335 530 3,579
Oceania 0 0 0 330 346 392 429 428 482 496 462 3,365
US&Canada 2,826 3,285 2,902 3,328 3,342 3,554 3,744 3,551 3,613 3,100 3,031 36,276
World 2,826 3,285 8,985 11,873 13,868 15,931 17,557 18,929 16,520 14,008 16,752 140,534
BRIC 0 0 1,086 1,577 1,806 2,368 2,809 3,553 2,069 1,968 2,493 19,729
Table 3. Final sample for EV/EBITDA REVAAM calculation
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total
Africa 0 0 255 268 288 338 385 411 184 226 263 2,618
Asia 0 0 2,561 4,958 6,297 7,772 8,007 8,570 7,708 9,425 11,607 66,905
Europe_east 0 0 0 389 462 577 880 1,242 266 388 460 4,664
Europe_west 0 0 2,648 2,514 2,802 2,866 3,076 3,223 2,713 2,281 2,481 24,604
Latam&Carib 0 0 500 430 456 456 519 558 515 544 586 4,564
Middle East 0 0 224 255 348 425 561 600 227 288 381 3,309
Oceania 0 0 0 338 380 428 401 395 438 458 454 3,292
US&Canada 3,382 3,752 3,615 3,951 3,929 3,969 4,179 3,892 3,914 3,811 3,654 42,048
World 3,382 3,752 9,803 13,103 14,962 16,831 18,008 18,891 15,965 17,421 19,886 152,004
BRIC 0 0 1,132 1,675 1,893 2,498 2,794 3,360 1,877 3,592 4,470 23,291
For both cases, the fnal sample used for the REVAAM
Model is substantially lower than the initial sample, but it
does contain suffcient data points to run linear regression
analysis.
For the case of Eastern Europe, both the original and the
fnal samples had few data points for year 2002; thus, we
decided no to include them for that particular year.
Results
From the samples obtained we ran the linear regression
model, frst using equations (5) and (6) to obtain the
adjusted multiples per year for each region. The following
are the resulting equations for each region. For additional
parameters, please refer to appendices 1 and 2.
Africa
Ln(EV/EBITDA) = 1.71E+12+ -0.3623*
Ln(ROC) + 0.1071* Ln(Pre-tax
Operating Margin) -1.71E+12*2002 -
1.71E+12*2003 -1.71E+12*2004 -
1.71E+12*2005 -1.71E+12*2006 -
1.71E+12*2007 -1.71E+12*2008 -
1.71E+12*2009 -1.71E+12*2010
(7)
2011 277
(8)
Ln(PE) = 7.60E+12+ -0.117* Ln(ROE) +
-0.0752* Ln(Net Margin) -
7.60E+12*2002 -7.60E+12*2003 -
7.60E+12*2004 -7.60E+12*2005 -
7.60E+12*2006 -7.60E+12*2007 -
7.60E+12*2008 -7.60E+12*2009 -
7.60E+12*2010
For the African region, we do not have information for
years 2000 and 2001, and the coeffcients for the intercept
and for each year are very high, and although using scientifc
notation for this fgures makes them appear equal among
them, the resulting numbers in the spreadsheet are different.
The resulting F-tests for both equations show that the linear
regression model can be used for the relationships we
proposed and the T-tests also show that our main variables
(ROC, ROE, Net margin, and Pre-tax Operating Margin) are
signifcantly different from zero. The R2 was 26.23% for
the EV/EBITDA multiple and 4.73% for the P/E multiple,
which means that the former, with the data available, has
an equation that explains better the corresponding multiple.
Asia
Ln(EV/EBITDA) = 8.37E+12 -0.3194*
Ln(ROC) + 0.1527* Ln(Pre-tax Operating
Margin) -8.37E+12*2002 -
8.37E+12*2003 -8.37E+12*2004 -
8.37E+12*2005 -8.37E+12*2006 -
8.37E+12*2007 -8.37E+12*2008 -
8.37E+12*2009 -8.37E+12*2010
(9)
Ln(PE) = -7.57E+11 -0.471* Ln(ROE) -
0.0444* Ln(Net Margin) +
7.57E+11*2002 + 7.57E+11*2003 +
7.57E+11*2004 + 7.57E+11*2005 +
7.57E+11*2006 + 7.57E+11*2007 +
7.57E+11*2008 + 7.57E+11*2009 +
7.57E+11*2010
(10)
For the Asian region, again we do not have information
for years 2000 and 2001. The resulting F-tests for both
equations show that the linear regression model can be
used for the relationships we proposed and the T-tests also
show that all main factors are signifcantly different from
zero. The R2 was 15.24% for the EV/EBITDA multiple
and 14.39% for the P/E multiple, so both show similar
proportion of variability explained by the available data.
We need to remember that the Asian region has the highest
number of companies in the whole sample.
Eastern Europe
(11)
Ln(EV/EBITDA) = 0.9011 -0.2276*
Ln(ROC) -0.0934* Ln(Pre-tax Operating
Margin) + 0.172*2003 + 0.1303*2004 +
0.4316*2005 + 0.3527*2006 +
0.3464*2007 -0.03506*2008 +
0.3574*2009 + 0.5927*2010
Ln(PE) = 1.534 -0.4556*
Ln(ROE) + 0.1702* Ln(Net
Margin) + 0.5978*2003 +
0.2182*2004 + 0.5321*2005 -
0.6714*2006 -1.0663*2007 -
0.2348*2008 + 0.4339*2009 +
0.7888*2010
(12)
For Eastern Europe we do not have information for
years 2000 and 2001, and as stated before we eliminated
year 2002 because data points for that year were few. The
resulting F-tests for both equations show that the linear
regression model can be used for the relationships we
proposed and the T-tests also show that all main factors are
signifcantly different from zero. The R2 was 7.14% for
the EV/EBITDA multiple and 4.98% for the P/E multiple.
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
278 Business Intelligence Journal July
Western Europe
Ln(EV/EBITDA) = 1.8465 -0.3479* Ln(ROC) +
0.1822* Ln(Pre-tax Operating Margin) -
0.0626*2002 + 0.0152*2003 + 0.1348*2004 +
0.2591*2005 + 0.4073*2006 + 0.1581*2007 -
0.4386*2008 + 0*2009 + 0.1693*2010
(13)
(14)
Ln(PE) = 1.4909 -0.3908* Ln(ROE) -0.1129*
Ln(Net Margin) + 0.0212*2002 + 0.1008*2003
+ 0.3085*2004 + 0.4554*2005 + 0.5943*2006
+ 0.2374*2007 -0.5214*2008 + 0*2009 +
0.2509*2010
In the case of Western Europe information starts in
2002 as well. In this case, the coeffcient or year 2009 was
calculated as zero in both equations even though there is
data for that particular year; this means that the model is
using 2009 as the base year. The resulting F-tests for both
equations show that the linear regression model can be
used for the relationships we proposed and the T-tests also
show that all main factors are signifcantly different from
zero, including the intercept value obtained. The R2 was
32% for the EV/EBITDA multiple and 22.28% for the P/E
multiple; a high correlation factor in fnancial regression
models.
Latin-America & the Caribbean
Ln(EV/EBITDA) = 1.5172 -0.3317* Ln(ROC)
+ 0.0895* Ln(Pre-tax Operating Margin) -
0.2196*2002 -0.1945*2003 -0.0642*2004 +
0*2005 + 0.216*2006 + 0.2106*2007 -
0.4391*2008 + 0.249*2009 + 0.3912*2010
(15)
Ln(PE) = 1.4069+ -0.5303* Ln(ROE) +
0.0559* Ln(Net Margin) -0.95*2003 -
0.9394*2004 -0.1834*2005 -0.0044*2006 -
0.5897*2007 -0.2693*2008 + 0.4695*2009 +
0.5748*2010
(16)
Regression equations for Latin-America and the
Caribbean are for years 2002 to 2010, where information
was available. F-tests show that both linear regression
models can be used for the relationships we proposed and
the T- tests also show that all main factors are signifcantly
different from zero, except for Net Margin in equation (16)
that suggests that this factor may not contribute to explaining
P/E multiple in this region. For purposes of this analysis and
in order to keep all models consistent among them, we keep
the variable in the model. Additionally, the use of dummy
variables for differentiating years could be misleading when
removing a critical factor such as Net Margin. The correct
procedure should be to run independent linear regression
models for each year and determine when is the factor not
important. R2 was 17.41% for the EV/EBITDA multiple
and 5.48% for the P/E multiple. This difference may be the
result of using Net Margin in the P/E model.
Middle East
Ln(EV/EBITDA) = 1.9015 -0.4348* Ln(ROC) +
0.1815* Ln(Pre-tax Operating Margin) -
0.0406*2003 + 0.2154*2004 + 0.3912*2005 +
0.0338*2006 + 0.1307*2007 -0.5468*2008 -
0.2345*2009 -0.1019*2010
(17)
Ln(PE) = -1.91E+11 -0.4064* Ln(ROE) -0.1208*
Ln(Net Margin) + 1.91E+11*2002 + 1.91E+11*2003 +
1.91E+11*2004 + 1.91E+11*2005 + 1.91E+11*2006 +
1.91E+11*2007 + 1.91E+11*2008 + 1.91E+11*2009 +
1.91E+11*2010
(18)
The Middle East region also has data available for years
2002 to 2010. F-tests also show that the linear regression
model is a good model for the proposed relationships. T-tests
also show that all main factors are signifcantly different
from zero, except for the intercept with a high probability
(95%) of being equal to zero. As in the equations for Latin-
America and the Caribbean we will keep this coeffcient.
The resulting R2 was 29.96% for the EV/EBITDA multiple
and 5.38% for the P/E multiple, which also shows that
maybe EV/EBITDA multiple for this region is better than
P/E when using the proposed relationships.
2011 279
Oceania
Ln(EV/EBITDA) = 1.9298 -0.3244* Ln(ROC)
+ 0.0872* Ln(Pre-tax Operating Margin) -
0.0629*2003 -0.0168*2005 + 0.1463*2006 -
0.007*2007 -0.7868*2008 + 0.0083*2009 -
0.1072*2010
(19)
Ln(PE) = 1.9966 -0.4642* Ln(ROE) -0.0399*
Ln(Net Margin) -0.0576*2003 -0.0141*2005 +
0.1294*2006 -0.1522*2007 -1.1194*2008 -
0.0343*2009 -0.2005*2010
(20)
Oceania region has data available for years 2003 to 2010.
As in the case of Western Europe, both resulting equation
calculate as zero the coeffcient for the base year, which
in this case is 2004. Both F-tests and T-tests show that the
linear regression model is a good model for the proposed
relationships and that all main factors are signifcantly
different from zero, including the intercept. The resulting
R2 was 30.17% for the EV/EBITDA multiple and 33.41%
for the P/E multiple, both high values that show similar
proportion of variability explained by the available data.
US&Canada
Ln(EV/EBITDA) = 1.6474 -0.2439* Ln(ROC) +
0.0492* Ln(Pre-tax Operating Margin) -
0.2262*2001 -0.3049*2002 + 0.0087*2003 +
0.1285*2004 + 0.1135*2005 + 0.1723*2006 +
0.0908*2007 -0.5673*2008 -0.1257*2009 -
0.0028*2010
(21)
Ln(PE) = 2.0608 -0.4683* Ln(ROE) + 0.1005*
Ln(Net Margin) -0.0359*2000 +0.0202*2001
+ 0.3556*2003 + 0.4529*2004 + 0.3781*2005
+ 0.4028*2006 + 0.2765*2007 -0.524*2008
+0.0466*2009+0.2228*2010
(22)
US and Canada is the second region with more data
available and it covers all years from 2000 to 2010. The
base year in both equations is 2002, where the coeffcient is
calculated at zero in both equations. All F-tests and T-tests
show that the linear regression model is a good model for
the proposed relationships and that all main factors are
signifcantly different from zero, including the intercept.
The resulting R2 was 12.81% for the EV/EBITDA multiple
and 23.64% for the P/E multiple, being the latter higher.
World
Ln(EV/EBITDA) = 1.6362 -0.3283* Ln(ROC) + 0.1219*
Ln(Pre-tax Operating Margin) -0.2074*2001 -0.0215*2002
+ 0.0957*2003 + 0.1519*2004 + 0.2482*2005 + 0.303*2006
+0.275*2007 -0.4338*2008+0.1507*2009+0.3397*2010
(23)
Ln(PE) = 1.6897 -0.4203* Ln(ROE) -0.0388*
Ln(Net Margin) + 0*2000 +0.0874*2001 +
0.0142*2002 + 0.0501*2003 + 0.1674*2004 +
0.2707*2005 + 0.2473*2006 + 0.0878*2007 -
0.4883*2008 + 0.1215*2009 + 0.2928*2010
(24)
World includes all available information from all
the previous regions, so data for years 2000 and 2001
correspond to the US and Canada region only. In both
equations the base year is 2000. F-tests and T-tests show
that the linear regression model fts the relationships for
both multiples, including intercepts. The overall R2 was
17.31% for the EV/EBITDA multiple and 10.87% for the
P/E multiple.
BRIC
Ln(EV/EBITDA) = 1.8929 -0.2861* Ln(ROC)
-0.0305* Ln(Pre-tax Operating Margin) -
0.0319*2003 -0.2788*2004 -0.207*2005 -
0.0374*2006 + 0.2486*2007 -0.4303*2008 +
0.2656*2009+0.4456*2010
(25)
Ln(PE) = -5.85E+11+ -0.5834* Ln(ROE) + 0.033*
Ln(Net Margin) +5.85E+11*2002 + 5.85E+11*2003 +
5.85E+11*2004 + 5.85E+11*2005 + 5.85E+11*2006 +
5.85E+11*2007 + 5.85E+11*2008 + 5.85E+11*2009 +
5.85E+11*2010
(26)
Finally, the BRIC region was constructed from a sample
including countries from several regions, so data was
available for years 2002 to 2010. Except for the intercept
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
280 Business Intelligence Journal July
Table 4. REVAAM EV/EBITDA multiples (short term)
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 7.33 7.84 8.33 10.42 13.96 13.29 4.93 6.93 7.92
Asia NA NA 12.98 11.14 10.45 11.39 11.86 12.72 5.73 10.45 12.07
Europe_east NA NA NA 6.05 6.18 7.77 7.44 7.28 5.03 7.44 9.24
Europe_west NA NA 8.35 10.06 11.25 12.38 14.16 10.87 5.49 8.04 9.31
Latam&Carib NA NA 7.10 7.03 7.72 8.16 9.94 9.99 5.05 9.31 10.17
Middle East NA NA 16.59 14.53 18.14 20.32 14.17 16.21 7.52 9.14 9.89
Oceania NA NA NA 11.55 11.86 11.41 13.62 11.65 4.98 9.94 9.25
US&Canada 7.57 5.91 5.53 7.75 8.68 8.41 8.87 8.17 4.25 6.53 7.53
World 7.60 5.90 7.98 9.39 9.82 10.55 11.16 10.82 5.20 8.89 10.42
BRIC NA NA 15.81 14.49 11.17 11.53 13.73 17.69 8.84 17.18 18.89
Table 5. REVAAM P/E multiples (short term)
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 7.33 7.70 6.85 13.06 16.15 18.83 7.16 11.15 15.03
Asia NA NA 16.85 16.68 18.22 18.89 19.51 19.55 10.08 19.70 20.99
Europe_east NA NA NA 16.44 9.97 13.31 4.00 2.65 6.02 13.87 17.46
Europe_west NA NA 15.14 15.18 18.51 20.76 23.81 15.79 7.76 14.60 18.13
Latam&Carib NA NA 11.25 4.45 4.23 8.40 10.18 5.56 7.81 15.74 16.04
Middle East NA NA 18.47 3.49 12.42 13.33 8.97 10.70 6.97 12.51 13.90
Oceania NA NA NA 19.14 19.41 18.72 21.54 15.80 6.42 19.34 17.65
US&Canada 14.31 16.10 16.56 23.51 25.84 23.50 23.75 20.82 9.42 16.32 20.97
World 13.86 15.98 15.62 16.45 18.02 19.13 18.62 15.55 9.09 17.23 19.77
BRIC NA NA 27.09 27.88 18.68 18.06 15.31 16.39 14.26 34.96 34.74
The following graphs show the multiples from tables 4
and 5 for each region and for each year. From Graph 1 it
can be seen that:
1. All regions reduced their multiples in 2008, in the
middle of the crisis
2. US and Canada region, show a decline in EV/EBITDA
multiples for 2001 and 2002 and then a recovery until
the fall of 2008.
3. From 2004 to 2007, the US and Canada region had
adjusted P/E multiples above other regions, even higher
than the BRIC region.
of equation (26) all other factors are statistically signifcant
different from zero. We will keep this coeffcient in our
formulas for purposes of our analysis. Also, coeffcients
for years and for intercept in equation (26) are very high.
R2 was 13.57% for the EV/EBITDA multiple and 9% for
the P/E multiple.
Now, for each year we apply the equations (7) to (25)
-those with odd number- to obtain the corresponding
multiples for each year. The factors (ROC, ROE, Pre-
tax Operating Margin and Net Margin) used in each case
correspond to the median values for each region in each
year. Table 4 shows the results for adjusted EV/EBITDA
multiples and Table 5 for adjusted P/E multiples.
2011 281
4. Middle East and BRIC regions have presented the
highest adjusted EV/EBITDA multiples. For the P/E
multiples, BRIC was the highest for years 2002-2003,
2008-2010.
5. The BRIC region had the highest EV/EBITDA and
P/E adjusted multiples during the crisis of 2008 and
recovered in one year and reached historical values in
2010.
6. Except for the BRIC region, variability of EV/EBITDA
multiple in 2002 has been reduced from 2002 to 2010.
Ranges in 2002, excluding BRIC, were between 5.5
and 16.5, whereas in 2010 multiples were between 7.5
and 12; that is a reduction in size of the range of 59%.
7. In the same token, variability of the P/E multiple was
also reduced in the same period in size of range of
64.5%, excluding the BRIC region.
0
5
10
15
20
25
30
35
40
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa Asia Europe_east Europe_west Latam&Carib
Middle East Oceania US-Canada World BRIC
Graph 2. Adjusted REVAAM P/E multiples per region per year
After we have obtained our adjusted multiples yearly
for each region, we will use equations (8) to (26) -those
with even number- to obtain the long term multiples of each
region. Then we calculate the multiples using the median
values of the factors (ROC, ROE, Pre-tax Operating Margin
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 10.24 9.22 9.22 9.42 9.40 9.46 9.19 8.42 8.29
Asia NA NA 12.24 11.67 11.29 10.93 11.00 10.91 10.73 10.06 9.56
Europe_east NA NA NA 7.04 7.51 6.98 7.23 7.12 7.23 7.23 7.10
Europe_west NA NA 9.61 10.77 10.68 10.37 10.22 10.06 9.18 8.64 8.44
Latam&Carib NA NA 8.99 8.72 8.46 8.39 8.27 8.34 8.13 7.65 7.30
Middle East NA NA 17.27 15.72 15.20 14.29 14.24 14.78 13.38 11.72 11.06
Oceania NA NA NA 11.05 10.63 10.39 10.53 10.49 9.77 8.80 9.20
US&Canada 7.12 6.95 7.04 7.23 7.18 7.07 7.03 7.02 7.07 6.97 7.11
World 8.67 8.28 9.29 9.73 9.61 9.38 9.40 9.37 9.14 8.71 8.45
BRIC NA NA 16.43 15.79 15.67 15.21 15.25 14.80 14.69 14.46 13.49
and Net Margin) in each year. Although the formulas for
the long term multiples do not differentiate each year, we
apply the value of the factors to obtain the multiple of that
year using the long term formulae. Tables 6 and 7 show
these results.
Table 6. REVAAM EV/EBITDA multiples (long term)
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
282 Business Intelligence Journal July
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 11.64 11.32 11.33 11.33 11.49 11.59 11.71 11.78 12.12
Asia NA NA 17.47 19.58 18.47 17.25 17.20 16.87 17.84 18.57 16.61
Europe_east NA NA NA 6.91 6.09 5.93 5.94 5.83 5.77 6.86 6.02
Europe_west NA NA 17.47 16.19 16.04 15.54 15.52 14.71 15.44 17.21 16.64
Latam&Carib NA NA 9.50 9.69 9.16 8.61 8.71 8.55 8.70 8.41 7.79
Middle East NA NA 12.05 11.16 10.71 9.98 9.38 10.13 9.62 11.44 11.42
Oceania NA NA NA 16.60 15.91 15.57 15.52 15.09 16.10 16.39 17.62
US&Canada 17.22 18.31 19.23 19.13 19.07 18.69 18.43 18.32 18.46 18.08 19.48
World 15.14 15.98 16.80 17.07 16.62 15.92 15.87 15.54 16.16 16.65 16.10
BRIC NA NA 25.50 25.48 23.81 20.35 20.51 19.33 20.35 21.02 18.15
Table 7. REVAAM P/E multiples (long term)
1. All regions long term multiples are smoothed and with
little variability, even though each year we calculated the
corresponding multiple using that year's median values
for ROE, ROC, Net Margin and Pre-tax Operating
Margin. This means that the long term equations are
not taking into consideration other external factors
rather than the fundamental and intrinsic aspects of
each company.
2. Average long term multiples for each region are:
Region EV/EBITDA P/E
Africa 9.21 11.59
Asia 10.93 17.76
Europe_east 7.18 6.17
Europe_west 9.78 16.08
Latam&Carib 8.25 8.79
Middle East 14.18 10.65
Oceania 10.11 16.10
US&Canada 7.07 18.58
World 9.09 16.17
BRIC 15.09 21.61
3. Mature markets, such as Western Europe, Oceania,
United States and Canada, and the World show a clear
gap between long term multiples EV/EBITDA and
P/E, being the former lower than the latter. In contrast,
emerging markets do show the same relationship, but
not as clear, and for the case of Middle East and Eastern
Europe, P/E multiple is lower than EV/EBITDA
Multiple.
4. Average long term multiples for mature markets are 16
to 18 in the case of P/E and 7 to 10 for EV/EBITDA,
whereas for emerging markets P/E ranges from 6 to 21
and from 7 to 15 in the case of EV/EBITDA.
5. All regions showed that for the crisis of 2007-2008
their short term or observed multiples were below the
long term multiples for that period, but all were above
the long term multiple in 2010. However, mature
markets' short term multiples for 2010 were slightly
above the long term multiples for that year; contrasting
with emerging markets whose observed multiples were
relatively higher than the long term multiples for 2010.
This may mean that investors cash was sent to emerging
markets seeking for better risk adjusted returns, but
also means that these regions are overvalued according
to long term multiples.
Table 8. Long term REVAAM average multiples
2011 283
Graph 3. Long and short term multiple comparison for African
region
Graph 4. Long and short term multiple comparison for Asian
region
Graph 5. Long and short term multiple comparison for Eastern
Europe region
-
5
10
15
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Asia
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
-
5
10
15
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Eastern Europe
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 6. Long and short term multiple comparison for Western
Europe region
-
5
10
15
20
25
30
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Western Europe
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 7. Long and short term multiple comparison for Latin-
America and the Caribbean region
-
5
10
15
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Latin-America and the Caribbean
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 8. Long and short term multiple comparison for Middle
East region
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Middle East
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
284 Business Intelligence Journal July
Graph 9. Long and short term multiple comparison for Oceania
region
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Oceania
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 10. Long and short term multiple comparison for Us and
Canada region
-
5
10
15
20
25
30
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
US and Canada
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 11. Long and short term multiple comparison for World
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
World
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 12. Long and short term multiple comparison for BRIC
region
-
10
20
30
40
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
BRIC
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Hypothesis 1 established that adjusted long term and
short term multiples for emerging or underdeveloped
regions should be higher than those in mature markets.
When we compared the adjusted short term EV/EBITDA
multiples we obtained a high variability year by year and
there is no suffcient evidence that emerging regions have
higher multiples than mature regions, except for the BRIC
and Middle East region; being the latter converging to
mature markets multiples after 2008. Then, looking at the
long term adjusted multiples, we found that for the EV/
EBITDA multiple, only Asia, Middle East and the BRIC
region were above the mature regions (United States and
Canada, Oceania and Western Europe). Regarding the
adjusted P/E multiple, fndings are not the same. For the
short term P/E adjusted multiples, only the BRIC region was
above mature markets multiples for all years except from
2004 to 2007. In 2010, BRIC region multiples sky rocketed
and also Asian multiples surpassed those of mature markets.
All other emerging regions' P/E multiple remained below
those of mature regions, but began increasing after 2008.
Now, comparing long term adjusted P/E multiples, the
only region above mature markets was the BRIC, which in
2010 converged to mature markets multiples. Additionally,
Asia's P/E adjusted multiple is within mature market range,
while the rest of the emerging regions are below mature
markets for all the period range.
Therefore, we could not fnd enough evidence to support
Hypothesis 1, thus we reject it. However, in individual
cases, we could on accept it for the BRIC region which
in all cases appears to have adjusted EV/EBITDA and P/E
multiples above those observed in mature markets.
2011 285
Hypothesis 2 suggests that yearly multiples - those
calculated from equations with odd number (7) to (25) - will
signifcantly differ from the adjusted long term multiples in
each market. By looking at graphs 3 to 12, we can see that
adjusted short term multiples do vary more than adjusted
long term multiples. The following table shows the standard
deviation for each multiple per region.
Table 9. Standard deviations for adjusted short and long term
multiples
Region
E
V
/
E
B
I
T
D
A
S
h
o
r
t
T
e
r
m
E
V
/
E
B
I
T
D
A
L
o
n
g
T
e
r
m
P
/
E
S
h
o
r
t
T
e
r
m
P
/
E
L
o
n
g
T
e
r
m
Africa 2.99 0.58 4.51 0.26
Asia 2.16 0.80 3.22 0.95
Europe_east 1.28 0.16 5.70 0.45
Europe_west 2.57 0.86 4.50 0.90
Latam&Carib 1.73 0.51 4.44 0.58
Middle East 4.35 1.93 4.34 0.92
Oceania 2.60 0.78 4.67 0.78
US&Canada 1.47 0.09 5.02 0.64
World 2.00 0.49 2.97 0.57
BRIC 3.37 0.86 8.24 2.67
So statistically speaking, deviations for adjusted short
term multiples are remarkably higher than those for adjusted
long term multiples, so we would accept Hypothesis 2 for
all regions. However, we need to notice that variability for
EV/EBITDA multiples is considerably lower than that for
P/E multiples, both in the short term. Another important
fnding is that, theoretically speaking, EV/EBITDA
multiples are lower than P/E multiples, because EBITDA
is higher than NI. In our adjusted multiples, we fnd that all
mature, BRIC and Asia regions present a clear gap between
both multiples; nevertheless, for Latin-America and the
Caribbean, and Africa the gap is small, and for Middle East
and Eastern Europe the relationship is inverse, adjusted P/E
multiples are higher than EV/EBITDA multiples.
It is important to mention that these analysis could have
been performed by using simple average for the multiples
used, however, using REVAAM model provides adjusted
multiples that should be more accurate than using simple
average multiples. Besides, what determines the fair market
value of a company is not just a multiple, but it should take
into consideration proftability parameters that in the end
determine the free cash fow for each specifc company,
and those companies with higher proftability should be
awarded a higher valuation, thus, higher multiples. In
appendix 5 we can see that using average multiples will
lead to higher values, even more for the P/E multiple.
Discussion and Conclusions
We used REVAAM model, with its limitations, to
calculate P/E and EV/EBITDA adjusted multiples for
regions of the world (some belong to emerging markets,
other to mature markets and few with countries in both
markets) and compare them both in the short run (year by
year) and in the long run. The use of REVAAM, through
multiple linear regression, let us differentiate proftability
factors at level company which in the end are what
determine the cash fow from each company, thus its value,
rather than applying just an arithmetic average of multiples.
General valuation literature suggests that multiples in
emerging markets are higher than mature markets, mainly
because risk-return analysis should award a better return in
economies that are not yet fully developed. However, as
we tested this hypothesis, we found that empirical evidence
does not support this proposal, except for the BRIC region.
One may argue that in order to get a higher return, investors
buy assets whose prices (multiples) are low and then try to
sell them at higher prices. Nevertheless, results showed
that emerging markets multiples were not higher than those
in mature markets, not in the short nor in the long term;
actually, in the long term multiples for emerging regions
were below those in mature regions.
We need to consider also that our time frame included
two crises, the one in 2001-2002 and the one in 20007-2008.
If we analyze the short term multiples, we could see that
in both periods multiples plunged but made a recovery in
the following years; two years for recovery after the 2001-
2002 crisis and just one year after that of 2007-2008, so
investors are already assuming that valuations in crises do
fall, but eventually need to recover along with government
intervention (i.e. US government bailout of banks and
European Central Bank funding to Greece fnancial
system). Additionally, variation in multiples was still high
among regions after the 2001-2002 crisis, whether after the
2007-2008 crisis this variability decreased signifcantly,
suggesting that valuation ranges for most regions are based
more on fundamentals or similar to mature markets. This
is not true for the BRIC region which during the 2007-
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
286 Business Intelligence Journal July
2008 crisis had the largest fall in multiples but presented
the highest multiples than any other region during recovery,
even higher than past levels. This effect could be caused
by investors preference or current trend to invest in these
countries due to their economic potential.
Besides comparing emerging to mature regions in
terms of valuation multiples, we wanted to compare long
term multiples which are not differentiated by year, but by
proftability factors. The reason behind this is that although
fnancial analysts believe that markets are rational there are
many external factors that affect valuations without really
affecting proftability factors of the companies. It is true
that economical shocks due affect interest rates, infation
and economic growth perspectives, and these variables
are also inputs in valuation; however, valuations should be
performed in the long run and must include these shocks in
the adequate short term frame. Actually, the most important
driver in free cash fow valuations is the terminal value
which takes into consideration long term proftability of the
company matched with long term conditions.
Results showed that adjusted short term multiples had
a considerably higher variability than adjusted long term
multiples. Since formulae for short and long term multiples
are differentiated by specifc years, we can conclude that
deviations from the adjusted long term multiple are more
speculation bubbles than true change in fundamental
value, and this is also supported by the fact that multiples
recovered in two years, and recently in only one year after
the crisis. Also, comparing short term to long term multiples
allows to somehow realize if a company, industry or region
is under or overvalued, so if we believe that in the long
run that company, industry or region should converge to the
long term multiple, we could decide to invest or short sell
to make a proft. In this token, I would say that the BRIC
region is overvalued. Of course, as we said later, other
macroeconomic and external factors do affect companies´
and industries´ valuations, but should not affect an entire
region. It is opinion of the author of this paper, that when
using multiple valuation analysts should consider the
following:
a. Using a method that incorporates factors that
differentiates one company from another, such as
REVAAM, rather than just using a simple average of
multiples, even within an industry. It is possible to add
additional factors such as macroeconomic or industry
specifc factors that affect valuation of our companies,
if information is available.
b. Regarding the last point, multiples should be used at
company level, as industry or region multiples are the
aggregated sum of all corresponding companies. For
our analysis, we ran regression models using raw data
from companies, not aggregated information.
c. It is better to use adjusted long term multiples than short
term multiples, because the former are less infuenced
by speculation and do adjust in time with a trend.
d. Prefer EV/EBITDA multiples over P/E multiples as the
latter present more variability than the former. The only
exception would be fnancial institutions and some
service companies in which EBITDA is not a natural
or easy measure.
e. If possible, use multiple valuation if free cash fow
valuation is not possible, as valuation is the present
value of actual and future cash fows.
It will be interesting to calculate multiples at the end of
2011 for all these regions in order to see how the Middle
East conficts affect valuations in nearby regions and in the
world, as it has affected oil prices and interest rates, but
even more investors perspectives on risk and speculations
about the future.
Finally, I would like to mention that using REVAAM
for adjusted multiples in the Middle East and Eastern
Europe resulted that P/E multiples were lower than EV/
EBITDA multiples, but when looking at average raw data
the opposite is true. This is because distributions for those
regions were skewed and regression analysis takes an
average of the distribution, which is not necessarily at the
50% of the sample.
References
Acosta-Calzado, Carlos et al. 2010. "REVAAM
Model to determine a company's value by multiple
valuation and linear regression anaylisis" Business
Intelligence Journal Volume 3, Number 2.
Schreiner, Andreas. 2007. “Equity Valuation Using
Multiples: An Empirical Investigation” Dissertation
der Universität St. Gallen (HSG)
Damodaran, Aswath. 2002. “Investment Valuation:
Tools and Techniques for Determining the Value of
Any Asset” Second Edition.
McKinsey & Company Inc., et al. 2005. “Valuation:
Measuring and Managing the Value of Companies”,
Edition, University Edition.
2011 287
Damodaran, Aswath. 2010. “The Dark Side of
Valuation: Valuing Young, Distressed, and Complex
Businesses”, 2nd Edition, Pearson Education, Inc.
Medlen, Craig. 2007. “The bubble machine: relative
capital valuation, distributive shares and capital
gains” An article from: Journal of Economic Issues.
Iwanowski, Raymond J. 1995. “An investor's guide
to foating-rate notes: Conventions, mathematics
and relative valuation (United States fxed-income
research. Portfolio strategues)”
Hoover, Scott. 2005. “Stock Valuation: An Essential
Guide to Wall Street's Most Popular Valuation
Models” McGraw-Hill Library of Investment and
Finance.
Frykman, David and Tolleryd, Jakob. 2003. “Corporate
Valuation: an easy guide to measuring value”.
Myers, Brealey. 2002. “Principles of Corporate
Finance”.
Pagano, Robert R. 1998. “Estadística para las Ciencia
del Comportamiento”, 5ª Edición.
Beaver, W. 1967. “Financial ratios as predictors of
failure”, Empirical Research in Accounting. (1966),
“Selected studies”, supplement to the Journal of
Accounting Research, January, 1966.
Greene W. 1993. “Econometric Analysis”, 2th Edition,
Mc. Millan Publishing Co.
Mandala G.S. 1983. “Limited Dependent and
Qualitative Variables in Econometrics”,
Econometric Society, Monographs number 3,
Cambridge, University.
Bloomberg Professional monitoring service.
Yahoo Finance (http://fnance.yahoo.com)
Capital IQ (http://www.capitaliq.com)
Value Line (http://www.valueline.com)
Ibkan Consultants Financial Data (http://www.ibkan.
com.mx)
Appendix 1
REVAAM Regression results for EV/EBITDA multiple for each region per year (short term)
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
1.71E+12 8.37E+11 0.9012 1.8465 1.5172 1.9015 1.9298 1.6474 1.6362 1.8929
?
ROC
(0.3623) (0.3194) (0.2276) (0.3479) (0.3317) (0.4348) (0.3244) (0.2439) (0.3283) (0.2861)
?
OM
0.1071 0.1527 (0.0934) 0.1822 0.0895 0.1815 0.0872 0.0492 0.1219 (0.0305)
?
2000
0.0000 0.0000 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
?
2001
0.0000 0.0000 0 0.0000 0.0000 0.0000 0.0000 (0.2262) (0.2074) 0.0000
?
2002
-1.7E+12 -8.4E+11 0 (0.0626) (0.2196) 0.0000 0.0000 (0.3049) (0.0215) 0.0000
?
2003
-1.7E+12 -8.4E+11 0.1720 0.0152 (0.1945) (0.0406) (0.0629) 0.0087 0.0957 (0.0319)
?
2004
-1.7E+12 -8.4E+11 0.1303 0.1348 (0.0642) 0.2154 0.0000 0.1285 0.1519 (0.2788)
?
2005
-1.7E+12 -8.4E+11 0.4316 0.2591 0.0000 0.3912 (0.0168) 0.1135 0.2482 (0.2070)
?
2006
-1.7E+12 -8.4E+11 0.3527 0.4073 0.2160 0.0338 0.1463 0.1723 0.3030 (0.0374)
?
2007
-1.7E+12 -8.4E+11 0.3464 0.1581 0.2106 0.1307 (0.0070) 0.0908 0.2750 0.2486
?
2008
-1.7E+12 -8.4E+11 (0.0351) (0.4386) (0.4391) (0.5468) (0.7868) (0.5673) (0.4338) (0.4303)
?
2009
-1.7E+12 -8.4E+11 0.3574 0.0000 0.2490 (0.2345) 0.0083 (0.1257) 0.1507 0.2656
?
2010
-1.7E+12 -8.4E+11 0.5927 0.1693 0.3912 (0.1019) (0.1072) (0.0028) 0.3397 0.4456
R
2
0.2623 0.1524 0.0714 0.3139 0.1741 0.2996 0.3017 0.1281 0.1731 0.1357
N 2,618 66,905 4,665 24,604 4,564 3,309 3,292 42,048 152,005 23,291
d.f. 2,606 66,893 4,654 24,593 4,553 3,298 3,282 42,035 151,992 23,280
F value 84.2193 1,093.219 35.79 1,124.991 95.9562 141.1046 157.5764 514.4430 2,650.741 365.0390
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.5558 1.6562 0.6839 100.9945 32.0016 29.3768 44.0296 101.9062 109.1892 54.3584
t
?ROC
(22.5541) (83.1291) (11.5578) (83.2199) (23.3648) (31.0492) (26.6464) (49.5357) (138.558) (36.248)
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
288 Business Intelligence Journal July
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
7.60E+12 -7.6E+11 1.5345 1.4909 1.4069 -1.9E+12 1.9966 2.0608 1.6897 -5.8E+11
?
ROC
(0.1170) (0.4710) (0.4556) (0.3908) (0.5303) (0.4064) (0.4642) (0.4683) (0.4203) (0.5834)
?
OM
(0.0752) (0.0444) 0.1702 (0.1129) 0.0559 (0.1208) (0.0399) 0.1005 (0.0388) 0.0330
?
2000
0 0 0.0000 0 0 0 0 (0.0359) 0 0
?
2001
0 0 0.0000 0 0 0 0 0.0202 0.0874 0
?
2002
-7.6E+12 7.57E+11 0.0000 0.0212 0 1.91E+12 0 0 0.0142 5.8E+11
?
2003
-7.6E+12 7.57E+11 0.5978 0.1008 (0.9500) 1.91E+12 (0.0576) 0.3556 0.0501 5.8E+11
?
2004
-7.6E+12 7.57E+11 0.2182 0.3085 (0.9394) 1.91E+12 0 0.4529 0.1674 5.8E+11
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
t
?NM
6.0593 38.7113 -4.2165 41.8399 6.0297 12.5162 7.0948 11.3174 51.1871 (3.5417)
t
2000
N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
t
2001
N/A N/A N/A N/A N/A N/A N/A (12.6226) (10.6053) N/A
t
2002
(0.5558) (1.6562) N/A (3.2943) (4.3463) N/A N/A (16.8566) (1.3046) N/A
t
2003
(0.5558) (1.6562) 0.1358 0.7854 (3.7166) (0.5643) (1.3230) 0.4941 6.0027 (0.8401)
t
2004
(0.5558) (1.6562) 0.1029 7.1434 (1.2448) 3.1926 N/A 7.2597 9.6603 (7.5224)
t
2005
(0.5558) (1.6562) 0.3409 13.8234 N/A 5.9976 (0.3754) 6.4263 15.9548 (5.8450)
t
2006
(0.5558) (1.6562) 0.2786 22.0669 4.3229 0.5403 3.2156 9.8756 19.5894 (1.0732)
t
2007
(0.5558) (1.6562) 0.2737 8.6446 4.2859 2.1150 (0.1525) 5.1209 17.8437 7.3126
t
2008
(0.5558) (1.6562) (0.0277) (23.2462) (8.7614) (7.3358) (17.6316) (32.0176) (27.7688) (11.5434)
t
2009
(0.5558) (1.6562) 0.2822 N/A 4.9946 (3.3132) 0.1858 (7.0525) 9.7150 7.7845
t
2010
(0.5558) (1.6562) 0.4680 8.8024 7.9484 (1.5135) (2.4138) (0.1566) 22.1371 13.2546
Prob(t
?EVEBITDA
)
0.5784 0.0977 0.4940 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?ROC
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004
Prob(t
2000
)
N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Prob(t
2001
)
N/A N/A N/A N/A N/A N/A N/A 0.0000 0.0000 N/A
Prob(t
2002
)
0.5784 0.0977 N/A 0.0010 0.0000 N/A N/A 0.0000 0.1920 N/A
Prob(t
2003
)
0.5784 0.0977 0.8920 0.4322 0.0002 0.5726 0.1859 0.6212 0.0000 0.4009
Prob(t
2004
)
0.5784 0.0977 0.9180 0.0000 0.2133 0.0014 N/A 0.0000 0.0000 0.0000
Prob(t
2005
)
0.5784 0.0977 0.7332 0.0000 N/A 0.0000 0.7074 0.0000 0.0000 0.0000
Prob(t
2006
)
0.5784 0.0977 0.7805 0.0000 0.0000 0.5890 0.0013 0.0000 0.0000 0.2832
Prob(t
2007
)
0.5784 0.0977 0.7843 0.0000 0.0000 0.0345 0.8788 0.0000 0.0000 0.0000
Prob(t
2008
)
0.5784 0.0977 0.9779 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
2009
)
0.5784 0.0977 0.7778 N/A 0.0000 0.0009 0.8526 0.0000 0.0000 0.0000
Prob(t
2010
)
0.5784 0.0977 0.6398 0.0000 0.0000 0.1302 0.0158 0.8756 0.0000 0.0000
Appendix 2
REVAAM Regression results for PE multiple for each region per year (short term)
2011 289
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
2005
-7.6E+12 7.57E+11 0.5321 0.4554 (0.1834) 1.91E+12 (0.0141) 0.3781 0.2707 5.8E+11
?
2006
-7.6E+12 7.57E+11 (0.6714) 0.5943 (0.0044) 1.91E+12 0.1294 0.4028 0.2473 5.8E+11
?
2007
-7.6E+12 7.57E+11 (1.0663) 0.2374 (0.5897) 1.91E+12 (0.1522) 0.2765 0.0878 5.8E+11
?
2008
-7.6E+12 7.57E+11 (0.2348) (0.5214) (0.2693) 1.91E+12 (1.1194) (0.5240) (0.4883) 5.8E+11
?
2009
-7.6E+12 7.57E+11 0.4339 0 0.4695 1.91E+12 (0.0343) 0.0466 0.1215 5.8E+11
?
2010
-7.6E+12 7.57E+11 0.7888 0.2509 0.5748 1.91E+12 (0.2005) 0.2228 0.2928 5.8E+11
R
2
0.0473 0.1439 0.0498 0.2228 0.0548 0.0538 0.3341 0.2364 0.1087 0.0900
N 2,880 62,193 4276 23,530 4,436 3,579 3,365 36,276 140,535 19,729
d.f. 2,868 62,181 4265 23,519 4,425 3,567 3,355 36,263 140,522 19,717
F value 12.94 949.91 22.3593 674.16 25.63 18.44 187.07 935.48 1,428.50 177.20
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.46 (0.99) 0.4197 49.79 7.95 (0.06) 33.15 95.32 55.14 (0.13)
t
?ROC
(2.90) (65.41) (6.8452) (46.55) (9.60) (8.05) (26.50) (73.84) (83.94) (29.81)
t
?NM
(1.98) (7.21) 2.8127 (16.02) 1.14 (2.99) (2.56) 17.58 (8.97) 1.77
t
2000
N/A N/A N/A N/A N/A N/A N/A -1.44 N/A N/A
t
2001
N/A N/A N/A N/A N/A N/A N/A 0.84 2.24 N/A
t
2002
(0.4566) 0.9918 N/A 0.6915 N/A 0.0565 N/A N/A 0.4330 0.1287
t
2003
(0.4566) 0.9918 0.1634 3.1325 (4.6847) 0.0565 N/A 14.8999 1.5742 0.1287
t
2004
(0.4566) 0.9918 0.0596 9.8559 (4.8645) 0.0565 0.0000 18.9897 5.3303 0.1287
t
2005
(0.4566) 0.9918 0.1455 14.6807 (0.9499) 0.0565 (0.2221) 16.0700 8.7124 0.1287
t
2006
(0.4566) 0.9918 (0.1837) 19.8478 (0.0238) 0.0565 2.0834 17.3060 8.0144 0.1287
t
2007
(0.4566) 0.9918 (0.2918) 7.9858 (3.2324) 0.0565 (2.4482) 11.7364 2.8571 0.1287
t
2008
(0.4566) 0.9918 (0.0642) (17.1806) (1.4704) 0.0565 (18.4664) (22.3193) (15.7557) 0.1287
t
2009
(0.4566) 0.9918 0.1187 N/A 2.5145 0.0565 (0.5692) 1.9188 3.8709 0.1287
t
2010
(0.4566) 0.9918 0.2157 7.9523 3.1689 0.0565 (3.2785) 9.1156 9.4601 0.1287
Prob(t
?EVEBITDA
)
0.6480 0.3213 0.6747 0.0000 0.0000 0.9549 0.0000 0.0000 0.0000 0.8976
Prob(t
?ROC
)
0.0038 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0483 0.0000 0.0049 0.0000 0.2537 0.0028 0.0107 0.0000 0.0000 0.0773
Prob(t
2000
)
N/A N/A N/A N/A N/A N/A N/A 0.1487 N/A N/A
Prob(t
2001
)
N/A N/A N/A N/A N/A N/A N/A 0.3989 0.0251 N/A
Prob(t
2002
)
0.6480 0.3213 N/A 0.4892 N/A 0.9549 N/A N/A 0.6650 0.8976
Prob(t
2003
)
0.6480 0.3213 0.8702 0.0017 0.0000 0.9549 N/A 0.0000 0.1154 0.8976
Prob(t
2004
)
0.6480 0.3213 0.9524 0.0000 0.0000 0.9549 1.0000 0.0000 0.0000 0.8976
Prob(t
2005
)
0.6480 0.3213 0.8844 0.0000 0.3422 0.9549 0.8242 0.0000 0.0000 0.8976
Prob(t
2006
)
0.6480 0.3213 0.8543 0.0000 0.9810 0.9549 0.0373 0.0000 0.0000 0.8976
Prob(t
2007
)
0.6480 0.3213 0.7704 0.0000 0.0012 0.9549 0.0144 0.0000 0.0043 0.8976
Prob(t
2008
)
0.6480 0.3213 0.9488 0.0000 0.1415 0.9549 0.0000 0.0000 0.0000 0.8976
Prob(t
2009
)
0.6480 0.3213 0.9056 N/A 0.0120 0.9549 0.5693 0.0550 0.0001 0.8976
Prob(t
2010
)
0.6480 0.3213 0.8292 0.0000 0.0015 0.9549 0.0011 0.0000 0.0000 0.8976
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
290 Business Intelligence Journal July
Appendix 3
REVAAM Regression results for EV/EBITDA multiple for each region for all years (long term)
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
1.71E+12 8.37E+11 0.9012 1.8465 1.5172 1.9015 1.9298 1.6474 1.6362 1.8929
?
ROC
(0.3623) (0.3194) (0.2276) (0.3479) (0.3317) (0.4348) (0.3244) (0.2439) (0.3283) (0.2861)
?
OM
0.1071 0.1527 (0.0934) 0.1822 0.0895 0.1815 0.0872 0.0492 0.1219 (0.0305)
R
2
0.2623 0.1524 0.0714 0.3139 0.1741 0.2996 0.3017 0.1281 0.1731 0.1357
N 2,618 66,905 4,665 24,604 4,564 3,309 3,292 42,048 152,005 23,291
d.f. 2,606 66,893 4,654 24,593 4,553 3,298 3,282 42,035 151,992 23,280
F value 84.2193 1,093.219 35.79 1,124.991 95.9562 141.1046 157.5764 514.4430 2,650.741 365.0390
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.5558 1.6562 0.6839 100.9945 32.0016 29.3768 44.0296 101.9062 109.1892 54.3584
t
?ROC
(22.5541) (83.1291) (11.5578) (83.2199) (23.3648) (31.0492) (26.6464) (49.5357) (138.558) (36.248)
t
?NM
6.0593 38.7113 -4.2165 41.8399 6.0297 12.5162 7.0948 11.3174 51.1871 (3.5417)
Prob(t
?EVEBITDA
)
0.5784 0.0977 0.4940 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?ROC
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004
Appendix 4
REVAAM Regression results for P/E multiple for each region for all years (long term)
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
1.71E+12 8.37E+11 0.9012 1.8465 1.5172 1.9015 1.9298 1.6474 1.6362 1.8929
?
ROC
(0.3623) (0.3194) (0.2276) (0.3479) (0.3317) (0.4348) (0.3244) (0.2439) (0.3283) (0.2861)
?
OM
0.1071 0.1527 (0.0934) 0.1822 0.0895 0.1815 0.0872 0.0492 0.1219 (0.0305)
R
2
0.2623 0.1524 0.0714 0.3139 0.1741 0.2996 0.3017 0.1281 0.1731 0.1357
N 2,618 66,905 4,665 24,604 4,564 3,309 3,292 42,048 152,005 23,291
d.f. 2,606 66,893 4,654 24,593 4,553 3,298 3,282 42,035 151,992 23,280
F value 84.2193 1,093.219 35.79 1,124.991 95.9562 141.1046 157.5764 514.4430 2,650.741 365.0390
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.5558 1.6562 0.6839 100.9945 32.0016 29.3768 44.0296 101.9062 109.1892 54.3584
t
?ROC
(22.5541) (83.1291) (11.5578) (83.2199) (23.3648) (31.0492) (26.6464) (49.5357) (138.558) (36.248)
t
?NM
6.0593 38.7113 -4.2165 41.8399 6.0297 12.5162 7.0948 11.3174 51.1871 (3.5417)
Prob(t
?EVEBITDA
)
0.5784 0.0977 0.4940 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?ROC
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004
2011 291
Appendix 5
Comparison graphs among REVAAM multiples and Average multiples
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
15
30
45
60
75
90
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Asia
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Eastern Europe
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Western Europe
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Latin-America & Caribbean
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
10
20
30
40
50
60
70
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Middle East
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
292 Business Intelligence Journal July
-
10
20
30
40
50
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Oceania
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
10
20
30
40
50
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
US and Canada
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
10
20
30
40
50
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
World
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
20
40
60
80
100
120
140
160
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
BRIC
EV/EBITDA ST
EV/EBITDA Average
P/E ST
PE Average
2011 293
MARKETING PRINCIPLES OF A NEGLECTED BANKING
SERVICE: SAFE DEPOSIT BOX RENTAL
Monireh Panahi
PhD candidate in Strategic Management in Shahid Beheshti University. Marketing consultant; author of
articles on marketing and strategic management in Iranian journals.
School of Management & Accounting, Shahid Beheshti University, Daneshjoo Blvd., Evin, Tehran.
Email: [email protected]
Mohammad Pakeniat
Researcher in SRRC, Sharif University of Technology. MBA from School of Management & Economics,
Sharif University of Technology. Business consultant and researcher in the fields of strategic
management in developing countries and service marketing.Shahid Rezaii Research Center,
Sharif University of Technology, Azadi Avenue, Tehran, Iran
Email: [email protected]
Abstract
Purpose of this paper is identifying factors which hinder demand for Safe Deposit Box service and factors that are important for
customers when they want to get Safe Deposit Box Service. Methodology consisted of the use of literature review, interviews and focus
groups to form our questionnaires; then two surveys were conducted to extract our results. Findings show (a) customer perception
of lack of safety and privacy are the main causes of little demand for Safe Deposit Box Service; (b) making the customers feel safe,
by giving the information or insuring the boxes can be very useful in marketing Safe Deposit Box service; (c) Bank CRM, personnel
behavior and skills and proximity to home/offce are also important factors for customers who demand SDBR service. There was no
prior formulation of SDBR service marketing practices openly published, so the use of general literature on bank marketing and focus
groups was performed. Focus groups take the ideas of some groups of people for granted; as few people can take apart in them. This
work provides a base for marketing Safe Deposit Box Rental Service. Although the direct income from this service may not seem to
be high, the effects of this service on customer satisfaction have been evaluated to be dramatic. Former research had focused on retail
bank marketing in general; we failed to fnd a research aimed at marketing Safe Deposit Box Services. Our work has directly focused on
marketing this service. Key words: Safe Deposit Box Rental, Marketing, Customer needs, Safety, Banking Services
Financial institutions have prepared a safe place to keep
the valuables of their customers, called “Safe Deposit Box”.
This service is ought to respond to the need for safety. People
use this service to protect their valuables from theft, fre,
etc. Today almost every important bank in the world and
also some non-bank companies are delivering this service.
This means that a huge sum of money has been invested in
building boxes, delivering SDBR services, managing and
planning to develop them.
Most banks have built limited numbers of safe boxes
in their selected, mainly prestigious, branches. Iranian
banks, among others have pursued the same policy. In Iran
state-owned as well as private banks provide this service in
selected branches. These banks have been able to market
few numbers of their boxes (in some cases only about
10%). Heavy investment and little sale have raised the
question about the reasons of low demand in the market and
the factors which are important for customers. Answering
these questions can help banks market their SDBR service
easier.
In bank marketing literature former researchers have
studied marketing related topics in retail banking in general
(e.g., Rosenberg & Czepiel, 1983; Lewis, 1989; Smith and;
Oliver, 1993; Avkiran, 1994; LeBalnc and Nguyen, 1998;
Ndubisi, 2003; wang et al., 2003; Manrai and Manrai,
2007); however, there has been little research about banking
services separately; and we have found no document to
discuss the marketing side of Safe Deposit Box Rental
(SDBR) service. To help fnancial institutions to understand
the behavior of customers of this service better, we have
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
294 Business Intelligence Journal July
conducted a research to answer the following questions:
Why few people use Safe Deposit Boxes? And what is
important for customers when they think about having a
Safe Deposit Box?
To answer these questions, we reviewed pertinent
literature, interviewed with experts and formed an expert
focus group; then we formed a SDBR customer focus group.
We used the outcomes of interviews and focus groups to
form a questionnaire. We conducted a survey to answer
each of the two questions in our fnal step. This research
has been done in different cities in Iran. Although our
respondents, especially in focus groups and surveys, were
limited to Iranians, we expect our fndings to bring about
ideas for other researchers who may try to fnd answers for
similar questions in other social and economic settings.
Our paper is organized as follows: First, we discuss the
history and importance of this service to understand if this
service is still valuable for banks and fnancial institutions.
Second, we discuss our research method, and next we
propose our fndings and conclusion comes at the end.
Importance of the service
In Today’s banking sector SDBR service is delivered
as a necessary banking service. Former research reveals
that in 1970’s about 97% of American banks provide this
service for their customers (Heggestad and Mingo, 1976).
Although most banks spend a lot to offer this service to
their special customers, we failed to fnd a research that
analyses the importance of this service.
To study the importance of this service we reviewed the
patents registered in this feld, and interviewed with experts
from different banks.
a. Patent Analysis
Patent analysis is a method used for several reasons. One
of the applications of this method is identifying the rate of
technological development of a particular product. High
number of patents that have been registered in a period of
time shows that there has been a high investment done in
building this product, which has yielded in many patents.
Low number of patents shows that there has been little
investment done in that product. Higher investment implies
that investors have assumed future market favorable and
expect market growth. Low investment means that frms
are hopeless to have a lucrative business in that product in
future.
In this section, we have reviewed patents and applications
in United States Patent and Trademark Offce (USPTO)
since 1850 that included Safe Deposit Box in their title.
Figure 1 shows the number of patents registered in USPTO
from 1850 till 2010.
Figure 1. Number of patents registered in USPTO about Safe Deposit Boxes
2011 295
As seen in Figure 1, the patents registered show two
upward cones at around 1920’s, a gradual growth since
1950, and a downward cone in 1990. Sharp growth at
1920’s could be attributed to the growth in the number of
customers. It could be argued that the increasing number of
patents related to Safe Deposit Boxes implies that investors
have estimated this product to be proftable.
b. Interviews
To identify the importance of this service for providers,
we carried out interviews with experts from different banks.
Interviewees were selected from globally recognized
fnancial institutions as well as Iranian banks. These banks
included HSBC, Lloyds and Barclays in UK, Standarder
and Banco Pastor from Spain and Melli (largest bank in
the Muslim world), Mellat, Maskan, Saman and ENBank
in Iran.
Results show that most banks do not identify this service
important. All of the interviewees except one admitted that
this service does not bring about direct benefts to banks.
More than half of them expressed that their banks identifes
this service as a means to increase customer satisfaction
rather than a separate service that is separately proftable. So
these banks offer this service to their premium or superior
customers, even sometimes for free. Others, although said
that they treat this service separately, identifed it not very
important to their banks.
Some of the answers that have leaded us to this
conclusion are:
“This service is costly. I think the costs of providing this
service is more than its income; we need to allocate a large
place for boxes, an operator for them, and run a 24 hour
safeguarding, but the price of this service is very low”.
“We offer this service to our selected customers. We
try to increase their satisfaction and we think this would
be proftable for us. Can you imagine the income that
this service will cause by increasing the satisfaction of a
millionaire, even only a little?”
“Today banks’ competition makes you to increase
customer loyalty. We don’t like to lose our customers,
especially premiums, because of lack of SDBR service.”
This evidence questioned the proftability of delivering
this service, but there were other evidence that ensured us
this service is lucrative, even if treated as a separate service,
not just as a means to increase customer satisfaction. This
evidence was non-bank institutions that deliver SDBR
services. Some of these companies sell this service with
other security services (e.g. UK Security Services
1
) and
some provid solely one service: Safe Deposit Box rental
(e.g. metropolitan Safe Deposit Boxes
2
, safe deposit
centers
3
). This evidence showed that this service makes
enough income itself that one runs a business for.
Literature Review
Intense competition has persuaded frms to put more
effective marketing methods into action. Studies performed
based on this want reveal that the cost of serving a loyal
customer is fve to six times less than the cost of fnding
and serving a new customer (Rosenberg & Czepiel, 1983;
Ndubisi, 2003). Therefore further research was carried out
to identify the reasons of customers’ satisfaction, loyalty and
switching behavior. (e.g., Smith and Lewis, 1989; Oliver,
1993; Avkiran, 1994; LeBalnc and Nguyen, 1998; wang et
al., 2003; Manrai and Manrai, 2007). The need for customer
satisfaction urged frms to increase their service quality.
As frms in service sectors decided to increase customer
loyalty and satisfaction, they had to increase their service
quality. Bolton and Drew (1991), Cronin and Taylor (1992)
and Spreng and Mckoy (1996) have argued that service
quality is the main foundation of customer satisfaction for
service companies. Therefore the literature on customer
satisfaction in service sectors has emphasized service
quality as the basis for customer satisfaction, especially
in banking sector. (Smith and Lewis, 1989; Oliver, 1993;
Avkiran, 1994; Levesque and McDougall, 1996; LeBalnc
and Nguyen, 1998; Newman, 2001; Caruana, 2002;
Kheng et al., 2010). Banks today identify service quality
as the main objective that helps them survive in today’s
competitive market (Wang et al., 2003).
Studying customer satisfaction in banking sector added
other factors to service quality. We propose a brief review
of the literature on customer satisfaction in banking sector
here:
There are related topics in banking sector being studied:
customer satisfaction (e.g., Spreng et al., 1995; Levesque
and Mcdougall, 1996; Fienberg, 1996; Mittal and Lassar,
1998; Mittal et al., 1998; Athanassopoulos, 2000; Yavas et
al., 2004), customer loyalty and switching behavior (e.g.,
Jain et al., 1987; Moutinho and Brownlie, 1989; Manrai
and Manrai, 2007). Colgate and Hedge (2001) have argued
1
www.uksecurityservices.com
2
www.metropolitansafedeposits.co.uk
3
www.Safedepositcenters.co.uk
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
296 Business Intelligence Journal July
that there are three main reasons for customer switching:
service failure, pricing and denied services. Chacravarty
et al. (2004) argue that the higher the relationship of
a customer with a bank (economic or other relations)
the lower the propensity of the customer to switch his
bank. Yavas et al. (2004) admit that service quality is the
foundation of customer satisfaction. Manrai and Manrai
(2007) conclude that dissatisfaction with service quality
is the main reason for switching banks. Financial offering
(interest rate) beside convenience and service are the main
parts of service quality identifed (Gwin and Lindgren,
1986; Manrai and Manrai, 1995, Chacravarty, 2004).
Manrai and Manrai (2007) conclude that there are
four overall dimensions that cause customer satisfaction
with bank services: Financial offerings, environment and
atmosphere, convenience and personnel related issues
(Manrai and Manrai, 2007). Although there has been much
debate about their priority (See for example Thwaites and
Vere, 1995).
SDBR service in Iran
Before discussing our research, we propose some
information about SDBR service in Iran, as most of our
feld data has been gathered in Iran.
First openly known Safe Deposit Box in Iran was
delivered in early 1940’s in Melli Bank. This service was
not openly promoted or used till 1980’s, during Iran-Iraq
war; when most of the rich used this place to keep their
valuables safe. The war increased the demand for SDBR
service, and this encouraged the banks to deliver this
service. Today this service is provided in several banks in
Iran; they provide this service at their selected branches.
Most of them offer similar prices, due to Central Bank
regulation. Customers leave some money as a deposit and
pay annual rent to get this service. In Iranian banks access is
limited to working hours but banks follow different criteria
to evaluate suitable customers to deliver SDBR service to
them. Neither of Iranian banks provides insurance for the
contents of safe boxes.
Research Design
In this research we have tried to answer to two questions:
First, Why few people are interested in SDBR service? And
what is important for customers when they think about
having a Safe Deposit Box, in other words, what are key
success factors in marketing Safe Deposit Box services?
To answer the frst question we needed to understand
how people who don’t use this service think about it and
why they don’t use it. To gather their ideas we needed to
conduct a survey. As there was no studies to show what
may be the factors that affect people’s behavior in this
regard, we extracted the factors from:
• A focus group session with people who don’t use this
service.
• A focus group session with bank experts.
In the frst focus group we tried to involve people from
different demographic/occupational groups. Seven people
were invited and the session lasted about 120 minutes.
In the second focus group we invited experts from
Melli (largest bank in the Muslim world), Mellat, Maskan,
Saman and EN banks. These included private as well as
state-owned banks, some of which had a small branch
network, and some a large one. All of these banks deliver
SDBR service. We invited these people as we thought they
are exposed to retail banking customer behavior, and may
have ideas why few banks’ customers demand this service.
The session lasted about 120 minutes.
Results of focus group discussions were analyzed using
content analysis techniques to identify the items of our
questionnaire. This open-ended structured questionnaire
asked about factors that cause few people to use Safe
Deposit Boxes. Randomly selected people who said they
don’t use this service flled out this questionnaire. We tried
to have different demographic groups in our respondents.
784 people answered this questionnaire in 11 selected cities,
to reach the confdence level of 95% at confdence interval
of 3.5. This questionnaire had options based on Lickert
scale, ranking from “frmly agree” to “frmly disagree”.
To answer the second question we needed to know
the idea of SDBR service customers to identify which
factors are important for them when they want to select a
SDBR service provider. To gather their ideas we needed
to conduct a survey. However, as we failed to fnd any
former research to study this phenomenon, we relied on
researches studying retail banking customer satisfaction
and marketing in general. We used the results of these
studies and checked the validity of these factors with bank
experts. As a result we extracted a list of factors that affect
customer satisfaction and successful marketing in retail
banking which were applicable to SDBR service. However,
to complete the factors we added results from:
2011 297
• Interviews with bank experts
• Bank experts focus group
• Customer focus group
Experts from HSBC, Lloyds and Barclays from UK,
Banco Pastor and Standarder from Spain and Melli, Mellat,
Maskan, Saman and EN banks from Iran were interviewed.
These banks deliver SDBR service. We interviewed these
people as we thought they are exposed to customer needs
and expectations from SDBR service providers, and may
have ideas about the factors which affect customer decision
when they want to select a bank to get this service from.
Each interview lasted about 90 minutes.
In bank expert focus group experts from Melli, Mellat,
Maskan, Saman and EN banks were invited. In this session
experts we asked to discuss about factors that are important
for customers and banks can success in marketing SDBR
service by paying attention to them. The session lasted
about 120 minutes.
Seven customers of SDBR service from different banks
were invited to participate in customer focus group. In
this focus group we tried to involve people from different
demographic/occupational groups. The session lasted about
120 minutes.
Results of interviews and focus groups’ discussions
were analyzed using content analysis techniques to identify
the items of our questionnaire. This open-ended structured
questionnaire was aimed at understanding the key factors
of success in marketing SDBR service. 784 current SDBR
service customers flled this questionnaire in 11 selected
cities to reach the confdence level of 95% at confdence
interval of 3.5. This questionnaire had options based on
Lickert scale, ranking the importance from “very low” to
“very high”.
Findings/Discussion
In this section we propose the fndings of our research.
Findings would be proposed in two sections: why people
may not use this service, and which attributes of this service
are important when they look for one.
Why few people use SDBR service?
In this section we present our fndings to answer the frst
question: The reasons of people’s little demand for SDBR
service. To fnd the reasons, as stated earlier we have used
interviews and focus groups with people who don’t use this
service and bank experts. Here we present some quotes
from our interviews and focus groups that we have found
important to derive our results from:
a. Experts’ Focus Group/interview:
“Few people know about the real price of this service;
they think this is a luxury service and should be costly”.
“Even when we introduce this service to our selected
customers, most of them think that this is a special service
for very rich people and they can’t get it.”
“I heard that a customer had told she desired to use this
service, but she thought this service is only for our special
customers”
“People may accuse banks of revealing customer
information to government agencies. People don’t like this,
especially because of tax. Although this does not happen,
but people are concerned about it”
“Compared to other banking services, banks rarely
introduce this service to their customers. This is because
they don’t identify it lucrative”.
b) Non-customers’ focus group
“I don’t use this service; because I may need my
belongings on holidays, when banks are closed”.
“Safe Deposit Boxes are safe, but not my way to the
bank.”
“One of my concerns about this service is my privacy.
When I want to use my box, other customers may know
what I have there. As far as I know, one should use his
SDB in front of others; there should be a private place for
customers”
“Insurance is utmost important for me. No bank provides
insurance on boxes [in Iran].”
“This service is for the rich. I prefer to keep my valuable
belongings in my house.”
“We use this service to increase the safety in keeping
our things, but banks have not assured us they can really
increase safety.”
“No bank in our neighborhood provides this service; I
will have to travel a long distance to get and use this service”
“Safe Deposit Boxes? I have nothing valuable to get
one”
These were quotes we derived from our interview and
results of our focus groups. We then implemented a content
analysis method to analyze these fndings. The factors that
were identifed could be classifed into: little awareness
about the price, little trust to banks, little awareness, lack of
access on holidays, unsafe way to/from bank, little privacy,
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
298 Business Intelligence Journal July
To analyze the results of table 1 we summed “frmly
disagree” and “disagree” percentages, and “frmly agree”
and “agree” percentages. Analysis showed that lack of safety
(85%) and lack of insurance (60%) are the outstanding
reasons. Unsafe way (54%) is ranked third. 47% think that
distance between their place and SDBR provider is a reason
they don’t use this service and 44% have claimed that they
don’t think they have valuables. Other factors don’t seem
to be so important.
The frst three important factors (bank safety, insurance
and safety of people’s way) have all targeted safety;
in other words people’s image of service safety is the
most important factor that affects their decision towards
demanding this service. Our fnding is in harmony with one
of our customers’ claim:
“The main feature of this product is safety. We decide to
use this because we need more safety”.
Geographic proximity is also important for people; as it
takes their time, and affects the safety of their ways to/from
bank. Assuming safety the value that SDBR service brings
about for customers, one would identify the results of the
survey acceptable.
Less than half the respondents have said that other
factors affect their decision. These factors include: people’s
image of privacy of SDBR service, their trust to banks,
their propensity to keep valuables at home, and their need
to access boxes on holidays, and lack of valuables. This
shows that these factors are not as important as formerly
mentioned factors (safety and proximity of providers to
home/offce).
Less than 20% have said that price is important for
them. This fact, beside the fact that most of the people
overestimate the price of this service shows that price does
not hinder people’s demand for SDBR service.
Which attributes of SDBR service are
important for customers?
In this section we discuss the factors that affect that
are important for customers when they want to select a
SDBR service provider. Identifying these factors would
help service providers to understand customer needs and
expectations to develop their services in harmony with
them.
To fnd these factors, we frst reviewed the literature, in
which we failed to fnd a fne-tune literature about SDBR
service. We therefore decided to review the literature on
retail banking as SDBR service is a retail banking service
Factor
F
i
r
m
l
y
d
i
s
a
g
r
e
e
D
i
s
a
g
r
e
e
N
o
i
d
e
a
A
g
r
e
e
F
i
r
m
l
y
a
g
r
e
e
I don’t have access on holidays 9% 20% 36% 25% 10%
My way to bank is not safe 7% 16% 23% 38% 16%
Others may know what I have 21% 34% 26% 15% 4%
I don’t trust banks 27% 32% 18% 15% 8%
There is no insurance 7% 11% 21% 28% 32%
I prefer to keep my valuables at home 14% 36% 18% 24% 8%
Banks are not safe 3% 3% 9% 46% 39%
I have nothing valuable to use SDBR 8% 29% 19% 27% 17%
I cannot aford it 7% 28% 46% 15% 4%
There is no SDBR in our neighborhood 4% 18% 31% 33% 14%
Beside the factors reported in Table 1 we tested
customers’ awareness about this service. We asked them if
they know how to get this service. 15% of respondents
said that they are not aware about this issue at all; and 57%
admitted they don’t know the conditions of getting this
service “exactly”.
In our focus groups we found out that people think
SDBR service is a luxury service and should be expensive.
However, real prices of this service didn’t seem so high.
So we also asked customers’ image about the price of
this service. Survey results recommend that only 27% of
respondents had answered the question about the price
correctly; 19% estimated the price fve times more, and
11% estimated the price ten times more. This implies that
people may imagine this service an expensive service.
Table 1 . Results of our survey about the reasons of little demand
for SDBR service
not insured boxes, propensity to keep belongings in house,
customer imagination of lack of safety and having little
valuables.
These were our most outstanding fndings among many.
To test our fndings from interviews and focus groups
and generalize them to the Iranian society, we designed a
survey based on our former fndings. We ran our survey in
11 selected cities in different parts of Iran. At confdence
level of 95% and confdence interval of 3.5 we collected
784 questionnaires. People that don’t use this service were
asked to fll out the questionnaire. Questions were ranked
in their answers from one (frmly disagree) to fve (frmly
support). Results are shown at table 2.
Asked “Why don’t you use Safe Deposit Box rental
service?” respondents have answered as follows:
2011 299
(provided in the literature review section 3). From the
literature review we derived that: Financial offerings,
environment and atmosphere, convenience and personnel
related issues are important factors in retail banking.
As these factors had not been developed specifcally for
SDBR service, we checked the validity of these four main
groups with bank experts. Based on experts’ ideas we then
concluded that fnancial offering in SDBR service might
not be the interest paid (similar to deposit services) but it
could be the price, which is composed of a deposit and an
annual rent. Environment and atmosphere in case of SDBR
service could be translated to branch safety. We assumed
convenience to be geographic proximity of SDBR service
provider to customers’ place. On personnel related issues,
we derived personnel skill and behavior (including friendly
behavior, caring customers’ needs, etc).
We then formed focus groups with customers and
experts to understand the expectations and important items
for SDBR service customers. As participants in focus
groups are limited, they should be selected in a way to
represent the ideas of different customer groups (Patton,
2001); therefore participants in our focus group were from
different demographic and occupational groups. Here we
present some quotes we have found important to derive our
results from:
“To select a SDBR service, I looked for the one with
insurance, but as I didn’t fnd one [in Iran] I got an uninsured
one, but if it was insured I would feel my belongings are
much safer.”
“I selected this bank, because I felt it is the safest place.”
“It is important to fnd one close to your house or offce.”
“I want to share my box with my father; we keep our
valuables together. Our bank provides us only with one
card. This means that every time the one who wants to use
the box should get the card from the other one. This is not
easy always and takes a lot of time”.
“The least expensive was my option.”
“When I selected this bank, I thought this is the only
bank that provides this service.”
“Every time I go to the bank, I meet a new guy. Bank’s
policy to rotate jobs causes me to feel unsafe.”
“We need the bank to inform us about things like the
need to extend our contract, etc.”
We analyzed our fndings using content analysis
techniques and added our fndings from the literature to it.
We identifed the following possible factors: cost, proximity
to home/offce, insurance, bank branch safety, possibility
of shared access, bank CRM activities, personnel skill,
personnel behavior, fxed SDB staff.
Factor
V
e
r
y
L
o
w
L
o
w
M
o
d
e
r
a
t
e
H
i
g
h
V
e
r
y
H
i
g
h
Cost 4% 7% 25% 26% 38%
Proximity to home/ofce 4% 5% 20% 32% 39%
Insurance 5% 3% 10% 21% 61%
Branch safety 2% 3% 8% 22% 65%
Possibility of shared access 7% 11% 22% 31% 29%
Bank CRM activities 2% 2% 12% 37% 47%
Personnel skills 5% 8% 15% 33% 39%
Personnel behavior 5% 6% 11% 30% 48%
Fixed SDB staf 7% 11% 25% 32% 25%
Table 2. Customer ideas about important factors in SDBR
service delivery
We designed a questionnaire based on the mentioned
factors. We conducted a survey in 11 cities in Iran. At
confdence level of 95% and confdence interval of 3.5 we
collected 784 questionnaires. Cities were selected from
different parts of the country. Respondents were customers
of SDBR service of different banks. We asked them to
evaluate the importance of the identifed factors. The
spectrum composed of “very low” to “very high” options.
It could be observed that bank branch safety and
insurance are the most important factors for customers.
This implies that safety is very important for customers.
Therefore to succeed banks should assure their customers
that their valuables are kept safe.
Also another important factor for customers is banks’
relation with them. As SDBR customers rarely go to bank,
it is important to keep them in touch with their banks to
succeed in increasing customer satisfaction and marketing
this service.
Table 2 shows that all the factors that have been
evaluated are ranked above average. This shows that all of
the factors we have introduced are important. To identify
which factors are more important than others, we used
Friedman test and ranked them. Results are shown below:
Table 3 . Friedman Test for factors identifed to affect SDBR
service marketing
Test Statistics
N 660
Chi-Square 747.819
Df 12
Asymp. Sig. .000
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
300 Business Intelligence Journal July
The results of Friedman test are reported in Table 4.
Table 4. Results of Friedman Test for important factors in
Marketing of SDBR Service
Rank Factor Mean Rank
1 Branch safety 8.79
2 Insurance 8.40
3 Bank CRM activities 7.67
4 Personnel behavior 7.33
5 Proximity to home/ofce 6.87
6 Personnel skills 6.63
7 Cost 6.59
8 Possibility of shared access 5.75
9 Fixed SDB staf 5.47
As the table shows the most important factors are:
Branch safety, insurance and bank CRM activities. This
ensures us that this service should assure customers that
their belongings are kept safe. Other important factors
include personnel behavior and proximity to home/offce.
Factors that ranked lower are: personnel skills, cost,
possibility of shared access and fxed SDB staff.
Conclusion
In this paper we tried to answer two questions that
can help bankers to market their Safe Deposit Box Rental
Service. The frst question was about the reasons few people
use this service. Second question was to identify the factors
that are important for customers to select a service provider.
Results of our research showed that there are several
factors that hinder people’s demand for SDBR service. One
main reason was that people had doubts about the safety of
this service. That the majority doubted the safety of boxes
and identifed insurance as an important factor show that
safety is an outstanding factor. Their concerns about the safe
way to/from bank also strengthen this assumption. Besides
people’s dissatisfaction with the long distance between
their place and bank implies that selecting branches to
locate Safe Boxes is an important issue that helps the bank
to market this service; this would also relieve customer
concerns about unsafe way, due to geographic proximity.
Some people think that the information about contents of
their boxes may be revealed to others (family, government,
etc.), something they resent. Some also think that other
customers may know what they have when they want to use
their boxes and question the private environment in banks.
Our research also showed that many people identifed
SDBR a luxury service. This could be derived from their
imagination about the price of this service (as about notable
proportion estimated the price fve or ten times more), and
that some think this service is dedicated to the very rich
class solely.
Other factors include propensity to keep valuables at
home, lack of access on holidays and lack of valuables to
use SDBR service.
To know which factors are important for customers in
marketing SDBR service (second question), our research
showed that branch safety is ranked frst. This means that
banks should try to transfer a safe image of their boxes. This
could be done by informing them about the construction
of boxes and their location and insuring contents of boxes.
As SDBR customers rarely go to bank (because they
rarely need what they have put in their safe boxes) bank
performance in customer relationship management (CRM)
is important for customers and they have ranked this issue
third. This could be achieved by informing customers about
new pertinent services, termination of their contracts, etc.
by calling, sending mails and emails, etc.
Other important factors which can help banks attract
customers are personnel related issues. This includes
personnel behavior (friendliness, courtesy, etc) and skill.
Banks can strengthen these factors by educating their
personnel in areas like how to do the checking process
faster and following special customer needs whenever
possible. One other factor that was identifed to affect
customer satisfaction in SDBR service delivery was fxed
SDB staff. However, results of our survey showed that this
factor is not very important.
Customers’ request to have this service close to their
home/offce could be answered by distributing the boxes
in a number of branches instead of locating a high number
of them in one branch. Another solution can be providing a
delivery service of boxes to customers’ place. This service
is being delivered in some banks, especially for small boxes.
Cost does not seem to be very important for SDBR
customers; this implies that they are mostly above average
income people. Also the cost of this service (at least in the
Iranian case) is not very high; the rent of the small size
is about 0.3% of the minimum labor income. Therefore
offering lower prices does not seem to be very effective in
marketing this service.
Some customers may need to use Safe Boxes together.
In these cases banks provide one access card for them; and
the one who needs to use the box will need to get the card
from the other; although both have been introduced to the
2011 301
bank as users and their information is registered. This could
be solved by providing one card for each of them. Although
possibility of shared access (with family, colleagues, and
friends) is of very low importance factors identifed in our
research.
It is worth noting that surveys were done in Iran and
the results of our research are based on Iran’s social,
cultural and economic setting. This research has been done
in a developing country; due to similarities in banking
system, social, cultural and economic situations across the
developing world, the results of our work can be generalized
in developing countries. Also, our results can bring about
ideas for further research in marketing SDBR service in
other countries.
Acknowledgement
We are grateful to Mr. Hashemi, Dr. Baseri, Mrs.
Mohseni, and Ms. Rezaeian from Maskan Bank Research
Center for their support of this research. We would also
like to thank Dr. Samadi for his comments on our work.
Mr. Khaksar, Mr. Mohammadzaman and Mr. Mesgari are
acknowledged for their cooperation in analyzing the results
of interviews, focus groups and surveys.
References
Athanassopoulos A. D., 2000, Consumer satisfaction
cue to support market segmentation and explain
switching behavior, Journal of Business Research,
Vol. 47(3), pp.191.
Avkiran N. K., 1994, Developing an instrument to
measure customer service in branch banking,
International Journal of Bank Marketing, Vol.
12(6), pp.10-18.
Bolton, R. N. and J. H. Drew, 1991, A Multistage
Model of Customers’ Assessments of Service
Quality and Value, Journal of Consumer Research,
17 (March), 375-384.
Caruana, A. (2002). Service loyalty: The effects of
service quality and the mediating role of customer
satisfaction, European Journal of Marketing, Vol.
36(7/8), pp. 811-828.
Chacravarty S., R. Feinberg and E. Rhee, 2004,
Relationships and individuals’ bank switching
behavior, Journal of Economic Psychology, Vol.
25(4), pp.507.
Colgate M., and R. Hedge, 2001, An investigation into
the switching process in retail banking services,
International Journal of Bank Marketing, Vol. 19
(4/5), pp.201-212.
Cronin, J. J. and Taylor, S. A. (1992), measuring
service quality: a reexamination and extension,
Journal of Marketing, Vol. 56, July, pp.55-68.
Fienberg R., 1996, Customer dissatisfaction and
delays: the robust negative effects of service delays,
Journal of consumer satisfaction, dissatisfaction
and complaining behavior, Vol. 9, pp.81-85.
Gwin J. M. and Lindgren Jr. J.H., 1986, reaching the
service-sensitive retail consumer, Journal of retail
banking, Vol. 8(3), pp.36-42.
Heggestad A. A. and J. T. Mingo, 1976, Prices, non-
prices and concentration in commercial banking,
Journal of Money, Credit and Banking, Vol. 8 (1),
pp. 107-117.
Jain A.R., Pinson C., Malhotra N. R. , 1987, customer
loyalty as a construct in the marketing of banking
services, International Journal of Bank Marketing,
Vol. 6, pp. 49-72.
Kheng L. L., O. Mahamad, T. Ramayah, 2010, The
Impact of Service Quality on Customer Loyalty: A
Study of Banks in Penang, Malaysia, International
Journal of Marketing Studies, Vol. 2(2), pp. 57-66.
LeBalnc G., and Nguyen N., 1998, Customers’
perception of service quality in fnancial institutions,
International Journal of Bank Marketing, Vol. 6(4),
pp.7-18.
Levesque T. and Mcdougall G.H.G, 1996, determinants
of customer satisfaction in retail banking,
International Journal of Bank Marketing, No. 7,
pp.12-20.
Manrai L.A. and Manrai, A.K.,1995, why do
consumers switch bank services: some managerial
implications (Abstract). In: H. Timmermans (Ed)
European Institute of Retailing and the Canadian
Institute of Retailing and Services Studies.
Manrai L.A. and A. K. Manrai, 2007, A feld study of
customers’ switching behavior for bank services,
Journal of Retailing and Consumer Services,
Vol.14, pp.208-215.
Mittal B., Ross W., Baldasare P.M., 1998, The
asymmetric impact of negative and positive
attribute-level performance on overall satisfaction
and repurchase intentions, Journal of Marketing,
Vol. 62 (1), pp.33-47.
Mittal B., Lassar W. M., 1998, why customers switch?
The dynamics of satisfaction versus loyalty, Journal
of Services Marketing, Vol. 12(3), pp.174-194.
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
302 Business Intelligence Journal July
Moutinho, L., Brownlie D., 1989, Customer
satisfaction with bank services-a multidimensional
space analysis, International Journal of Bank
Marketing, Vol. 7 (5), pp.23-27.
Newman, K. (2001). Interrogating SERVQUAL: a
critical assessment of service quality measurement
in a high street retail bank, International Journal of
Bank Marketing, Vol. 19(3), pp. 126-139.
Ndubisi N. O., 2003, Service quality: understanding
customer perception and reaction, and its impact
on business, International Journal of Business, Vol.
5(2), pp.207-219.
Oliver, R.L., 1993, A conceptual model of service
quality and service satisfaction: compatible goals,
different concepts, in Swartz T.A., Bowen D.A. and
Brown S.W.(Eds), Advances in Service Marketing
and Management, Vol.2, JAI Press, New York, NY,
pp.65-85.
Rosenberg, L. J., & J. Z. Czepiel, 1983, A marketing
approach for customer retention, The Journal of
Consumer Marketing, Vol. 1, pp.45-51.
Smith A. M. , Lewis B. R., 1989, Customer care in
fnancial service organizations, International
Journal of Bank Marketing, Vol. 7 No.5, pp.13-
22 MacKenzie S.B. and Olshavsky, R.W., 1995,
A reexamination of the determinants of consumer
satisfaction, Journal of Marketing, Vol.60 (3), p.15
M. Q. Patton (2001) Qualitative Research and
Evalutaion Methods, SAGE Publications.
Spreng R. and R. Mckoy, 1996, An empirical
examination of a model of perceived service quality
and satisfaction, Journal of Retailing, vol.72, pp.
201-214.
Thwaites D., Vere L., 1995, Bank selection criteria-a
student perspective, Journal of Bank Marketing,
Vol. 11 (1-3), pp. 133-149.
Wang Y., H. P. Li and Y. V. Hui, 2003, The antecedents
of service quality and product quality and their
infuences on bank reputation: evidence from
banking industry in China, Managing Service
Quality, Vol. 13(1), pp.72-83.
Yavas U., Benkenstein M., Stuhldreier., 2004,
Relationships between service quality and
behavioral outcomes: a study of private bank
customers in Germany, International Journal of
Bank Marketing, Vol. 22 (2/3), pp.144.
2011 303
STRATEGIC MARKETING AND FIRMS PERFORMANCE: A
STUDY OF NIGERIAN OIL AND GAS INDUSTRY
Akinyele, S.T. (PhD)
School of Business, Covenant University,Ota-Nigeria.
Email: [email protected]
Abstract
The purpose of this paper is to investigate the impact of strategic marketing and frms performance of the Nigerian oil and
gas industry. This study adopted a survey research methodology to examine strategic marketing and frms performance of Nigerian
oil and gas marketing companies in an attempt to attain their desired level of performance. One hypothesis was formulated from the
statement of research problem. Analysis of Variance, Pearson Moment Correlation Analysis, Factor Analysis among other statistical
tools were used in testing the hypotheses. The overall results suggest that strategic marketing is a driver of organizational positioning
in a dynamic environment, and that it helps to enhance the development of new product/service for existing markets. These fndings,
along with other interesting fndings of the study, are discussed. From the empirical and anecdotal managerial evidence as well
as from the literature implications are drawn for the effcient and effective strategic marketing in the Nigerian oil and gas industry.
Key words: Strategic Marketing, Firms, Deployment, Resources, Performance
The sensitivity of petroleum resource is clearly refected
in the fact that it has remained or continued to be the goose
that lay a golden eggs for the Nigerian economy as well
as the supreme foreign exchange earner contributing over
80% of government revenues and helps the development
of Nigeria’s infrastructures and other industries (Chukwu
2002). However, due largely to the highly technical nature of
exploration and production, the sector depends substantially
on imported technologies and facilities for most of its
operations. In view of the critical importance of the sector
to the nation’s economy and its capacity to generate far-
reaching multiplier effect, the grooming of highly skilled
indigenous manpower to participate keenly in the activities
of the sector to redress the foreign dominance becomes
imperative (Baker 2006). With a current production capacity
of about 30 million barrels per day (bpd), Nigeria plans to
increase her production capacity to about 40 million bpd
by 2010 (Mathiason 2006). Already, Nigeria is the leading
oil and gas producer in Africa, currently ranked the seventh
highest in the world (The Guardian 2006).
In addition to the above, Nigeria which is widely
referred to as a gas province, has natural gas reserves that
triple crude oil reserves, being estimated in excess of 187.5
trillion standard cubit feet(scf) (Africa Oil and Gas 2004).
Baker (2006) observes that Nigeria began exporting oil
in 1958 with crude oil production of 5000 barrels per day
(bpd) rising by 1979 to a peak of 2.3 million bpd.. However,
gas resources are largely untapped and Nigeria’s gas
reserves place it among the top ten countries in the world
in that category ( Ekpu 2004). Ekpu (2004) also observe
that other energy resources such as hydro power, wind
energy, and coal, which is produced in Enugu and Benue
States abound in the country. Energy consumption is in the
area of petroleum products, which according to (Dixton et
al 2005), accounted for between 70% and 80% of total
energy consumed in Nigeria between 1970 and 1980, the
major consumers being the transportation, household and
industrial sectors.
Marketing has been defned and conceptualized in
various ways, depending on the author’s background,
interest, and education (Osuagwu 1999). For example,
marketing can be seen as a matrix of business activities
organized to plan, produce, price, promote, distribute, and
megamarket goods, service, and ideas for the satisfaction
of relevant customers and clients. Achumba and Osuagwu
(1994) also posit that marketing is important for the success
of any organization, whether service- or product-oriented.
Mathiason (2006) argues that the oil and gas service
sector constitutes a service industry that has currently
been changed by aggressive strategic marketing behaviour.
According to (Li et al 2000), indigenous Nigerian oil
and gas marketing companies were not profoundly
entrepreneurial at the beginning for the following reasons:
lack of trained manpower, poor infrastructural development,
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
304 Business Intelligence Journal July
lack of adequate or suffcient capital base on the part of the
indigenous oil and gas marketing companies and intense
competition from superior foreign companies.
This study is intended to expand the body of knowledge
in respect of the application of strategic marketing to the
oil and gas sector especially in a developing economy like
Nigeria that earns over 80% of her foreign exchange from
oil and particularly, as the Federal Government is putting
in place policies and strategies to improve the oil sector’s
contributions to the Nigeria economy (Garuba 2006).
Statement of Research Problem
The problem statement, describes the content for the
study and it also identifes the general analysis of issues
in the research necessitating the need for the study
(Osuagwu 1999). The research is expected to answer
questions and provide reasons responsible for undertaking
the particular research (Pajares 2007). Many research
efforts in the area of marketing practices in developing
economies have dealt with macro issues and emphasized
the management of company’s structure and strategies,
conduct and performance of marketing activities as they
relate to performance indices such as market share, growth,
effciency and well being of consumers and clients (Samli
and Kaynak 1994) lament that the key defect with this static
and macroanalysis of marketing practices in developing
economies is that it minimizes the impact of marketing
environment on the achievement of performance measures.
Also, although some research efforts have been undertaken
to explain marketing practices in developing economies at
the organizational level (Chukwu 2002), many of these
research efforts do not provide answers to issues pertaining
to the impact of company’s structure and strategies on the
performance of mineral prospecting industries particularly
the oil and gas marketing companies. The deregulation of
the Nigerian economy through the Structural Adjustment
Programme (SAP) affected the oil and gas sector in Nigeria
in many ways . Consequently, the Nigerian oil and gas
companies incorporated the usage of various market mix
elements to improve their market outreach/ coverage, new
product ratio, price positioning, competitive orientation,
etc to survive and grow (Johne and Davies 2002). The
poor condition of some oil and gas marketing companies
in Nigeria is a function of some interrelated problems.
According to (Li et al 2000 ), the causes of the oil and gas
marketing companies failure or poor performance, are due
to microeconomic or macroeconomic factors (performance
industry environmental factors indice coupled with the
management of marketing content and product marketing).
Garuba (2006) have, however, posited that oil and gas
companies poor performance in Nigeria is a function of
industry environmental factor indices and marketing of oil
and gas services. It evolves as a result of the interplay of
the marketing mix elements and the environmental factors,
which impact on these elements (Li et al 2000 ; Mavondo
2000). Therefore, the function of marketing strategy
deals with determining the nature, strength, direction,
and interaction between marketing mix elements and the
environmental factors in a particular situation (Osuagwu
2001; 2001; 2004). However, achieving effcient and
effective product marketing strategy by an organization
is diffcult, as a result of the ambiguity and instability
of environmental factors . Sound and robust marketing
commitment on the part of organization and sales-people
are important to the survival and growth of the oil and gas
industry, considering the subtle, unstable and seemingly
hostile business environments in which contemporary
business organizations operate (Osuagwu 1999). In order
to formulate and implement effective and effcient goal
actualization and inter-industry marketing commitment in
product distribution, oil and gas companies should have
a thorough and continuous understanding of the relevant
environment that impacts on their marketing strategies.
Objectives of the Study
The main focus of marketing activities of oil and gas
marketing companies is the identifcation and satisfaction
of clients’ needs and wants. These objectives can be
attained by identifying the likely marketing mix variables
and strategies, including relevant environmental impacts
on them. There is, therefore, the need to carry out this
research given the enormity of the problem facing the oil
and gas industry. Specifcally the study sought to explore
the possible differences in the organizational structure and
strategies adopted and how they affect the performance of
Nigerian oil and gas marketing companies.
Research Hypothesis
The organizational structure and strategies adopted
do not affect the market share of Nigerian oil and gas
marketing companies.
2011 305
Literature Review
A service may therefore be seen as an activity or beneft
which can be offered to an organization or individual by
another organization or individual and which is essentially
intangible. It is a separately identifable but intangible
offer which produces want-satisfaction to the client, and
which may or may not be necessarily tied to the sale of
a physical product or another service (Osuagwu 1999).
Services include a wide range of activities and form some
of the growing sectors of the economies of developed and
developing countries. Services include professional services
(legal, accounting, medical, management consulting,
etc), general services (insurance, postal, telephone,
transportation, internet ,tourism, etc), maintenance and
repair services, and services from marketing researchers
and product manufacturers, among others. Oil and gas
service is not a tangible thing like food, clothing and car.
Sound and robust marketing strategies are important to
the survival and growth of any business, including oil and
gas business, considering the increasingly subtle, unstable
and seemingly hostile business environments in which
contemporary business organizations operate . Therefore,
in order to formulate and implement effcient and effective
marketing strategies, business organizations should have
a thorough and continuous understanding of the relevant
environment that impacts on their marketing strategies.
According to (Mavondo 2000), marketing strategy has
been a salient focus of academic inquiry since the 1980s.
There are numerous defnitions of marketing strategy
in the literature and such defnitions refect different
perspectives ( Li et al 2000). However, the consensus is
that marketing strategy provides the avenue for utilizing the
resources of an organization in order to achieve its set goals
and objectives. Generally, marketing strategy deals with
the adapting of markeing mix-elements to environmental
forces. It evolves from the interplay of the marketing mix
elements and the environmental factors (Li et al 2000).
Therefore, the function of marketing strategy is to determine
the nature, strength, direction, and interaction between the
marketing mix- elements and the environmental factors in
a particular situation . According to (Levie 2006), the aim
of the development of an organization’s marketing strategy
development is to establish, build, defend and maintain
its competitive advantage. Against this background, the
present research attempts to assess the marketing strategies
of Nigerian oil and gas marketing companies, the impacts
of environmental factors on such strategies and the
effectiveness of the marketing strategies. The marketing
strategies of Nigerian oil and gas marketing companies are
expected to be adaptable to these environmental factors
in order to achieve set performance goals. The oil and gas
industry seems to have witnessed some form of corporate
performance over the years which can be attributed to their
distinct level of market share (Akinyele, 2010 ).
Marketing strategies and tactics are concerned
with taking decisions on a number of variables to
infuence mutually-satisfying exchange transactions and
relationships. Typically, marketers have a number of
tools they can use, these include megamarketing and
the so-called 4Ps of marketing , among others. Marketing
seems easy to describe, but extremely diffcult to practice
(Kotler and Connor 1997). Marketing is one of the salient
and important organic functions which help to service
organizations to meet their business challenges and
achieve set goals and objectives . The word “ service” is
used to describe an organization or industry that “does
something” for someone, and does not “ make something”
for someone (Silvestro and Johnston 1990). “Service” is
used of companies or frms that meet the needs and wants
of society such organizations are essentially bureaucratic .
“ Service” may also be described as intangible its outcome
being perceived as an activity rather than as a tangible
offering. The question of the distinction between services
and tangible products is based on the proportion of service
components that a particular offering contains.
Definition of Strategic Marketing
The early strategic marketing - performance studies
date from the time of rapid expansion of formal strategic
marketing in the 1960s (Levie 2006). Although same studies
employed diverse methodologies and measures, they shared
a common interest in exploring the fnancial performance
consequences of the basic tools, techniques, and activities
of formal strategic marketing i.e. systematic intelligence-
gathering, market research, SWOT analysis, portfolio
analysis, mathematical and computer model of formal
planning meetings and written long- range plans. Osuagwu
(2004) observes that strategic marketing is a disciplined
effort to produce fundamental decisions and actions that
shape and guide what an organization is, what it does, and
why it does it, with a focus on the future. Akinyele (2010),
reports that newly invented strategic marketing displaced
long range planning because of the growing discontinuity of
the environment. Strategic marketing on the other hand does
not necessarily expect an improved future or extrapolatable
past. Grewal and Tansuhaj (2001), argued that a manager
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
306 Business Intelligence Journal July
Table 2 presents the descriptive statistics of the
effectiveness of strategic marketing practices of the studied
oil and gas industry. The fnding shows that strategic
marketing practices have been reasonably effective in oil and
gas industry, with strategic marketing effectiveness being
very high from the analysis above. The essence of strategic
marketing is to achieve set objectives, and these objectives
can be measured in terms of proft, market share, marketing
cost, gross earnings, capital employed, asset quality, quality
of marketing management, liquidity, turnover of marketing
staff, and management of departmental crisis. The
Table 2: Descriptive Statistics of Effectiveness of Strategic
Marketing Using Qualitative Measures of Performance(n=
286).
Variables Mean
Std.
Dev.
Variance Skewness
Company makes proft by
selling large quantities of
goods/services
5.17 .96 .89 -.1.27
Experience to cut costs is an
important goal
4.67 .99 .99 -.79
Sales executive move faster
than competitors in responding
to customers needs
4.93 .98 .87 -.99
Develops an exhaustive set
of alternatives before making
improvement management
decision
4.79 .89 .79 -.69
Demands better services
provided by customers
4.17 1.23 1.29 -.39
Emphasize opening up new
branches
4.69 1.23 1.27 -.32
Ability to gain market share is
high
3.57 1.39 1.79 -.34
is not necessarily a strategist and that a manager’s vision is
also not an entrepreneurial vision. He explains that while
the manager would rather have an orientation point of
guiding a company in a specifc direction, an entrepreneur
having strategic competence should state his vision clearly,
aggressively and in an optimistic manner.
Methodology
A cross- sectional survey was selected for this study
because it was easy to undertake compared to longitudinal
survey and the result from the sample can be inferred
to the larger population. In addition, some extraneous
factors could have manifested in the observed change
other than the independent variable concerned. The target
population in the study was the employees of product
pipeline marketing companies in Lagos, Nigeria. The
questionnaire was pre-tested with respondents in other
product marketing company, to authenticate reliability.
The pre-testing was done to avoid any possible infuence
on trial respondents before the actual survey. The analyzed
data was presented using descriptive statistics, frequency
tables, Analysis of Variance, and Correlation coeffcients.
Descriptive statistics allow the generalization of the data to
give an account of the structure or the characteristics of the
population as represented by the sample.
Data Analysis, Finding and Discussions
Table 1: Reliability Coeffcients of Research Measures
(Cronbach’s Alpha)
S/N Variable Measure
Cronbach’s Alpha
Coefcients
1 Management of Marketing strategy 0.76
2 Oil and Gas Performance Measurement 0.73
3 Efect of Environmental factors on
marketing strategies
0.65
4 Organizational structure and strategi 0.84
Table 1 above shows Cronbach’s alpha coeffcients of
the major research measures. “Management of marketing
strategy contructs” and “Oil and gas performance
measurements” met Nunnally’s (1978) internal consistency
(reliability) standard for newly- developed research
measures, while “Effect of environmental factors on
marketing strategies” failed to meet Nunnally’s (1978)
standard for reliability. Specifcally, Nunnally (1978)
recommended 0.70 Cronbach alpha value (internal
consistency) for newly developed research instruments.
Therefore, subject to the specifc and usual limitations
associated with this type of research, the research instrument
appears reliable and valid.
This study has provided empirical evidence pertaining
to the perception placed on oil and gas marketing strategies,
and oil and gas performance measures and impact of
environmental factors on such strategies. The research
fndings show that product and mega marketing strategies
received relatively low perception. These fndings have
managerial and research implications.
2011 307
effectiveness of strategic marketing practices in the studied
oil and gas industry is encouraging. These are the CAMEL
measures of performance. According to (Osuagwu 1999),
the effectiveness of oil and gas strategies determines the
survival and growth of downstream sector in Nigeria,
especially in an ever- changing environment. Effective oil
and gas management through strategic marketing assists
in the employment of capital raised, and manages the oil
and gas asset portfolio in viable business options so that
the assets are seen to be performing and yielding returns.
The marketing strategies of oil and gas, in order to show
reasonable levels of effectiveness along the CAMEL
measures, have to emphasize a marketing management
team with foresight, experience, and commitment towards
the survival and growth of the oil industry, among others.
Osuagwu(2001), posit that the most widely accepted
measure of performance of oil and gas industry is current
proftability, which is measured in terms of return on assets
and return on equity. Nigerian oil and gas industry that
creates comparatively large amounts of value (in relation
to its equity) through it strategic marketing practices can
be said to show high level of effectiveness. And as Table
3 shows, the studied oil and gas industry’s have shown
appreciable levels of effectiveness using the identifed
measures of performance.
Table 3: Comparison of Environmental Characteristic
a
Dimension of the Environment
C
o
m
p
a
n
y
A
T
o
t
a
l
M
e
a
n
C
o
m
p
a
n
y
B
O
a
n
d
o
M
e
a
n
C
o
m
p
a
n
y
C
T
e
x
a
c
o
m
e
a
n
C
o
m
p
a
n
y
D
A
g
i
p
M
e
a
n
Markets
Product diversity 4.64 4.29*** 3.22 3.12
Geographical diversity 5.40 4.22*** 2.34 3.21
Level of product information 4.89 4.80 4.33 4.02
Diversity of promotional media 4.92 4.46* 4.06 3.42
Competition
Intensity of rivalry 5.69 5.36** 4.34 4.54
Inability to infuence market
conditions
4.21 3.60*** 3.34 3.18
Average proftability of the principal
market
4.32 4.26 4.21 4.65
Entry barriers to the principal market 4.69 5.42*** 4.32 4.24
Constraints imposed by inter-industry
relationships with major stockholders
2.84 3.09 3.33 3.24
With major distributors and customers 2.76 3.78*** 3.43 3.11
With major suppliers-subcontractors 2.49 3.71*** 3.34 3.54
With government 4.38 3.27*** 3.12 3.43
With competitors 2.14 2.64*** 2.42 2.56
Notes: a. The higher the mean score, the more typical is the characteristics
* Signifcant at 0.5 level by t-test of means
** Signifcant at 0.1 level by t-test of means
*** Signifcant at.001 level by t-test of means
Dimension of the Environment
C
o
m
p
a
n
y
A
T
o
t
a
l
M
e
a
n
C
o
m
p
a
n
y
B
O
a
n
d
o
M
e
a
n
C
o
m
p
a
n
y
C
T
e
x
a
c
o
m
e
a
n
C
o
m
p
a
n
y
D
A
g
i
p
M
e
a
n
Ability of labour market
For managers 3.64 1.79*** 3.11 3.23
For technological experts 3.47 1.99*** 2.23 2.65
From the above table, there is also a signifcant difference
in labour market-ability between the four companies. Total
frms face a less mobile labour market than oando, Texaco
and Agip frms. Not new, the fndings is consistent with the
prevalent view that the Total labour market is less mobile
because of its many tangible incentives incorporated
into their employment system. The strengths and range
of constraints imposed by interrelationships with other
organizations are also different in Total and other oil and
gas companies under study. Oando and other oil and gas
frms face stronger constraints imposed by government and
regulatory bodies, while Total frms feel the constraints
imposed by their relationships with distributors, customers,
suppliers and competitors to a greater degree than Oando,
Texaco and Agip frms. This result suggests that Total
companies create closer inter-organizational networks with
various kinds of organizations. The networks, although
constraining decisions within organizations, may have a
number of benefts including risk-sharing and long term
stabilization of business. The strong constraints imposed by
the government upon oil and gas companies probably stem
from the relatively adverse historical relationship between
business and government in Nigeria as well as from motives
to protect the public and promote competition.
To sum up, Total frms face a less diverse, less competitive,
more volatile and high opportunity environment, and less
mobility of market. They are, moreover, constrained by
interrelationships with other organizations to a greater
extent than the other oil and gas marketing frms under
study. A frm must set operational objectives, the priorities
of which are contingent upon the opportunities provided
and constraints imposed by its environment matched
against the internal capabilities of the organization.
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
308 Business Intelligence Journal July
a
R Squared= .382 (Adjusted R Squared=.361)*
signifcant at 0.005 two tailed test.
The fndings in Table 5 indicate that there was a
signifcant difference in performance at f (4,285)= 5.457, at
0.05 signifcant level. This implied that the frst hypothesis
was rejected and the alternative hypothesis retained, which
meant that the company’s structure and the strategies
adopted by Total yielded a better market share than that
of their counterparts in the industry.
Discussion of the Findings
As stated earlier, the discussion of this study followed
the hypotheses raised and tested and they are presented
below:
The organizational structure and strategies adopted
do not affect the market share of Nigerian oil and gas
marketing companies.
The null hypothesis one which stated that” the
organizational structure and strategies adopted do
not affect the market share of Nigerian oil and gas
marketing companies” was rejected. This implied that
the organizational structure and strategies adopted by
oil and gas marketing companies affect market share
positively. Several empirical studies have concluded that
an organizational structure and strategies adopted is indeed
an important element in companies’ success. (Levie 2006),
gave evidence that the level of organizational structure and
strategies is positively related to company effectiveness,
while (Grewal and Tansuhaj 2001 ) reported that more
successful companies have well defned organizational
structures, in sharp contrast to less successful companies.
Focusing on large frms, (Ekpu 2004); found a positive
relationship between the unstructured organizational
patterns and large frm fnancial performance. The fndings
by (Osuagwu, 2004; Akinyele 2010) underscore this
fnding, as they established that changes in the market
environment, business strategy and organizational structure
have impact on companies performance.
Table 4: Mean and Standard Deviation on Company’s Structure
and Strategies Adopted and Market share Attained by Nigerian
Oil and Gas Marketing Companies
Questions Freq. Mean
Standard
Deviation
Company’s structure and strategies adopted by
Total frm
286 70.40 2.18
Company’s structure and strategies adopted by
Oando frm
286 68.40 1.18
Company’s structure and strategies adopted by
Texaco frm
286 42.64 1.22
Company’s structure and strategies adopted by
Agip frm
286 68.20 0.22
Oil and gas companies performance 286 64.40 1.16
Table 4 above shows the difference between means of
company’s structure and strategies and resultant market
share of Nigerian oil and gas marketing companies. Total
frm with a mean of 70.40 and standard deviation of 2.18
had a better structure compared with other counterparts in
the industry with mean of 68.40 and standard deviation of
1.18 respectively. The Table also shows that with the mean
of 64.40 and standard deviation of 1.16 when divided
between the companies, the market share of Total was better
than that of the other counterparts. The question then is,
was the difference signifcant enough or was it the result
of a sampling error? The answer is presented in Table 5.
Table 5 Summary of Analysis of Variance on Company’s
Structure and Strategies Adopted and Market share Attained by
Nigerian Oil and Gas Marketing Companies Respectively
Source
TypeIII
Sum of
squares
Df
Mean
square
F Sig.
Corrected Model 294.808 9 32.644 35.415 .000
Intercept 1887.508 1 1887.508 1749.596 .000
Company Types 20.014 1 5.654 6.240* .000
Company’s structure
&strategies
22.757 4 5.684 6.231 .000
Company Types*
Company’s structure &
strategies
24.745 4 5.086 5.457 .001
Source: Field Survey, 2007.
Hypothesis Testing
The frst hypothesis tested in the study states that ‘
The organizational structure and strategies adopted do not
affect the market share of Nigerian oil and gas marketing
companies.
Source
TypeIII
Sum of
squares
Df
Mean
square
F Sig.
Error 581.174 283 .816
Total 5758.000 285
Corrected Total 667.984 286
2011 309
Conclusions
This section elaborates on the conclusion of the research.
Based on the fndings of this research, the following
conclusions are warranted:
1. The evidence from fndings suggested that oil and gas
marketing companies have comparative advantages in
adopting various marketing strategies using different
technologies. Oil and gas marketing companies
appeared to specialize in the use of traditional methods
of marketing, which is based on “soft” information
culled from close contacts by marketing and sales
department rather than the use of the specialized
strategic marketing methods that are based on “hard”
quantitative information.
2. Most of the fndings of the research are consistent with
previous normative and empirical works. For instance,
the companies face a less diverse, less competitive,
more volatile and high opportunity environment, and a
less mobility of market. They are however, constrained
by interrelationships with other organizations to a
greater extent.
3. The research instrument shows high validity and
reliability.
4. This study has provided empirical evidence pertaining
to the perception of oil and gas marketing strategies, and
the industry environmental factors on such strategies..
Managerial and Research Implications
The fndings of this study have several managerial
implications for Nigerian downstream oil and gas. First,
Nigerian oil and gas managers are advised to place more
emphasis on the adoption of various marketing strategies
using different technologies. Second, all organizations
face an external business environment that constantly
changes. Changes in the business environment create both
opportunities and threats to an organization’s strategic
development, and the organization cannot risk remaining
static. It must monitor its environment continually in
order to: build the business, develop strategic capabilities
that move the organization forward, improve the ways in
which it creates products/services and develops new and
existing markets with a view to offering its customers
better services. Third, anticipating competitors’ actions
and reactions to the organizational’ moves may be the key
determinant of success for any marketing strategy. Fourth,
with the competitive downstream oil and gas industry of
today, participants can put more emphasis on relationship
marketing to ensure effectiveness. This essentially entails
personalizing the oil and gas services offered to clients,
attending to clients’ cultural and social activities, in relation
to other non- business activities, which are of interest to
clients. Fifth, Nigerian oil and gas marketers should be
sensitized to the importance of their offerings to their
clients, including the impressions their clients have of
those offerings. As oil and gas clients are demanding more
quality from their petroleum product marketing companies
(PPMC), it may be strategic to inject the idea of total
quality management (TQM) and its variants among product
marketers.
Recommendations, Limitations and
Suggestions for Further Studies
This study indicates that strategic marketing have
a signifcant impact on performance variables and that
they interact with the different components to facilitate
performance. It also indicates that different performance
factors moderate strategic marketing practice. Therefore,
organizations hoping to enhance corporate performance
in a dynamic business environment should consider the
following:
Recommendations
a. The concepts and principles of total quality
management (TQM) are recommended for holistic study,
in addition to contemporary marketing management issues
such as relationship marketing, value analysis, business
process re-engineering, megamarketing, re-marketing, co-
marketing, bench marketing, and permission marketing,
among others.
b. Efforts should be made by organizational marketers
to understand the relevant factors that affect both clients’
behaviours, and the strategic options to be adopted to cope
with such behaviours.
c. Oil and gas marketing scholars or researchers
should endeavour to study holistically the relevant business
functions and activities which may enhance or hinder the
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
310 Business Intelligence Journal July
Journal of Management and Business Research,
vol.10(6):60-77.
African oil and gas(2004). Energy policy agenda
.International energy services Ltd, Lagos.
Baker, M.(2006). Petroleum technology development
fund: Intervention to discover and develop
technocrats for the energy sector of tomorrow.
Petroleum Journal, Vol. 5(3),112-115.
Chukwu, I.(2002). Crude oil development, The Post
Express, Lagos, Thursday, October 1, p. 24-25.
Dixton, D. , Donald, F. , & Roger, A. (2005). Basic
Marketing: Concepts, Decisions and Strategies,
Englewood Cliffs, N.J. : Prentice- Hall Inc.
Ekpu, R.(2004). Associated gas utilization, Nigeria’s
oil and gas, 3(8) August,pp.17-18.
Garuba, D.S.(2006). Survival at the margins:
Economic crisis and coping mechanisms in rural
Nigeria, Local Environment, Vol. 11, (1), p.17-36.
Grewal, R. and Tansuhaj, P.(2001). Building
organizational capabilities for managing economic
crisis: The role of market orientation and strategic
management. Journal of Marketing, 65: 67-80.
Johne, A. and Davies, R.(2002). Innovation in medium-
sized oil companies: How marketing adds value,
The International Journal of Bank Marketing, 18(1).
Lavie, D.(2006). Capability reconfguration: An
analysis of incumbent responses to technology
change. Academy of Management Executive,
31(1):153-174.
Li, S. , Kinman, R., Duan, Y. and Edwards, J.S.(2000).
Computer-based support for marketing strategy
development, European Journal of Marketing,
34(5/6).
Mathiason, N.(2006). Oil delta burns with hate, The
observer(UK), Sunday, January, 29(http://observer.
Guardian.co.uk/business.
Mavondo, F.T.(2000).Marketing as a form of
adaptation: Empirical evidence from a developing
economy, Marketing Intelligence and Planning,
18(5),256-272.
Nunnally, J.(1978).Psychometric Theory, 6th Edition,
New York:McGraw –Hill.
0suagwu, L.(1999). Marketing: Principles and
Management, Lagos: Grey Resources Limited.
understanding and subsequently applicability of relevant
modern management concepts and principles to oil services
marketing.
d. Firms that are not operating in a dynamic business
environment need not adopt a strategic marketing practice
as this may cause the frm to look inconsistent in the eyes of
its customers and eventually reduce effective performance.
e. The need for the identifcation of options and
resources and of capabilities of deployment constitutes an
impetus to effective strategic marketing implementation,
since the practice derives from capabilities in assembling
and maintaining an appropriate resource portfolio and
coupling the resource portfolio with the identifcation and
recognition of options.
Suggestions for further studies
This research leads to some observations that might be
of interest to future researchers, as they represent the seeds
from which future research can be developed.
a. This same research can be carried out in other
nations so that a broad comparison of the concepts of
strategic marketing as it affects frm performance can be
made.
b. Research into the combined effects of strategic
marketing practice and performance factors as mediators
of frm performance relationship.
c. Research into the effects of key characteristics of
industries environmental indices and marketing strategy
could be carried out to further explain the differences in the
frm’s adoption of strategic marketing.
d. Finally, future research works are to be undertaken
in order to refne the cobwebs found in the present research,
and orient it to more specifc contexts (business, time,
location, etc) in Nigeria’s oil and gas industry.
References
Achumba, I.C. and Osuagwu, L.(1994). Marketing
Fundamentals and Practice, Rock Hill: USA: Al-
Marks Educational Research, Inc.1-400.
Akinyele, S.T. (2010). Signifcance of strategic
marketing to enterprise performance: An empirical
analysis of Nigerian oil and gas industry. Global
2011 311
Osuagwu, L.(2001). An evaluation of the marketing
strategies of Nigerian insurance companies,
Academy of Marketing Studies Journal, Vol.
5(2),pp.17-30.
Osuagwu, L.(2001). Adoption of marketing strategies
in Nigerian institutions, Academy of Marketing
Studies Journal, Vol. 5(1): 1-16.
Osuagwu,( 2004). Relationship marketing strategies in
Nigerian companies, The Marketing Management
Journal, Vol. 14(2): 114-128.
Pajares, F.(2007). Elements of a proposal, available
from the author at www.aes.emory.edu/mfp/
proposal.html, download on 10th May, 2008.
Samli, A.C and Kaynak, E.(1994). Marketing practices
in less developed countries, Journal of Business
Research, 12:5-18..
Silvestro, R. and Johnson, R.(1990). The determinants
of service quality-enhancing and hygiene factors,
Proceedings of the QUISS 11 Symposium, New
York: St. John’s University.
The Guardian(2006). What is wrong with the Nigerian
oil industry. Wednesday, January 25,p.1-2, Lagos.
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
312 Business Intelligence Journal July
INTUITIVE MANAGERIAL DECISION MAKING IN MALAYSIA
AND THE UNITED STATES
Dr. Isola Oluwabusuyi (MBA, MA, DBA)
School of Business and Accounting
Brown Mackie College, Atlanta
Email: [email protected]
Abstract
The purpose of this paper is to examine and compare the factors that infuence intuition as a decision-making tool for managers
in Malaysia and in the United States. The underlying study examined the relationships among gender, management level, country of
operation, and the reported use of intuition in decision making. Agor’s Intuitive Measurement Survey (AIM) survey was adapted (with
permission from copyright owner) from Weston Agor’s study to measure the relationship between a manager’s reported use of intuition
in decision making and the manager’s management level, the manager’s gender, and the manager’s country of operation. The research
shows signifcant relationship between research variables. Male managers in Malaysia’s reported use of intuition in decision making
was signifcantly lower than US managers reported use. Limitations of the study include the following: Samples were a combination of
criterion, maximum variation and convenience based selection of companies in Malaysia and the United States. United States samples
were selected from the East and West coast of the United States. The limitations could impact study’s external validity. Study’s fndings
are quite signifcant to global business managers intending to shift more of their activities to Asia in the near future. Multinational
corporations would have to provide more data for their Malaysian managers. This Study had fve signifcant fndings. Business schools
in the West may need to redesign their curriculum as more business managers feel more comfortable with intuitive decision making
techniques. Key words: Decision Making, Intuition, Global Management, Culture
Managerial decision making has always been a subject of
passionate academic discussion. What is the most effective
way of choosing between alternatives? Should managerial
decision making be intuitive or rational? Are managers
of companies facing relatively stable environments more
intuitive in their decision making than their counterparts
managing companies facing more unstable environments?
Are female managers more intuitive in their managerial
decision making than their male counterpart? Is managerial
decision making infuenced by a manager’s country of
operation?
Khatri &Ng (2000) discovered that top executives
of computer companies surveyed reported using more
intuition than their counterparts in banks and utilities.
The study also showed that computer industry was more
unstable than banking industry which was moderately
unstable and utilities’ industry which was relatively stable.
Financial performance in the computer industry was also
positively related to the use of intuitive decision making by
top executives in the industry.
Cappon (1993) tested over 3000 individuals and found
out that women did not have more intuition than men. He
believed that everyone had intuition and that it could be
developed in individuals (Fields, 2001). Cappon’s request
to administer his research instrument was turned down by
many intuition-sensitive companies but companies in the
manufacturing industry were quite receptive to his request.
This might have skewed his fndings. Cappon was a medical
doctor and psychotherapist.
Dane &Pratt (2007) in examining factors contributing to
effectiveness of intuitive decision making expatiated on the
role of domain knowledge, implicit and explicit learning,
and task characteristics on the effectiveness of intuition.
They concluded that as tasks become more judgmental, the
relationship between complex, domain relevant schemas
and effective decision making becomes stronger.
Intuition and gender
One widely-held view is that successful managers
are aggressive, forceful, competitive, self confdent,
independent and have a high need for control (Hayes et al,
2004). Loden (1985) argued that women have a lower need
for control and are more cooperative than men. Green and
2011 313
Cassell (1996) suggest that women are often characterized
as relatively submissive, nurturing, warm, kind and selfess.
In a study of sex stereotypes and leader behavior; Brenner
and Bromer (1981) reported that men are described as being
more analytical and logical and women as more intuitive.
Sex differences have been cited as the reason why
women are under-represented in management; they lack
the qualities for success and cannot perform as effectively
as male managers (Hayes et al, 2004). In agreeing with
this more compassionate and intuitive gendered¬ view of
women, Clare (1999), referred to intuition as one of the
valuable contributions that women bring to management.
Studies conducted to investigate the validity of this
stereotyping did not produce consistent results.
In a study by Wajcman (1996), successful women
managers were found to be in most respects, indistinguishable
from men in equivalent positions. Alban-Metcalfe and West
(1991) found a remarkable similarity in the way female and
male managers perceived themselves at work. Donnell and
Hall (1980) found no signifcant difference between male
and female managers in their study of 1,000 matched pairs
of female and male managers. However, Eagly and Johnson
(1990) found support for the absence and presence of
differences in their Meta analysis of studies of gender and
leadership style (Hayes et. al., 2004). Pacini and Epstein’s
(1999) study showed that women perceived themselves
as intuitive. They report that women are more likely than
men to identify themselves as engaging in experiential
processing and to judge themselves as being good at it
(Aarnio & Lindeman, 2005).
In summary, the search for gender differences in
information processing style has so far produced mixed
results. Self report studies produced results showing that
men and women support the existence of gender differences
in information processing; however some of the results are
actually contradictory. Some studies showed that women
see themselves as more intuitive than men while a few
self-report and in-depth studies showed men as being more
intuitive than women. In making sense of the results, Hayes
et al. (2004), suggested that observed pattern “appears to
lend support to the utility of the structural (Kanter, 1977)
and gendered culture (Green and Cassell, 1996) approaches
to understanding behavior in organizations”. The fact that
female managers showed more intuition than their non-
managing counterparts was construed to be their way of
adapting to a male dominated environment in which success
was determined by conformity to certain modes of conduct.
Country of Operation
Merriam-Webster online dictionary (2007) had four
different defnitions for a country: an indefnite usually
extended expanse of land, the land of a person's birth,
residence, or citizenship, a political state or nation or
its territory, the people of a state or district, and rural as
distinguished from urban areas. One of the variables
studied in this research is the country of operation. Research
question four examined the impact of country of operation
on reported use of intuition by executives. The literature
review on country of operation is hence focused on culture
as countries’ distinguishing factor.
Culture can be defned as the way of life of a group of
people. Damen (1987) also defned culture as “learned and
shared human patterns or models for living”. These patterns
and models pervade all aspects of human social interaction.
The use of proverbs to study cultures is a well-known
method in Anthropology. Lovell (2001) said “proverbs
can be the eyes that provide a window to a culture’s soul.”
Prahlad (2001) did a study of Jamaican culture through
Jamaican Proverbs gathered from Reggae music. In
studying Malaysia and the United States as countries of
operation, a review of literatures containing Western and
Malaysia proverbs was conducted.
Malaysia is a country of three major ethnic groups;
Malays, Chinese and Indians. The book of Analects is one of
the most revered sources of information on Chinese culture.
The book contains most of the sayings of Confucius and
other highly respected Chinese teachers. Relevant contents
of the book were contrasted with Hofstede’s (2003)
cultural universals to develop a comprehensive outlook
on the Chinese culture. The same procedure was carried
out for Western, Indian and Malay cultures to develop a
comprehensive view of those cultures.
Hofstede (1980) defned culture as a kind of collective
programming of the mind which distinguished members
of one category of people from another. In the 1970's, he
measured elements of national cultural systems that impact
behavior in work situations. His studies produced a total
of 116,000 questionnaires in two surveys held in 1968
and 1972 (Hofstede, 2003). The studies revealed four
main dimensions on which country cultures differ. They
were labeled power distance (PDI), uncertainty avoidance
(UAI), individualism (IDV), and masculinity (MAI). Later
research, which dealt with Asians as the subject, added
the dimension called "long-term orientation”. These fve
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
314 Business Intelligence Journal July
dimensions were used to compare the two cultures through
the eyes of their proverbs.
Power distance index (PDI) focuses on the degree of
equality, or inequality, between people in the country's
society. A high power distance ranking indicates that
inequalities of power and wealth have been allowed to grow
within the society. “Malaysia’s power distance score is 104
compared to the other Far East Asian countries average of
60. This is indicative of inequality of power and wealth
within the society (Hofstede, 2003)”. A review of the book
of Analects revealed the following Chinese proverbs.
“Yu Tzu said: There are few who have developed
themselves flially and fraternally who enjoy offending
their superiors. Those who do not enjoy offending superiors
are never troublemakers. The Superior Man concerns
himself with the fundamentals. Once the fundamentals are
established, the proper way (Tao) appears. Are not flial
piety and obedience to elders fundamental to the enactment
of humaneness?" (Lau, 1992, p. 1) The concept of superiors
practiced by the Chinese is a refection of the people’s belief
in power distance. Yu Tzu in this proverb demonstrates the
need to avoid offending those who occupy higher positions
in the power structure of the society. Parents are the most
prominent superiors in Chinese worldview.
Malay proverbs dealing with power distance are listed
below:
A lost wife can be replaced, but the loss of character
spells ruin. A deep look at this saying reveals a stratifed
society that values female fairly less than their male
counterparts. It is not of the characteristics of societies in
which power distance is widely accepted.
Individualism (IDV) focuses on the extent to which
the society reinforces individual or collective achievement
and interpersonal relationships (Hofstede, 2003). A
high individualism ranking indicates that individuality
and individual rights are paramount within the society.
Individuals in these societies may tend to form a larger
number of looser relationships. A low individualism
ranking typifes societies of a more collectivist nature with
close ties between individuals. Malaysia ranked low on
individualism with a score of 26, third highest for Far East
Asian countries, behind Japan's 46 ranking, and compared
to an average of 24 for Asian countries. Corresponding
quotes from the book of Analects are stated as follows:
“Tzu Kung asked, ‘Does the Superior Man also have
things that he hates?’ Confucius said; He does. He hates
those who advertise the faults of others. He hates those
who abide in lowliness and slander the great. He hates
those who are bold without propriety. He hates those who
are convinced of their own perfection, and closed off to
anything else. How about you, what do you hate? Tzu
Kung said I hate those who take a little bit of clarity as
wisdom; I hate those who take disobedience as courage;
I hate those who take disclosing people's weak points to
be straightforwardness." (Lau, 1992, p.17) Confucius here
demonstrates the need to avoid exposing others weaknesses.
The need to outdo others is almost non-existing in Chinese
worldview. This is in sharp contrast to Western worldview
of self before others.
Malay proverbs dealing with the topic of individualism
are listed below:
If you have, give; if you lack, seek.
The body pays for a slip of the foot, and gold pays for a
slip of the tongue.
A heavy load should be borne together as well as a light
load
As a bamboo conduit makes a round jet of water, so
taking counsel together rounds men to one mind. The need
for collectivism is emphasized by these Malay proverbs.
Masculinity (MAS) focuses on the extent to which a
society reinforces, or does not reinforce, the traditional
masculine work role model of male achievement, control,
and power (Hofstede, 2003). A high masculinity ranking
indicates the country experiences a high degree of
gender differentiation. In these cultures, males dominate
a signifcant portion of the society and power structure,
with females being controlled by male domination. A low
masculinity ranking indicates the country has a low level
of differentiation and discrimination between genders. In
these cultures, females are treated equally to males in all
aspects of the society. Quotes relevant to masculinity in the
book of Analects are stated as follows:
“Being robbed, Chi K'ang Tzu was upset, and
questioned Confucius about what to do. Confucius said, If
you were desire less, they wouldn't steal from you, even
if you were to offer them a reward to do so." (Lau, 1992,
p. 12) The authentic original Chinese culture does not
favor materialism. In fact, the society esteems peace and
harmony above competitiveness. Confucius here seems to
convey the need to shun materialism.
“Do not wait for the rice to be served at the knee” is
a Malay proverb that seems to extol hard work and shun
laziness.
“He who works as a slave, eats as a king” is an Indian
proverb extolling the virtues of hard work.
Uncertainty avoidance index (UAI) focuses on the level
of tolerance for uncertainty and ambiguity within the society
- i.e. unstructured situations (Hofstede, 2003). A high
2011 315
uncertainty avoidance ranking indicates the country has a
low tolerance for uncertainty and ambiguity. This creates a
rule-oriented society that institutes laws, rules, regulations,
and controls in order to reduce the amount of uncertainty.
A low uncertainty avoidance ranking indicates the country
has less concern about ambiguity and uncertainty and has
more tolerance for a variety of opinions. This is refected
in a society that is less rule-oriented, more readily accepts
change, and takes more and greater risks.
Malaysia is relatively low in uncertainty avoidance.
The country’s (UAI) is only 36, compared to an average of
63 for the Far East Asian countries. A search through the
book of Analects revealed the following sayings about the
dimension:
Chi Lu asked about serving the spirits. Confucius said,
"If you can't yet serve men, how can you serve the spirits?"
Lu said, "May I ask about death?" Confucius said, "If you
don't understand what life is, how will you understand
death?" (Lau, 1992, p. 11) Confucius in this teaching shows
how Chinese people are generally intolerant of ambiguity.
The emphasis in this teaching is to focus on what you know
and leave the ambiguous for other people.
Malay proverbs dealing with this dimension are listed
below:
Do not empty the water jars just because you hear the
thunder in the sky
Do not leave the tortoise at your feet and hunt for the
turtle on the sea shore. These proverbs also demonstrate the
society’s disdain for uncertainty.
Indian proverbs relevant to uncertainty avoidance index
were also gathered and listed below:
Don't bargain for fsh which are still in the water.
He who is a guest in two houses, starves.
Long-term orientation (LTO) focuses on the extent to
which the society embraces, or does not embrace long-term
devotion to traditional, forward thinking values. High long-
term orientation ranking indicates the country subscribe
to the values of long-term commitments and respect for
tradition. This is thought to support a strong work ethic
where long-term rewards are expected as a result of today's
hard work. A low long-term orientation ranking indicates
the country does not reinforce the concept of long-term,
traditional orientation. In this culture, change can occur
more rapidly as long-term traditions and commitments do
not become impediments to change. LTO related quotes
from the book of Analects are stated below:
“Confucius said: The superior man stands in awe of
three things: He is in awe of the decree of Heaven; He is in
awe of great men; He is in awe of the words of the sages.
The inferior man does not know the decree of Heaven;
he takes great men lightly, and laughs at the words of the
sages” (Lau, 1992, p. 16) Respect for tradition is implied by
the reference to great men in this teaching. Malaysia’s score
on this index is not available.
Only seven (7) countries in the Geert Hofstede (2003)
research have individualism (IDV) as their highest
dimension: USA (91), Australia (90), United Kingdom
(89), Netherlands and Canada (80), and Italy (76). The high
individualism (IDV) ranking for the United States indicates
a society with a more individualistic attitude and relatively
loose bonds with others. The populace is more self-reliant
and individuals look out for themselves and their close
family members. Review of Western proverb literatures
produced two groups of proverbs relating to individualism.
The frst group promotes individualism while the second
group promotes cooperation. The two groups are presented
below.
Every man must carry his own cross If you want a thing
done right, do it yourself If you want breakfast in bed,
sleep in the kitchen Paddle your own canoe Good fences
make good neighbors You are responsible for you The
need for individuals to fetch for themselves is expressed
in these proverbs. The central theme of the six proverbs is
that individuals should be prepared to solve their problems
without relying on others for help. This is in line with the
USA’s high score of 91 on this cultural dimension. However,
the literature review also revealed some proverbs that
promote cooperation more than individualism. Examples
of proverbs in this category are presented below.
A bicycle can't stand on its own because it's two-tired.
Honey catches more fies than vinegar.
No man is an island
The nail that sticks out gets pounded
A big tree attracts the woodsman's axe
The central theme of these fve proverbs seems to be at
odds with the earlier six. They encourage cooperation rather
than individualism. The possibility of some of the proverbs
being foreign is also real since the American society is not
entirely white. The most unusual of the proverbs is fourth
one that talks about nails getting pounded. The researcher
felt this particular proverb might have been imported from
Asia.
The next highest Hofstede (2003) dimension for the
United States is masculinity (MAS) with a ranking of 62,
compared with an average of 50 for all countries. This
indicates the country experiences a higher degree of gender
differentiation of roles. The male dominates a signifcant
portion of the society and power structure. This situation
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
316 Business Intelligence Journal July
generates a female population that becomes more assertive
and competitive, with women shifting toward the male role
model and away from their female role.
The literature review for this section turned up proverbs
that mostly support USA’s score on the dimension. The
supporting proverbs are presented before those that are not
supportive of the position.
Half a loaf is better than none.
If at frst you don't succeed, try, try again.
It's the early bird that gets the worm.
Make hay while the sun shines.
The need to keep trying until success is achieved is
expressed in the second proverb, this is indicative of the
male achievement model valued by the culture. The essence
of timeliness in trying to achieve the desired societal status
is emphasized by the third proverb. Two proverbs that are
not supportive of the male dominant, achievement oriented
tendency are listed below. The two proverbs look more like
what one would fnd in Asian cultural literatures.
Winning isn't everything.
Health is better than wealth.
The United States was included in the group of countries
that had the long term orientation (LTO) dimension
added. The LTO is the lowest dimension for the US at 29,
compared with an average of 45 for all countries. This low
LTO ranking is indicative of the societies' belief in meeting
its obligations. The frst group of proverbs gathered from
the literature review supports the low score of the country
on the dimension, while the second group seems to suggest
that the score on the dimension should have been higher.
The supporting proverbs are presented hereby presented.
Never put off till (until) tomorrow what you can do
today.
No time like the present.
A stitch in time saves nine
The second proverb in this second group is actually
more favorable towards long term orientation as people
are encouraged to make attempts to understand the past in
dealing with the future. The third is also similar in its theme
as people are encouraged to focus more on the long run.
All things come to him who waits.
He who fails to study the past is doomed to repeat it.
History repeats itself.
Nature, time, and patience are three great physicians.
The next lowest ranking Dimension for the United
States is power distance (PDI) at 40, compared to an
average of 55 for all nations. This is indicative of a greater
equality between societal levels, including government,
organizations, and even within families. This orientation
reinforces a cooperative interaction across power levels and
creates a more stable cultural environment.
A cat may look at a king.
Green leaves and brown leaves fall from the same tree.
If you want to judge a man's character, give him power.
Power corrupts; absolute power corrupts absolutely.
The society’s disdain for power permeates through the
four proverbs. This is also refected in the country’s very
low PDI score. Egalitarianism is one of the central tenets
of the American society. The frst proverb relates human
freedom to look to that of a cat. “If a cat may look at the king
- then I have a right to look where I please” Egalitarianism
is also further stressed by the fact that we all emanate from
the same source as expressed in the green leaf proverb. The
proverb implies that we are all the same inside regardless of
what we look like outside.
The last Geert Hofstede (2003) Dimension for the US is
uncertainty avoidance (UAI), with a ranking of 46, compared
to an average of 64 for all countries. A low ranking in the
uncertainty avoidance dimension is indicative of a society
that has fewer rules and does not attempt to control all
outcomes and results. It also has a greater level of tolerance
for a variety of ideas, thoughts, and beliefs. Researcher’s
review of Western proverb literatures produced two groups
of proverbs relating to uncertainty. The frst group promotes
risk taking while the second group promotes the need to
exercise caution. The two groups are presented below.
It's easier to ask forgiveness than permission.
It is better to die on one's feet than live on one's knees.
He who hesitates is lost.
He who dares wins
Don't cross a bridge before you come to it.
A watched pot never boils.
A coward dies a thousand times before his death.
The valiant never taste of death but once.
This frst group of proverbs encourage risk taking.
The ffth proverb claims fretting about future problems is
superfuous and the seventh proverb decries cowardice. It
teaches that worrying about a forthcoming disaster may
cause as much (or even more) pain as the disaster when it
occurs (but does neither change it nor make it easier). The
central theme of this group of proverbs is that fretting is
more destructive than risk taking. This position is affrmed
by the country’s low UAI score.
The second group promotes cautious approach to risk
taking and the proverbs in this group are presented below.
A bird in the hand is worth two in the bush.
A picture is worth a thousand words.
Cobbler, stick to thy last.
2011 317
Don't burn your bridges before they're crossed.
Don't count your chickens before they're hatched.
It's better to be safe than sorry.
The frst proverb in this second group asserts that what
you already have is worth more than what you dream about
and the third proverb tries to encourage people to stick to
what they know. The fourth proverb suggests that people
should not act in ways that would leave them with no
alternatives. The mitigating effect of this second group of
proverbs might be responsible for the middle of the road
ranking given to USA on the index.
Methodology
The quantitative research employed Agor’s Intuitive
Measurement Survey (with permission by copyright owner).
The AIM survey was administered to 100 participants from
the US and 100 participants from Malaysia. Questions on
the survey were developed to measure all independent and
dependent variables.
Questions on the survey measured the following
variables: Gender, Management Level, Country of
Operation, and Reported use of intuition in decision
making. Reported use of intuition in decision making was
the only dependent variable in the study, the remaining
three variables were independent variables.
Validity and Reliability of the Instrument
AIM Survey (Agor’s Intuitive Measurement Survey)
is a modifed MBTI (Myers-Briggs Type Indicator). The
instrument, therefore, uses the reliability and validity of
MBTI (Agor, 1984). Studies have found strong support
for construct validity, internal consistency, and test related
reliability of MBTI instrument (Thompson & Borello,
1986). Further, the instrument was designed to best measure
all the variables in this study. The questionnaire was
modifed and simplifed so it contained clear instructions,
questions, and possible answers.
Research Questions
The study’s three research questions explored the
relationship between the study’s independent and
dependent variables, using the following subjects and their
respective management experience: Malaysia’s business
owners, executives, managers and supervisors; US business
owners, managers, executives and supervisors; Malaysia’s
male and female business owners, executives, managers
and supervisors as well as United States male and female
business owners, executives, managers and supervisors.
Research Question 1: What is the relationship between
management level and reported use of intuition in decision
making?
Research Question 2: What is the relationship between
country of operation and reported use of intuition in
decision making?
Research Question 3: What is the relationship between
country of operation and use of intuition in decision
making?
Assumptions and Limitations
The following assumptions were formulated that were
central to the design of this research.
1. Respondents understand the questions and are able
to answer all of them in the questionnaire.
2. Answers to the questionnaire are given with the
respondent’s knowledge and that answers were truthful.
3. Data collection process was reliable.
4. Analysis tools were accurate
Below are some limitations that may infuence the
results. These limitations include:
1. The samples were a combination of criterion,
maximum variation and convenience based selection of
companies in Malaysia and the United States.
2. The United States samples were selected from the
East and West coast of the United States.
Findings
Research Question 1: What is the relationship between
management level and reported use of intuition in decision
making?
Finding 1.1: Malaysia’s male supervisors’ use of
intuition is lower than US male supervisors’. Mean scores
of Malaysia’s male supervisors 6.40; US male supervisors
6.45. Signifcance is at the .000 level.
Finding 1.2: Malaysia’s male supervisors’ use of
intuition is lower than US male managers’. Mean scores
of Malaysia’s male supervisors 6.40; US male managers
7.8889. Signifcance is at the .000 level.
Research Question 2: What is the relationship between
sex and reported use of intuition in decision making?
Finding 2.1: Malaysia’s male supervisors’ use of
intuition is lower than US female business owners. Mean
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
318 Business Intelligence Journal July
scores of Malaysia’s male supervisors 6.40; US female
business owners 7.0476. Signifcance is at the .000 level.
Research Question 3: What is the relationship between
country of operation and use of intuition in decision
making?
Finding 3.1: Malaysia’s male supervisors’ use of
intuition is lower than US female managers’. Mean scores
of Malaysia’s male supervisors 6.40; US female managers
6.8333. Signifcance is at the .002 level.
Finding 3.2: Malaysia’s male supervisors’ use of
intuition is lower than US female executives’. Mean scores
of Malaysia’s male supervisors 6.40; US female executives
7.5200. Signifcance is at the .000 level.
Conclusion
The study discovered that United States managers
were less rational in their managerial decision making
than Malaysian managers in fve of fve categories. These
fndings can be quite signifcant to global business managers
intending to shift more of their activities to Asia in the near
future. Multinational corporations would have to provide
more data for their Malaysian managers in order to make
them more comfortable with their decision making tasks
as they tend to rely more on data based decision-making
techniques.
Business schools in the West may also need to redesign
their curriculum as more business managers feel more
comfortable with intuitive decision making techniques.
This will make business schools more relevant to what
obtains outside school walls and produce graduates that will
be more amenable to top executive position appointments.
The author would also like to recommend more
comparative studies in managerial decision-making so that
a clearer picture of intuitive managerial decision making
can emerge.
References
Aarnio, K., & Lindeman, M. (2005). Superstition,
education and thinking styles. Personality and
Individual DiVerences,39, 1227–1236.
Agor, W.H. (1984). Intuitive management: Integrating
left and right brain management skills. Englewood’s
Cliffs, NJ: Prentice Hall.
Alban Metcalfe B and West M (1991) "Women in
Management" in J Firth-Cozens and M West (eds)
Women at Work Milton Keynes: Open University
Press, 154-171.
Brenner, O.C., & Bromer, J. A. (1981) Sex stereotypes
and leaders behavior as measured by the agreement
scale for leadership behavior. Psychological
Reports, 48, 960-962.
Cappon, D. (1993). The Anatomy of Intuition:
What's your Intuition Quotient? Psychology
Today, Retrieved August 15, 2006 from http://
www.psychologytoday.com/articles/199305/the-
anatomy-intuition.
Clare, D. K. (1999) Workplace devaluation: Learning
from Experience Digital Library and archives (etd-
042699-162003).
Damen, L. (1987). Culture learning: The ffth
dimension on the language classroom. Reading,
MA: Addison-Wesley.
Dane, E., & PRATT M. G.(2007) Exploring intuition
and its role in managerial decision making Academy
of Management Review 32: 33–54.
Eagly, A. H., & Johnson, B. T., (1990). Gender and
leadership style: a meta-analysis. Psychological
Bulletin, 108, 233-256.
Fields A. F. (2003) A Study of Intuition in Decision-
Making using Organizational Engineering
Methodology. Dissertation Abstracts International,
66 (07), 2474. (UMI No. 3182016)
Green, E. & Cassell, C. (1996) Women Managers,
Gendered Cultural Processes and Organizational
Change Gender, Work & Organization, 3: 168–178.
doi: 10.1111/j.1468-0432.1996.tb00057.x
Hayes, R. H., Pisano, G. P., Upton, D. M., &
Wheelwright, S. C. (2004) Operations, Strategy,
and Technology: Pursuing the Competitive Edge
Somerset, New Jersey: Wiley.
Hofstede, G. (1980), Culture’s consequences:
International differences in work related values,
Thousand Oaks, CA: Sage Publications, Inc.
Hofstede, G. (2003) Geert Hofstede cultural dimensions
Itim International Retrieved August 10, 2006 fromhttp://www.geert-hofstede.com/hofstede_china.
shtml.
Kanter, R.M. (1977). Men and women of the
corporation New York: Basic Books.
Khatri, N., & Ng, H. A. (2000). The Role of Intuition
in Strategic Decision Making Human Relations.53:
57-86.
Lau, D.C. (1992). The Analects by Confucius
Angelfre.com Retrieved October 02, 2006 http://
www.angelfire.com/ego/nolimitz/confucius/
analects.html.
2011 319
Loden, M. (1985) Feminine leadership, or, How to
succeed in business without being one of the boys
New York: Times Books.
Lovell, J. (2001). The proverbs of a culture refect
much of its attitudes. Bigpond.net.au Retrieved
August 10, 2006 fromhttp://users.bigpond.net.au/
bwi28/pro/julie.html
Pacini, R., & Epstein, S. (1999). The relation of
rational and experiential information processing
styles to personality, basic beliefs, and the ratio-
bias phenomenon. Journal of Personality and Social
Psychology, 76, 972-987.
Prahlad, A. (2001) Reggae wisdom: Proverbs in
Jamaican music. Jackson, MS: University ' Press of
Mississippi
Thompson, B., and Borello, G. (1986a). "Construct
validity of the Myers-Briggs Type Indicator".
Educational and Psychological Measurement 46:
pp 745-752.
Wajcman, J. (1996), Desperately Seeking Differences:
Is Management Style Gendered? British Journal of
Industrial Relations, 34: 333–349.
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
320 Business Intelligence Journal July
FRANCHISING AND ORGANIZATIONAL PERFORMANCE:
EMPIRICAL INVESTIGATION OF SELECTED FAST FOOD
RESTAURANTS IN NIGERIA
Olu Ojo
Department of Business Administration College of Management
and Social Sciences Osun State University
P. M. B. 2008, Okuku
Osun State, Nigeria
Email: [email protected]
I. A. Irefin
Technology Planning And Development Unit Obafemi Awolowo University
Ile-Ife, Osun State, Nigeria
Email: [email protected]
Abstract
This study tries to gain insight into the relationship between franchising and organizational performance using selected fast food
restaurants in Nigeria as case study. It specifcally investigates the effects of franchising types and franchising ownership on organizational
performance. This research work was undertaken in order to fll the existing gap in literature with regards to the effects of franchising
types and franchising ownership on organizational performance. Thus, this study will add to the existing level of understanding on
the subject of investigation and add to the existing literature on the subject especially in underdeveloped countries where such study
is not popular. The survey research design was employed. Primary data were used and they were collected through the administration
of question to our respondents. Data collected were analyzed using descriptive method and simple percentage. Three hypotheses were
advanced and tested with the aid of correlation coeffcient. The fndings show that there is positive relationship between franchising types
and organizational performance and that positive relationship also exists between franchising ownership and organizational performance.
The author concludes by saying that franchising as a business form is fast becoming the in thing in Nigerian business arena and it is
therefore a good investment opportunity for upcoming entrepreneurs. Key words: Franchising, business organizations, franchising
relationships, organizational performance, franchising types, and franchising ownership.
Franchising occupies a prominent position in
contemporary business world and increasingly provides a
common channel for frm growth, entrepreneurial wealth
creation, and national development. In a competitive
market each business organization must strive to sell its
products and/or services by creating a competitive edge
over its competitors via multi product development, price
variation and market segmentation.
The various franchising types have their cost and beneft
owing to product differentiation, creation and affliation
requirements and therefore, the most suitable type to be
chosen by various interested parties becomes critical.
Therefore people seek to fnd the most suitable and most
productive for their franchise business that will yield returns
and improve quality and effectiveness in a competitive
environment. The ownership of the franchise relationship
is also very important. Countries with centralized and rigid
organizational system like France and China will prefer
that ownership be given to the franchisee and royalties be
paid to the franchiser periodically and this could lead to
lower quality levels or higher prices in franchised outlets.
The other which is ownership with the agent, will also
increase contracting cost and thus costlier to write contracts
and enforce, than those with employed managers by the
franchiser. Therefore, this confict in the moderate, total and
even non ownership of the franchise by either the principal
or the agent has become a problem to be solved.
Based on the above research problems, this study is
expected to provide answers to the following questions:
(i) How does franchising types affect the performances of
2011 321
Ojo O., Irefn I. A. - Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants in Nigeria
Olu Ojo, I. A. Irefn
both the franchiser and franchisee? (ii) Does franchising
ownership model have effect on return on investment of
the franchiser and franchisee? And (iii) What is the effect
of franchising relationship on organizational performance?
In order to answer the above research questions, the
general objective of this research is to explore the extent
to which franchising relationship can affect organizational
performance. The specifc objectives of this study are
(i) To examine the effect of franchise types on franchise
performance within a highly competitive environment. (ii)
To ascertain ownership models of franchising relationship
towards organizational performance. And (iii) To examine
the effect of franchising relationship on organizations
performance.
Franchising has received a great deal of attention over
the past few years. Despite the plethora of research on
franchising, some important gaps exist in our understanding
of this important organizational form. Although some
studies provide a useful starting point for research, there
are other factors associated with success that are unique to
franchising that should be considered. These include the
effect of franchising types on the performances of both
the franchiser and franchisee, and the effect of franchising
ownership model on the return on investment of the
franchiser and franchisee?
A number of studies provide support for the assumption
of greater franchise success (for example, Castrogiovanni,
Justis, and Julian 1993; Justis, Castrogiovanni, and Chan
1992). However, other recent studies are critical of this
position and argue that franchises exhibit higher rates of
frm discontinuance and lower mean proftability than
independent businesses (Bates 1995). To date the evidence is
mixed. Because of these contradictory results, the question
of whether franchising type and franchising ownership
improves or worsens franchising performance is still worthy
of further research such as the one undertaken in this study.
Besides, most of the existing studies on franchising were
conducted in developed countries using archival data. This
means that there is a major gap in the relevant literature on
developing and underdeveloped countries in which Nigeria
rightly belongs to that has to be covered by research. This
research tries to fll this gap by studying the situation of
Nigerian fast food industry and providing more empirical
evidence on franchising performance.
Literature Review
A number of strategies are available to the entrepreneur
for the expansion of his venture. However, one of these
strategies that are commonly ignored by entrepreneurs
in the new venture expansion is franchising (Ojo, 2008).
Although, franchising is a remarkable organizational
form, in the common economic parlance, it is a hybrid
organizational form of business ownership (Ojo, 2009).
Different defnitions have been given to franchising by
different people over time. Hisrich, Peters and Shepherd
(2005) see franchise as a form of new entry that can
reduce the risk of downward loss for the franchise. They
see franchising as an alternative means by which an
entrepreneur may expand his or her business by having
others pay for the use of the name, process, service etc. It
can be used as a growth mechanism by the organization
(i.e. franchiser).
Franchising dates back to at least 1850’s; when Isaac
Singer, who made improvements to an existing model of a
sewing machine, wanted to increase the distribution of his
sewing machines. His efforts, though unsuccessful in the
long run, was among the frst franchising effort in the United
States. Slightly later, yet much more successful, example
of franchising was John Pemberton’s franchising of Coca-
cola. Early American examples include the telegraph
system, which was operated by various railroad companies
but controlled by western union, an exclusive agreement
between automobile manufactures and operators of local
dealership. Modern franchise came to prominence with the
rise of franchised-based food service establishments. This
trend started as early as 1919 with quick service restaurants
such as A&W Root Beer. In 1935, Howard Deering
Johnson teamed up with Reginald Sprague to establish
the frst modern restaurant franchise. The idea was to let
independent operators use the same name, food, supplies,
logo and even building design in exchange for a fee. The
growth in franchises picked up steam in the 1930”s when
such chains as Howard Johnson started franchising motels.
The 1950”s saw a boom of franchise chains in conjunction
with the development of Americans interstate highway
system. Fast food restaurant, motel chains exploded. In
regards to contemporary franchise chains, Mc Donald’s
is arguably the most successful worldwide with more
restaurant units than any other franchise network (Ike-
Okoh, 2006).
Business format franchising, a type of franchise in 19th
century, took off in 1850’s.
Business format franchising is the form of franchising
most commonly associated with the franchise concept
(Alon, 2004). A franchiser licenses an entire way of doing
business under a brand name. This variety of franchising
is prevalent in accounting services, auto accessories, auto
Business Intelligence Journal - July, 2011 Vol.4 No.2
322 Business Intelligence Journal July
rentals, campgrounds, cleaning systems, fast food, food
retailing, motels/hotels, real estate, and schools. Business
format franchising involves packaging a mode of business,
attracting a supply of capable and dedicated entrepreneurs,
selecting superior prospects, training them in the minute
details of the business operations, providing assistance in
setting up the business at specifc outlets, and maintaining
an ongoing business that is proftable for the franchisor and
the collective franchises. The relationship entails continued
provision of benefcial services such as advertising and
new product development by the franchises and continued
provision of royalties from the franchisees to the franchisors.
Business format franchising represents a complete
package that allows the franchisee to use the format
provided by the franchiser, while retaining independence
as a business. The franchiser typically sells the franchisee
a right to use his intellectual property in return for a lump
sum payment and an annual royalty fee based on sales
for a specifed period of time (Miller and Grossman,
1990). In addition, the franchisee usually agrees to adhere
to franchiser requirements for product mix, operating
procedures and site selection (Rubin, 1978). Business
format franchising is a popular example of a hybrid
organizational format that incorporates elements of both
markets and hierarchies (Williamson, 1991). It is a hybrid
alternative since the franchiser and franchisee both retain a
degree of ownership and authority over the use of the trade
name, operating procedures and the location of outlets and
contracts with independent entrepreneurs to operate the
units (Child, 1987).
Franchising as a business seems to have achieved more
respectability in the fast food industry than ever before as
some franchised brands command noticeable lead today in
terms of awareness, product strength and value. The growth
of the fast food industry in Nigeria has been energetic with
Tantalizer, Captain Cook, Sweet Sensation and Mr. Biggs
are among the most recognized franchise chains featuring
more up from market local menu. It is important to note that
the Nigerian fast food market is still in its infant stage, far
from maturity.
What May be Bought in A Franchise
According to Hisrich et al. (2005), what you may buy in
a franchise as a franchisee includes:
• A product or service with established market ad
favorable market,
• A patented formula or design,
• Trade name or trade marks
• Financial management system for controlling the
fnancial revenue
• Managerial advice from experts in the feld
• Economies of scale for advertising and purchasing
• Head offce services
• A tested business concept
Methods and Materials
This section describes the research methodology used
in the study. Survey research design was utilized in this
study. The theoretical population of study consists of all the
franchised fast food restaurants in Lagos State. The choice
of Lagos State stems from the fact that it is the commercial
centre of Nigeria and that overwhelming majority of
franchised fast food restaurants were concentrated in Lagos
State. The offce in charge of regulating franchise business
in Lagos State reveals that there are about twenty fve
(25) franchised outlets existing in Lagos State. Therefore
these franchised outlets were the theoretical population.
Our study population consists of the entire staff of 25
registered franchised fast food restaurants in Lagos State.
Both probability and nonprobability sampling methods
were used. Stratifed probability sampling method was
used to classify the staff of the franchised restaurants
into three groups: lower level management, middle level
management, and top level management. Combination
of convenience and judgmental nonpropability sampling
methods were used to select our sample elements. These
methods allow a large number of respondents to participate
in the research over a short period of time (Ojo, 2003).
Primary method of data collection was used in this study.
The data consists of a number of items in structured
questionnaire that was administered to the respondents. We
also utilized 5-point Likert scale in our questionnaire. The
decision to structure the questionnaire is predicated on the
need to reduce variability in the meanings possessed by the
questions as a way of ensuring comparability of responses.
The questionnaire is titled “Franchising and Organizational
Performance Questionnaire.” Two hundred and seventy-
fve (275) respondents were selected for the study. However,
only 230 of them flled their questionnaire and returned it
and were used for fnal analysis in the study. This means
that the return rate of completed questionnaire was 83.6%.
To ensure the validity and reliability of the questionnaire
used for the study, even numbers of experts were consulted
to look at the questionnaire items in relations to its ability
to achieve the research objectives, level of coverage,
2011 323
According to the table 3 above, 4.3% of the respondents
strongly disagree with the statement. 17.4% of the
respondents disagree with the statement, and 4.3% of the
respondents are undecided about the statement. Of the
remaining 73.9% of the respondents, 43.5% of them agree
with the statement and 30.4% of them strongly agree with
the statement. This indicates that majority of the respondents
are either agree of strongly agree with the statement.
comprehensibility, logicality, and suitability for prospective
respondents. Data collected from the questionnaire were
analyzed with the aid of descriptive statistical techniques
such as total score and simple percentage, while inferential
statistics such as correlation coeffcients was used to proof
the level of signifcance in testing stated hypotheses.
Data Presentation, Analysis and Results
All the items in the questionnaire were analyzed.
However, only the key ones featured in this section.
Table 1: Age Distribution of Respondents
Age of
Respondents
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid Below 20 Years 20 8.7 8.7 8.7
21- 40 Years 180 78.3 78.3 87.0
41 - 60 Years 30 13.0 13.0 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
From the above table it can be observed that 8.7% of the
respondents fell within the age bracket of less than 20years.
Overwhelming majority of the respondents (78.3%) are in
the age bracket of 21 to 40 years. The remaining 13.0%
of the respondents are within the age bracket of 41 – 60
years. This clearly shows that the larger percentages of the
respondents are young persons in the middle of their career.
Table 2: Respondents’ Position in the Organization
Respondents’ Position in the
Organization
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid Valid Lower Level
Management
40 17.4 17.4 17.4
Middle Level
Management
180 78.3 78.3 95.7
Top Level
Management
10 4.3 4.3 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Table two above presents the position of the respondents
on the organizational hierarchy. The table reveals that 17.4
of the respondents are at the lower level management. The
bulk of the respondents (78.3%) are in the middle level
management, while the remaining 4.3% of the respondents
are in the top level rung of the organization. This shows the
highest responses came from the middle level managers in
the organization.
Table 3: This type of Franchising Relationship Increases profts
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 10 4.3 4.3 4.3
D 40 17.4 17.4 21.7
U 10 4.3 4.3 26.0
A 100 43.5 43.5 69.5
SA 70 30.4 30.4 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Table 4: Type of Franchising has Effect on Money Invested in
the Business
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 30 13.0 13.0 13.0
D 20 8.7 8.7 21.7
U 20 8.7 8.7 30.4
A 100 43.5 43.5 73.9
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Ojo O., Irefn I. A. - Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants in Nigeria
Olu Ojo, I. A. Irefn
Business Intelligence Journal - July, 2011 Vol.4 No.2
324 Business Intelligence Journal July
Table 4 above reveals that 13% of the respondents
strongly disagree that the type of franchising has effect on
money invested in the business, 8.7% of the respondents
disagree with the statement, while another 8.7% of the
respondents are undecided about the statement. Majority
of the respondents (43.5%) agree with the statement while
additional 26.1% of the respondents strongly agree with
the statement. The inference that can be drawn here is that
the type of franchising has effect on money invested in the
business if we go by the respondents’ opinion.
Table 5: The Ownership by Franchisee Agents Increases Return
on Capital Employed
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 10 4.3 4.3 4.3
D 20 8.7 8.7 13.0
U 40 17.4 17.4 30.4
A 100 43.5 43.5 73.9
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
In table 5 above, 4.3% of the respondents strongly
disagree that ownership by franchise agent’s increases return
on capital employed. A total of 8.7% of the respondents
disagree with the statement. 17.4% of the respondents are
undecided about the statement. 43.5% of the respondents
agree with the statement. The remaining 26.1% strongly
agree with the statement. Based on the above data, we
can confdently say that ownership by franchise agents
increases return on capital employed.
According to the table 6 above, 4.3% of the respondents
strongly disagree that proft is maximized when higher
ownership is vested with the franchisee. In addition,
another 4.3% of the respondents disagree with the
statement. 13% of the respondents are undecided about the
statement. However, overwhelming majority (52.2%) of the
respondents agree with the statement while the remaining
26.1% of the remaining respondents strongly agree with the
statement. Thus, we can conclude that proft is maximized
when higher ownership is with the franchisee.
Table 7: Franchising Leads to Positive Corporate Performance
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 30 13.0 13.0 13.0
D 20 8.7 8.7 21.7
U 20 8.7 8.7 30.4
A 100 43.5 43.5 73.9
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Table 7 above presents the respondents feedback to the
effect of franchising on corporate performance. 13.0% of
the respondents strongly disagree that franchising leads
to corporate performance, while another 8.7% of the
respondents disagree with the statement. Another 8.7% of
the respondents are undecided about the statement. Among
the remaining respondents, 43.5% of them agree with the
statement while 26.1% of them strongly agree. Thus, we
can inferred that franchising leads to positive corporate
performance.
Testing of Hypothesis
Our three hypotheses were tested using correlation
coeffcient.
Table 6: Proft is Maximized when Higher Ownership is with the
Franchisee
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 10 4.3 4.3 4.3
D 10 4.3 4.3 8.6
U 30 13.0 13.0 21.6
A 120 52.2 52.2 73.8
Source: Field Survey, 2009.
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
2011 325
Table 10: Correlation Between Franchising Relationship and
Organizational Performance
Types Performance
Franchising
Relationship
Pearson
Correlation
1 .535(**)
Sig. (2-tailed) .009
N 230 230
Performance Pearson
Correlation
.535(**) 1
Sig. (2-tailed) .009
N 230 230
Table 8: Correlation Between Franchising Types and
Organizational Performance
Types Performance
Types Pearson
Correlation
1 .363(**)
Sig. (2-tailed) .089
N 230 230
Performance Pearson
Correlation
.363(**) 1
Sig. (2-tailed) .089
N 230 230
(**) Correlation is signifcant at the 0.01 level (2-tailed).
The analysis above shows that there is a positive
correlation between the two variables: franchising types
and organizational performance [r = .363(**); N = 230],
meaning that the relationship between them is positive.
Hypothesis 2
H
1
: The franchisee ownership has effect on organizational
performance.
Here franchisee ownership was correlated against
corporate performance. The outcome is presented in table
9 below.
Table 9: Correlation Between Franchisee Ownership and
Organizational Performance.
Types Performance
Owners hip Pearson
Correlation
1 .570(**)
Sig. (2-tailed) .005
N 230 230
Performance Pearson
Correlation
.570(**) 1
Sig. (2-tailed) .005
N 230 230
(**) Correlation is signifcant at the 0.01 level (2-tailed).
The analysis above shows that there is a positive
correlation between the two variables, franchisee ownership
and organizational performance [r = .570(**); N = 23],
meaning that the relationship is positively related.
Hypothesis 3
H
1
: Franchising relationship has effect on organizational
performance.
In order to test the signifcant relationship between
franchising relationship and organizational performance
the Pearson product moment correlation was used. Our
data were combined and analyzed to check the relationship
and strength of the relationship between franchising and
organizational performance. The analysis is as presented
below:
Hypothesis 1
H
1
: There is a signifcant relationship between franchise
types and organizational performance.
(**) Correlation is signifcant at the 0.01 level (2-tailed).
From the above table, the analysis shows that there is a
positive correlation between the two variables, franchising
and organizational performance [r = .535(**); N = 230),
meaning that franchising and organizational performance
are positively correlated.
In all the three instances, we accept our hypotheses.
Discussion of Findings
As said earlier, this study focused on a research into
franchising and how it affects the profts of an organization
operating it. Based on analyzed data, the fndings in this
study include the followings:
i. The larger percentage (78.3) of the respondents
are young persons in the middle of their career. Their age
bracket is 21to 40 years.
ii. The bulk of the respondents (78.3%) are in the
middle level management of the organization..
Ojo O., Irefn I. A. - Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants in Nigeria
Olu Ojo, I. A. Irefn
Business Intelligence Journal - July, 2011 Vol.4 No.2
326 Business Intelligence Journal July
iii. Majority of the respondents either agree (43.5%)
or strongly agree (30.4%) with the statement that business
format type of franchising relationship has effect on money
invested in the business and also increases profts.
iv. In addition, 43.5% of the respondents agree while
another 26.1% of the respondents strongly agree that
ownership by franchisee agent’s increases return on capital
employed.
v. It is also discovered that 52.2% of the respondents
agree and 26.1% of the respondents strongly agree that
proft is maximized when higher ownership is with the
franchisee.
vi. Besides, 43.5% of the respondents agree and
another 26.1% of the respondents strongly agree that
franchising leads to positive corporate performance.
vii. Finally, all the three hypotheses tested reveal
positive correlation between business format type of
franchising and organizational performance, ownership
type and organizational performance, and franchising
relationship and organizational performance.
Conclusion
Recent trends in the business environment have brought
about innovative ways in which frms can take the lead
in their industry even in the face of great competitors.
The conclusion that is drawn from this study is that
organizations that are currently operating franchising
as a business arrangement are making large profts and
are gaining even stronger brand names through it. By
maintaining the same standards operated in the parent
company, in all their franchises they stand a greater chance
of maintaining their customer loyalties. Franchising is fast
becoming the thing in the business arena in Nigeria as more
and more companies are being drawn to it as a good strategy
for distributing their products and raising better awareness
for their companies.
References
Alon, I. (2004). Global Franchise and Development
in Emerging and Transitioning Market”, Journal of
Macro-Marketing, 24 (2), pp. 156-167.
Bates, T. (1995). Analysis of Survival Rates among
Franchise and Independent Small Business
Startups. Journal of Small Business Management
33(2), pp. 26-36.
Castrogiovanni, G. J., R. T. Justis, and S. D. Julian
(1993). Franchise Failure Rates: An Assessment of
Magnitude and Infuencing Factor. Journal of Small
Business Management 16: pp.105-114.
Child, J. (1987). Information Technology, Organization
and the Response to Strategic Challenges, California
Management Review, 30, pp. 35-50.
Hisrich, R. D., M. P. Peters, D. A. Shepherd (2005).
Entrepreneurship, New York: McGraw-Hill/Irwin.
Ike-Okoh, C. (2006). Franchising: Forwarding-
Looking Form of Entrepreneurship, Businessday
March 28, pp. 2-3.
Justis, R. T., G. J. Castrogiovanni, and P. Chan
(1992). Examination of Franchise Failure Rrates.
Proceedings of the Society of Franchising.
Minneapolis, MN: University of St. Thomas.
Miller, A., and T. Grossman (1990). Business Law,
Glenview: Scott Foresman:
Ojo, O. (2003). Fundamentals of Research Methods,
Lagos: Standard Publications.
Ojo, O. (2008). Franchising: Hybrid Organizational
Arrangement for Firm Growth and National
Development, Lex et Scientia International Journal,
Nr.XV, Vol. 2, pp. 113-120.
Ojo, O. (2009). Fundamentals of Business
Management, Lagos: Standard Publications.
Rubin, P. H. (1978). The Theory of The Firm and The
Structure of The Franchise Contract, Journal of
Law and Economics, 21, pp. 223-234.
Welsh, D.H.B., Alon, I. and Falbe, C.M. (2006), “An
Examination of International Retail Franchising
In Emerging Markets” Journal of small Business
management, 44 (1), pp. 130-149.
Williamson, O. E. (1991). Comparative Economic
Organization: The Analysis of Discrete Structural
Alternatives, Administrative Science Quarterly, 36,
pp. 269-296.
2011 327
RELATIONSHIP BETWEEN REWARDS AND EMPLOYEE’S
MOTIVATION IN THE NON-PROFIT ORGANIZATIONS OF
PAKISTAN
Nadia Sajjad Hafiza, Syed Sohaib Shah, Humera Jamsheed
MS Scholar, Department of Management Sciences, COMSATS Institute of Information Technology,
Abbottabad, Pakistan.
Khalid Zaman (Corresponding author)
MS Scholar, Department of Management Sciences, COMSATS Institute of Information Technology,
Abbottabad, Pakistan.
Email: [email protected], [email protected]
Abstract
This study empirically examines the relationship between rewards and employee’s motivation in the non-proft organizations
of Pakistan. Employees of three organizations (PERRA, World Vision and SUNGI Development Foundation) working in Khyber
Pakhtonkhuwa province of Pakistan were taken as sample of the study. Self designed questionnaire was used for data collection. 125
questionnaires were distributed and 107 were returned, hence the response rate was 85.6%. The data was analyzed using the techniques
of rank correlation coeffcient and multiple regression analysis. All the fndings were tested at 0.01 and 0.05 level of signifcance. The
result concludes that there is a direct relationship between extrinsic rewards and the employee’s motivation. However, intrinsic rewards
found an insignifcant impact on employee motivation. Key words: Rewards, Employee motivation, Non-Proft Organization (NGO),
Correlation, Regression, KPK province, Pakistan.
In today’s competitive business environment companies
are facing many challenges and among those challenges
acquiring right workforce and retaining it, is of utmost
importance. Nowadays, human asset is considered to be
the most important asset of any organization. In order to
get the effcient and effective result from human resource,
employee motivation is necessary.
Employee will give their maximum when they have a
feeling or trust that their efforts will be rewarded by the
management. There are many factors that effect employee
performance like working conditions, worker and employer
relationship, training and development opportunities, job
security, and company’s over all policies and procedures
for rewarding employees, etc. Among all those factors
which affect employee performance, motivation that comes
with rewards is of utmost importance. Motivation is an
accumulation of different processes which infuence and
direct our behavior to achieve some specifc goal (Baron,
1983).
This study will examine the employee’s motivation of an
organization with the rewards (both extrinsic and intrinsic)
given to them. For the study, particularly employees
of non-proft organizations of Pakistan will be selected,
and as a sample employees would be taken from three
non-proft organizations of Abbottabad, KPK, Pakistan,
namely, PERRA, World Vision and SUNGI Development
Foundation.
Rewards can be extrinsic or intrinsic, extrinsic rewards
are tangible rewards and these rewards are external to the job
or task performed by the employee. External rewards can
be in terms of salary/pay, incentives, bonuses, promotions,
job security, etc. Intrinsic rewards are intangible rewards or
psychological rewards like appreciation, meeting the new
challenges, positive and caring attitude from employer, and
job rotation after attaining the goal. Frey (1997) argues that
once pay exceeds a subsistence level, intrinsic factors are
stronger motivators, and staff motivation requires intrinsic
rewards such as satisfaction at doing a good job and a sense
of doing something worthwhile.
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
328 Business Intelligence Journal July
Desired performance can only be achieved effciently
and effectively, if employee gets a sense of mutual gain
of organization as well as of himself, with the attainment
of that defned target or goal. An organization must
carefully set the reward system to evaluate the employee’s
performance at all levels and then rewarding them whether
visible pay for performance or invisible satisfaction. The
concept of performance management has given a reward
system which contains; needs and goals alignment between
organization and employees, rewarding employees both
extrinsically and intrinsically. The system also suggests
where training and development is needed by the employee
in order to complete the defned goals. This training or
development need assessment of employee gives them an
intrinsic motivation.
The objective of this study is to fnd out the relationship
between rewards and employees’ performance in non-proft
organization in Pakistan. More specifc objectives are to
fnd out:
• The effect of intrinsic rewards on employee’s
performance.
• The effect of extrinsic rewards on employees'
performance and
• The effect of both intrinsic and extrinsic rewards on
employees' performance
Based on the above objectives, the present study seeks
to test the following hypothesis:
H1: There is a direct relationship between intrinsic rewards
and employee motivation.
H2: There is a direct relationship between extrinsic rewards
and employee motivation.
The signifcance of the study would be an idea about
relative importance of extrinsic and intrinsic rewards.
Management can get a better idea while preparing its reward
system that what kind of reward would be given the most
importance and at what stage through such a type of study.
This paper is organized as follows: after introduction
in section 1, literature review is carried out in section 2.
Research framework and methodology is mentioned in
section 3. Result and discussion is provided in section 4.
Final section concludes the study.
Literature Review
Human resource is one of the most important resources of
gaining competitive advantage over competitors for a frm.
And this resource can be retained and optimally utilized
through motivating it using different techniques among
which reward is of signifcance importance. Carraher et al
(2006) advocates that there should be an effective reward
system to retain the high performers in the organization
and reward should be related to their productivity. A lot of
work has been done on evaluating the relationship between
rewards and employee motivation and there exist a large
number of studies in the literature describing impact of
reward on employee motivation. In order to maximize the
performance of the employees organization must make
such policies and procedures and formulate such reward
system under those policies and procedures which increase
employee satisfaction and motivation. Bishop (1987)
suggested that pay is directly related with productivity and
reward system depends upon the size of an organization.
Organizations in today’s competitive environment want to
determine the reasonable balance between employee loyalty
and commitment, and performance of the organization.
The existing literature on individuals’ innovative
performance reveals a wide array of individual and
organizational antecedent factors. Among many individual
antecedents that infuence employees’ innovative
performance are attitudes (Williams, 2004), cognitive styles
(Scott and Bruce, 1994), personality and demographic
characteristics such as age, education background, and
prior R&D experience (Roberts, 1991 and Rothwell, 1992).
In terms of organizational antecedents, expenditure on
R&D (Hadijimanolis, 2000), cooperation with external
technology provider, leader’s infuence (Hage and Dewar,
1973), and reward system (Eisenberger and Cameron,
1996; Janssen, 2000; Mumford, 2000) are commonly cited
as factors that affect individuals’ innovative performance.
Effcient reward system can be a good motivator but an
ineffcient reward system can lead to demotivation of the
employees. Reio and Callahon (2004) concludes that both
intrinsic and extrinsic rewards motivates the employee
resulted in higher productivity.
Most of the organizations have gained the immense
progress by fully complying with their business strategy
through a well balanced reward and recognition programs
for employee. Deeprose (1994) argued that the motivation of
employees and their productivity can be enhanced through
2011 329
providing them effective recognition which ultimately
results in improved performance of organizations. The
entire success of an organization is based on how an
organization keeps its employees motivated and in what
way they evaluate the performance of employees for job
compensation.
Sometimes management pays more attention to extrinsic
rewards but intrinsic rewards are equally important in
employee motivation. Intangible or psychological rewards
like appreciation and recognition plays a vital role in
motivating employee and increasing his performance.
Andrew (2004) concludes that commitment of employees is
based on rewards and recognition. Lawler (2003) argued that
prosperity and survival of the organizations is determined
through how they treat their human resource. Ajila and
Abiola (2004) examine that intrinsic rewards are rewards
within the job itself like satisfaction from completing a
task successfully, appreciation from the boss, autonomy,
etc, while extrinsic rewards are tangible rewards like pay,
bonuses, fringe benefts, promotions, etc. Filipkowski
and Johnson (2008) examined the relationships between
measures of job insecurity, organizational commitment,
turnover, absenteeism, and worker performance within a
manufacturer. A positive relationship was found between job
insecurity and intentions to turnover, and a small negative
correlation was found between measures of job insecurity
and organizational commitment. Tosti and Herbst (2009)
discuss about behavior systems approach which can be
used to achieve a customer centered organization through
examples and reports from consultation cases. Johnson et
al (2010) establish the effects of presenting organizational
information through implicit and explicit rules on sales-
related target behaviors in a retail setting. Results indicated
that when organizational information was presented in a
specifc form, productivity was increased and maintained
longer than when presented in other forms.
Research Design and Methodology
As this study examines the impact of extrinsic and
intrinsic rewards on employee motivation, the employees
of the non-proft organizations working in KPK has been
taken as population. Employee motivation is taken as
dependent variable and extrinsic and intrinsic rewards are
taken as independent variable. The framework of the study
is given in Figure 1, 2 and 3 respectively.
Employee
Motivation
Challenging
Tasks
Recognition and
Appreciation
Empowerment
and Autonomy
Figure 1: Intrinsic Rewards and Employee´s Motivation
Source: Self Construct
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
330 Business Intelligence Journal July
Employee
Motivation
Pay/Salary
Fringe Benefits
Bonuses
Promotions
Figure 2: Extrinsic Rewards and Employee’s Motivation
Source: Self Construct
Sampling
Using convenience method of sampling, 125
questionnaires were distributed among the employees of
three selected non-proft organizations namely, PERRA,
World Vision and SUNGI working in Abbottabad. 107
questionnaires were returned, so the response rate was
85.6%.
Data Collection
Self Designed questionnaire has been developed
for data collection. Self-designed questionnaire was
divided into two parts; one containing socio-demographic
questions and the second part containing questions related
to variables that are extrinsic and intrinsic rewards and
employee motivation. Parameters for measuring extrinsic
rewards are pay, bonuses, fringe benefts, and promotions
while parameters for measuring intrinsic rewards are
empowerment and autonomy, recognition and appreciation,
and challenging tasks. Five point Likert Scale ranging
from 1 (strongly disagree) to 5 (strongly agree) was used
to measure responses. The respondents’ scores for each
construct were obtained by summing across all the item
scores of the individual variables. The Cronbach’s Alpha
reliability coeffcients for the sample are given in Table 1.
Intrinsic Rewards
Extrinsic Rewards
Employees’ Motivation
Figure 3: Research Framework
Table 1: Cronbach’s Alpha Reliability Coeffcients
Items Cronbach’s Alfa
Extrinsic Rewards 0.83
Intrinsic Rewards 0.90
Results and Discussion
Frequency Distribution
Table 2 shows the demographic information of the
respondents. Most of the respondents are falling in the age
group of 26-30 year with the percentage of 50% and then 31-
35 years of age with 21%. The demographic characteristics
also show a gender division of the respondents, majority
of the respondents’ are males, i.e. 80% representing a
bigger part of the sample group. However, 20% percent
respondents are females.
2011 331
Table 2: Frequency Distribution of Demographic Variables
Variables Frequency Percentage
Age
21-25 10 9.3
26-30 55 51.4
31-35 25 23.3
36-40 09 8.4
41-45 05 4.6
46-50 02 1.8
51-55 01 1.2
Gender
Males 77 71.9
Females 30 28.1
Education
Under graduate 03 2.8
Graduate 42 39.2
Masters 61 57.0
Others 01 1.0
Descriptive Statistics
Table 3 explains that majority of the responses regarding
pay, fringe benefts, appreciation and challenging tasks
were neutral i.e., mean value is less than 3.5. The responses
of promotions, bonuses, empowerment and motivation
show that employees consider this factor slightly more
important than other factors as mean value is greater than
3.5. Standard deviation of challenging tasks and pay shows
that these variables have extensive responses than its mean
as value indicates 1.33 and 1.24 respectively. One of the
reasons for this deviation could be the selection of three
different organizations in the sample having different pay
packages.
Variables Mean Std. Deviation
Pay (PAY) 3.2857 1.2411
Fringe Benefts (FB) 3.3571 .8934
Promotions (PRO) 3.5714 .7086
Bonuses (BON) 3.3714 .8456
Empowerment (EMP) 4.0500 .7668
Appreciation (APP) 2.9464 .8511
Challenging Task (CT) 2.9089 1.3373
Motivation (MOT) 3.9857 .4761
Table 3: Descriptive Statistics
Correlation Analysis
Spearman’s Correlation was performed to study the size
and magnitude of the relationship between the variables.
The relationship between extrinsic, intrinsic and employee
performance are shown in the Table 4.
Table 4 shows that all extrinsic rewards i.e., pay,
fringe benefts, promotions, bonuses are positive and
signifcant related with employee motivation, whereas
intrinsic rewards i.e., empowerment appreciation and
challenging task have a negative but insignifcant impact
with employee motivation. Correlation coeffcient between
fringe benefts and motivation (0.319) is the highest among
all the variables and is signifcant at 99%. Pay (0.281) and
promotions (0.254) are also signifcant at 99%. Bonuses
have a weakest correlation among extrinsic rewards but
it is signifcant at 95%. Among intrinsic rewards all are
insignifcant relationship with employee motivation.
V
a
r
i
a
b
l
e
s
M
O
T
P
A
Y
F
B
P
R
O
B
O
N
E
M
P
A
P
P
C
H
A
MOT 1.000
PAY .281** 1.000
FB .319** .342** 1.000
PRO .254** .328** .317** 1.000
BON .190* .076 .423** .330** 1.000
EMP -.042 .070 -.114 .074 .240* 1.000
APP -.019 -.179 -.172 .155 .116 .451** 1.000
CT -.023 -.250* -.289** -.166 .100 .683** .829** 1.000
Table 4: Correlation Matrix
** Correlation is signifcant at the .01 level.
* Correlation is signifcant at the .05 level.
Incremental Regression
The incremental regression is performed by removing
individual independent variables from the model and by
checking the effect on the value of R-squared. Among all
the variables removed, fringe beneft has altered the value
of R-squared to a highest degree i.e., 8.4% decreases in the
portion of the dependent variable explained by independent
variables as the value for the R-squared changes from
54.2% to 45.8%. This importance is also highlighted in the
regression result as the value of coeffcient of the variable
is highest among all the variables in their three models
respectively. The result is presented in Table 5.
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
332 Business Intelligence Journal July
Hypothesis Testing
H1: There is a direct relationship between intrinsic rewards
and employee motivation.
The results of correlation matrix rejects the hypothesis as
it gives a negative and weaker relationship of empowerment
(-0.042), appreciation (-0.019) and challenging tasks
(-0.023) with employee motivation. Intrinsic rewards are
insignifcant relationship with employee motivation.
Incremental regression also rejects this hypothesis, which
shows that empowerment and appreciation has a negative
impact on employee motivation. And only empowerment
is signifcant among intrinsic variables. Chris and Awonusi
(2004) supports that extrinsic rewards have a signifcant
impact on employee motivation while intrinsic rewards
don’t have any signifcant impact on employee motivation.
Table 5: Incremental Regression Dependent Variable: Employee Motivation
Variables OLS1 OLS2 OLS3 OLS4 OLS5 OLS6 OLS7 OLS8
Pay .145 _ .240*** .188 .131 .077 .140 .108
FB .347* .387* _ .364* .380* .374* .355* .346*
PRO .203 .236*** .235*** _ .220*** .180 .154 .134
BON .097 .077 .214*** .129 _ .039 .076 .060
EMP -.286** -.246*** -.324** -.268** -.263** _ -.272** -.205
APP -.217 -.205 -.263 -.071 -.184 -.163 _ .069
CH .377 .316 .374 .233 .337 .154 .176 _
R square .542 .528 .458 .516 .535 .494 .533 .519
F-value 3.680* 4.056* 2.786* 3.817* 4.221* 3.381* 4.162* 3.877*
D-W 1.773 1.799 1.611 1.753 1.789 1.716 1.740 1.712
*, ** and *** indicates signifcance at 0.01, 0.05 and 0.09 percent level.
The result further shows goodness of ft in the
incremental regression after removing fringe benefts in
Table 6. This decrease in the value of the R-squared shows
the importance of fringe benefts in the model.
Table 6: Results of Incremental Regression removing Fringe
Benefts
Models Values
R-squared (original) 0.542
R-squared (after the removal) 0.458
H2: There is a direct relationship between extrinsic rewards
and employee motivation.
The results of correlation matrix have supported the
hypothesis that there exist a positive/direct relationship
between extrinsic rewards and employee motivation.
Incremental regression analysis also confrms the
hypothesis, as the most signifcant variable is fringe beneft
which is an extrinsic reward.
Conclusion
Human resource is considered to be the most important
resource of an organization to remain competitive in today’s
competitive business world. Acquiring the right workforce
and then retaining that force is one of the challenges faced
by organizations and their management. The results from
this study reveal that there is a signifcant and positive
relationship between extrinsic rewards and employee
motivation but it has been observed that organizations are
not offering right amount of fnancial rewards (extrinsic
rewards) to their employees in this sector. Pay is a signifcant
factor which affects employee motivation but the results
moderately supports the hypothesis due to difference
between the pay packages of three different organizations.
Fringe benefts are very important in motivating employees
according to this study, so organizations must have to
provide all the essential fringe benefts to their employees,
it also increase their job effciency.
2011 333
On the other hand, intrinsic rewards have a weaker
impact on employee motivation. Empowerment has
negative effect on employee motivation, due to the lack
of trust between employee and his/her boss through which
employee thinks that his/her boss has over burden him
instead of thinking himself empowered. There is an indirect
relationship between appreciation and employee motivation
as employees of the organizations are not satisfed with their
pay packages. So in the absence of extrinsic rewards which
is the basic source of motivation for employee, intrinsic
rewards like recognition, appreciation and empowerment
is of little importance. Pay is potentially powerful tool
to employee motivation so the employees can only be
intrinsically motivated to perform an activity when they are
fully satisfed with the pay they are getting.
There are certain limitations of the study which can
be taken into account for further studies in the future, like
sample size was too small and only organizations working in
Abbottabad were considered. Another important limitation
was that the pay structure of the three organizations was
different from one another, so responses regarding pay
variable were deviating, generating slightly signifcant
impact of pay on employee motivation. These limitations
can be avoided in the future studies carried out in this feld,
and a more clear picture can be obtained regarding impact
of extrinsic and intrinsic rewards on employee motivation.
References
Ajila, C and Abiola, A. (2004). Infuence of Rewards
on Workers Performance in an Organization,
Journal of Social Science, 8(1), pp.7-12
Andrew, D and Kent, R. (2007). ‘The impact of
perceived leadership behaviors on satisfaction,
commitment, and motivation: An expansion
of the multidimensional model of leadership’,
International Journal of Coaching Science, 1(1), p
35-56.
Baron, R. A. (1983). Behavior in organizations, p. 123,
New York: Allyn & Bacon, Inc.
Bishop, J. (1987). The recognition & Reward
of Employee Performance, Journal of Labor
Economics Vol. 5, No. 4 Part 2: The New Economics
of Personnel pp. S36-S56.
Carraher, R, Gibson, A. & Buckley R (2006).
.Compensation in the Baltic and the USA, Baltic
Journal of Management Vol. 1, pp 7-23.
Chris A., Awonusi A. (2004). Infuence of Rewards on
Workers Performance in a Organization, Journal of
Social Science, 8(1): 7-12
Collins, J. (1995). Motivating Your Organization, 1st
edition. McGraw-Hill International, p 187.
Deeprose, D. (1994). How to recognize and reward
employees. New York: AMACOM.
Eisenberger, R., and Cameron, J. (1996). Detrimental
Effects of Reward, American Psychologist, Vol.51,
No.11, pp.1153-1166.
Filipkowski, M. and Johnson, C. M. (2008).
Comparisons of Performance and Job Insecurity
in Union and Nonunion Sites of a Manufacturing
Company, Journal of Organizational Behavior
Management, Vol.28, No.4, pp.218 – 237.
Frey. B. (1997). On the Relationship between Intrinsic
and Extrinsic Work Motivation. International
Journal of Industrial Organization, 15, p 427 – 439.
Gaertner, K. & Gaertner, G. (1985). Performance-
contingent Pay for Federal Managers.
Administration & Society, 17, 1, p 7-20.
Hadjimanolis, A. (2000). An Investigation of
Innovation Antecedents in Small Firms in the
Context of a Small Developing Country, R&D
Management, Vol.30, No.3, pp.235-245.
Hage, J., and Dewar, R. (1973). Elite Values Versus
Organizational Structure in Predicting Innovation,
Administrative Science Quarterly, Vol.18, No.3,
pp.279-290.
Janssen, O. (2000). Job Demands, Perceptions of Effort-
Reward Fairness and Innovative Work Behavior,
Journal of Occupational and Organizational
Psychology, Vol.73, No.3, pp.287-302.
Johnson, R.A., Houmanfar, R. and Smith, G.S.
(2010). The Effect of Implicit and Explicit Rules
on Customer Greeting and Productivity in a Retail
Organization, Journal of Organizational Behavior
Management, Vol.30, No,1, pp38- 48
Lawler, E. E. (2003). Treat people right. San Francisco:
Jossey-Bass Inc. McGraw-Hill Irwin.
Mumford, M.D. (2000). Managing Creative People:
Strategies and Tactics for Innovation, Human
Resource Management Review, Vol.10, No.3,
pp.313-351.
Reio, G, T. & Callahon, J. L. (2004). Affect, Curiosity,
and socialization-related Learning; a path analysis
of antecedents to job performance, Journal of
Business and Psychology, Vol.19, pp3-22.
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
334 Business Intelligence Journal July
Roberts, E.B. (1991). Entrepreneurs in High
Technology: Lessons from MIT and Beyond, New
York: Oxford University Press.
Rothwell, R. (1992). Successful Industrial Innovation:
Critical Factors for the 1990s, R&D Management,
Vol.22, No.3, pp.221-239.
Scott, S.G. and Bruce, R.A. (1994). Determinants of
Innovative Behavior: A Path Model of Individual
Innovation in the Workplace, Academy of
Management Journal, Vol.37, No.3, pp.580-599.
Tosti, D. and Herbst, S.A. (2009). Organizational
Performance and Customer Value, Journal of
Organizational Behavior Management, Vol.29,
No.3, pp.294- 314.
Williams, S.D. (2004). Personality, Attitude, and
Leader Infuences on Divergent Thinking and
Creativity in Organizations, European Journal of
Innovation Management, Vol.7, No.3, pp.187-204.
2011 335
COMPARING BANKRUPTCY PREDICTION MODELS IN IRAN
Ali Ebrahimi Kordlar, Nader Nikbakht
Faculty of management, Tehran university, Tehran, Iran
Email: [email protected]
Abstract
This paper compares the predictive power of a number of previous research models on bankruptcy prediction in Tehran Stock
Exchange (TSE) from 2001 to 2009. To compare the predictive power of these models, the adjusted R
2
(as In-sample metric) and
root mean squared error (RMSE), mean absolute error (MAE) and mean absolute percent error (MAPC) (as out-of-sample prediction
metrics). Finally, based on the estimation results of previous models in Iran, we present a fnal version logit type model that has higher
performance than other models. The empirical results of In-sample and out-of-sample prediction power indicate that our presented model
has higher adjusted R
2
and lower RMSE, lower MAE and lower MAPC, too. Key words: Bankruptcy prediction model, Logit model,
Voung test, Tehran Stock Exchange
Since Beaver (1966), an expanded literature on
bankruptcy prediction has emerged, and its impact has
spilled into the commercial world, where it has been used
in the development of several commercially employed
bankruptcy prediction models. Out of this literature have
come a number of competing empirical models with
alternative explanatory variables and alternative statistical
methodologies for model estimation.
The dependent variable in these models is commonly a
dummy variable where ''frm fled for bankruptcy" is set to
1 and other is set to 0. The independent variables are often
accounting ratios extracted from fnancial statements and
include measures of proftability, liquidity, and leverage.
Some studies also include market-based variables such
as the volatility of stock returns and past excess returns.
The accounting-based models developed by Altman (1968)
and Ohlson (1980) have emerged as the most popular
bankruptcy prediction models and are often used by
empirical accounting researchers as indicators of fnancial
distress.
Altman (1968) employs multivariate discriminate
analysis (MDA) on a list of fnancial ratios to identify
those ratios that are statistically associated with future
bankruptcy. Ohlson (1980) uses a logit ?model, which uses
less restrictive assumptions than those taken by the MDA
approach. Zmijewski (1984) adopts a probit approach that
is also based on accounting data but uses a different set
of independent variables. All of these approaches predict
future bankruptcy based on accounting ratios drawn from
frm’s fnancial statements.
More recently, Shumway (2001) has proposed a discrete-
time hazard ? model to predict a frm's bankruptcy using
both accounting and market variables. The main difference
between this model and the static logit ? model is that the
hazard model can be estimated within the logit ?framework
while using the entire life span of information (all frm-
years) for each frm. By contrast, the static logit ? model
can only incorporate one frm-year for each observation
(i.e., each observation consists of a single set of variables
observed at a single point in time).
Another stream of the bankruptcy prediction literature
focuses on market-based information. For example,
Hillegeist et al. (2004) have developed a BSM-Prob
bankruptcy prediction model that is based on the Black-
Scholes-Merton option-pricing model. Their results
indicate that the BSM-Prob model outperforms the models
of Altman (1968) and Ohlson (1980) in a series of tests.
There are also a number of papers that propose various
frm-characteristics that may be useful to future bankruptcy
prediction. For instance, Rose (1992) presents a model of
frm diversifcation in which managers use diversifcation
to reduce the risk of bankruptcy, particularly where the
ratio of the manager's frm-specifc human capital to his
non-frm-specifc human capital is high. Denis et al. (1997)
measure corporate diversifcation by the number of business
segments. Beaver et al. (2005) propose that, other things
equal, large frms have a smaller probability of bankruptcy
and that a part of this explanation is related to corporate
diversifcation. That is, corporate diversifcation and frm-
size are two frm-characteristics that may help to predict
future bankruptcy.
Hillegeist et al. (2004) compare the performance of their
BSM-Prob model against the Altman and Ohlson models in
a series of in-sample and out-of-sample tests, concluding
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
336 Business Intelligence Journal July
that the BSM model outperforms the accounting-based
models. Similarly, Chava and Jarrow (2004) examine the
relative performance of Shumway's hazard model against
the Altman and Zmijewski models, concluding that the
hazard model outperforms static logit models.
Wu, Gaunt and Gray (2010) build a new model
comprising key variables from each of the previous models
(Altman 1968; Ohlson 1980; Zmijewsky 1984; Shumway
2001; and Hillegeist et.al 2004) and add a new variable that
proxies for the degree of diversifcation within the frm.
The degree of diversifcation is shown to be negatively
associated with the risk of bankruptcy. This more general
model outperforms the existing models in a variety of in-
sample and out-of-sample tests.
Base on Zmijewski and Shirata models, Mehrani,
Mehrani, Monsef and Karami (2005) present a new
prediction model to bankruptcy prediction in Tehran
Stock Exchange (TSE). To reach this model, research
hypotheses are divided to two groups. The frst group is on
the power of classifying corporations in to bankrupt and
non-bankrupt frms. These hypotheses confrm the power
of correct classifying corporations in two bankrupt and
non-bankrupt groups. The second group is on the difference
between importances of fnancial ratios as an independent
variable. These hypotheses confrm the difference between
importances of independent variables in predicting
corporate failures.
Etemadi, Anvarirostami and Dehkordi (2009)
investigate the application of Genetic Programming
(GP) for bankruptcy prediction modeling. They use GP
to classify 144 bankrupt and non-bankrupt Iranian frms
listed in Tehran stock exchange (TSE). Then a multiple
discriminant analysis (MDA) was used to benchmarking
GP model. Genetic model achieved 94% and 90% accuracy
rates in training and holdout samples, respectively; while
MDA model achieved only 77% and 73% accuracy rates in
training and holdout samples, respectively.
Rahnamai Roud Poshti, Alikhani and Maranjouri (2009)
compare the results of prediction power of Altman and
Fulmer models and fnd that there is a signifcant difference
between the performances of two models and show that
Altman model has done more conservatively than Fulmer
model as well.
Makian, Almodaresi and Karimi (2010) fnd that an
Artifcial Neural Network (ANN) model performs much
better than the discriminant analysis and logistic regression
techniques. Moreover, the results confrm that the accuracy
of ANN model is higher than the discriminant analysis and
logistic regression techniques for predicting of bankruptcy.
The analysis also shows that none of the frms will bankrupt
in the year after the period covered in this study. Komijani
and Saadatfar (2007) fnd that in comparison with other
models, applying neural network models can improve
the potentials of fnancial managements to stand against
economic fuctuations and bankruptcy.
The rest of paper is organized as follows. Research
models and data collection process are described in section
2. Section 3 presents descriptive statistics, correlation
analysis and regression results. The last section of study
provides conclusion remarks.
Methodology
Research models
There are a number of key models that have been
developed by various researchers and presented in the
bankruptcy prediction literature over the last three decades.
A number of the most prominent models in bankruptcy
prediction models are as follows:
a. Altman (1968) multiple-discriminate analysis model
it it it it it it it
X X X X X Failing ? ? ? ? ? ? ? + + + + + + =
5 5 4 4 3 3 2 2 1 1
(1)
Where:
1
9 8 7
6 5 4
3 2 1
exp 1
?
¦
)
¦
`
¹
¦
¹
¦
´
¦
|
|
|
.
|
\
|
(
(
(
¸
(
¸
+ + + +
+ + +
+ + +
? + =
it it it it
it it it
it it it
it
CHIN INTWO FUTL
NITA OENEG CLCA
WCTA TLTA OLSIZE
Failing
? ? ? ?
? ? ?
? ? ? ?
(2)
Failing: is 0 for failed frm-years and 1 for other frm-years.
X1: Net working capital/total assets.
X2: Retained earnings/total assets.
X3: Earnings before interest and taxes/total assets.
X4: Market value of equity/book value of total liabilities.
X5: Sales/total assets.
b. Ohlson (1980) logit model
2011 337
Where:
OLSIZE: Log (total assets/GNP price-level index). The index
assumes a base value of 100 for 1988.
TLTA: Total liabilities divided by total assets.
WCTA: Working capital divided by total assets.
CLCA: Current liabilities divided by current assets.
OENEG: 1 if total liabilities exceed total assets, 0 otherwise.
NITA: Net income divided by total assets.
FULT: Funds provided by operations (income from operation
after depreciation) divided by total liabilities.
INTWO: 1 if net income was negative for the last 2 years, 0
otherwise.
CHIN: (NI
t
-NI
t-1
)/(|NI
t
|+|NI
t-1
|), where NIt is net income for
the most recent period. The denominator acts as a level
indicator. The variable is thus intended to measure the
relative change in net income.
The variable of Failing is defned similar to previous
models.
c. Zmijewski (1984) probit model
( )
it it it it it
CACL TLTA NITL Failing ? ? ? ? ? + + + + ? =
3 2 1
(3)
Where:
NITL: Net income divided by total liabilities
CACL: Current assets divided by current liabilities.
and TLTA is defned in Ohlson’s model.
d. Shumway (2001) hazard model
Now, Based on previous models, we present a logit ?type
combined model as follows:
e. Combined model
1
6 5
4 it 3
3 2 1
1
2
exp 1
?
¦
¦
)
¦
¦
`
¹
¦
¦
¹
¦
¦
´
¦
|
|
|
|
.
|
\
|
(
(
(
¸
(
¸
+ + +
+ +
+ +
? + =
? it
it it
it
it it
it
CACL NITL
INTWO OENEG
X X
Failing
? ? ?
? ?
? ? ?
(5)
Where, all variables are defned in previous models. We
expect that the predictive power of this model is higher than
that of other models.
To compare the predictive power of research models
to future bankruptcy, we use the adjusted R
2
and related
Voung (1989) test (as In-sample prediction metric) and root
mean squared error (RMSE), mean absolute error (MAE)
and mean absolute percent error (MAPC) (as out-of-sample
prediction metrics)
Data
We use the 2009 version of Tadbirpardaz (the Iranian
database of Tehran Stock Exchange) annual data fles and
sample all frms in Tehran Stock Exchange between 2001
and 2009 with suffcient data available to calculate the
research variables. In some cases whereby the required
data is incomplete, we use the manual archive in the TSE’s
library. We eliminate banks and fnancial institutions from
sample. Imposing all the data-availability requirements
yields 1,532 frm-years over the period 2001–2009. This is
the full sample that we use for testing research hypotheses.
Research results
Descriptive statistics
Table 1 provides descriptive statistics of research
variables in failed and continuous frm-years. There are
142 failed frm-years and 1390 continuous frm-years. In
our available statistical population the mean (median) of
X1, X2, X3, X4 and X5 in failed frms are -0.50 (-0.38),
-0.63 (-0.42), -0.07 (-0.05), 0.40 (0.27) and 0.61 (0.53). The
mean and median of all these variables for failed frms are
lower than the mean and median of mentioned variables
for continuous frms. The mean (median) of OLSIZE for
failed frms, -0.71 (-0.77) is lower than that of OLSIZE
1
1 5
1 4
it 3 2 1
RESIZE
exp 1
?
?
?
¦
)
¦
`
¹
¦
¹
¦
´
¦
|
|
|
.
|
\
|
(
(
(
¸
(
¸
+ +
+
+ + +
? + =
it it
it
it it
it
LAGSIGMA
LEXRETURN
TLTA NITL
Failing
? ?
?
? ? ? ?
(4)
Where:
RESIZE: Log (the number of outstanding shares multiplied
by year-end share price then divided by total market
value).
LEXRETURN: Cumulative annual return in year t-1 minus the
value-weighted TSE index return in year t-1.
LAGSIGMA: Standard deviation of the residual derived from
regressing monthly stock return on market return in
year t-1.
and other variables are defned in previous models.
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
338 Business Intelligence Journal July
for continuous frms, -0.07 (-0.15), too. Failed frms on
average have a higher debt ratio (1.31) than continuous
frms (0.65). Also, the mean of ratio of current assets on
current liabilities (CLCA) for failed frms (1.77) is greater
than this ratio for continuous frms (0.95). On average,
failed frms (-3.72) are smaller than the continuous frms
(-3.10). The mean of WCTA, CLCA and TLMTA for failed
frms (0.67, 1.77 and 0.77, respectively) is higher than the
mean of these variables for continuous frms (0.60, 0.95,
and 0.50, respectively).
The mean of NITA, FULT, CHIN, NITL and CACL (-0.14,
0.01, -0.05, -0.10 and 0.68) for failed frms is lower than the
mean of mentioned variables for continuous frms (0.13,
0.22, 0.03, 0.23 and 1.19, respectively). Furthermore, the
mean (median) of LEXRETURN and LAGSIGMA are -0.05
(-0.02) and 0.10 (0.03), respectively and are lower than
the mean (median) value of these variables for continuous
frms, 0.18 (0.15) and 0.12 (0.10), respectively.
Correlation analysis
The Pearson correlation coeffcients between research
variables are provided in Table 2. In this table, the
coeffcients that are marked with asterisk (*), are not
signifcant at the ordinary signifcance level.
Table 1. Descriptive statistics
Failed frms: 142 Obs. continuous frms: 1390 Obs.
Mean Median Maximum Minimum St.Dev Mean Median Maximum Minimum St.Dev
X1 -0.50 -0.38 0.21 -2.63 0.52 0.04 0.04 0.49 -0.69 0.17
X2 -0.63 -0.42 -0.05 -2.97 0.59 0.11 0.08 0.50 -0.20 0.11
X3 -0.07 -0.05 0.17 -0.43 0.10 0.16 0.14 0.55 -0.19 0.10
X4 0.40 0.27 5.47 0.03 0.64 1.68 0.93 15.51 0.03 2.05
X5 0.61 0.53 1.85 0.02 0.34 0.83 0.78 2.59 0.04 0.37
OLSIZE -0.71 -0.77 2.69 -3.20 1.16 -0.07 -0.15 4.41 -3.20 1.32
TLTA 1.31 1.19 3.34 0.57 0.51 0.65 0.66 1.20 0.19 0.15
WCTA 0.67 0.70 0.93 0.20 0.17 0.60 0.62 0.94 0.11 0.19
CLCA 1.77 1.52 5.49 0.64 0.91 0.95 0.88 4.95 0.26 0.39
NITA -0.14 -0.12 0.19 -0.62 0.13 0.13 0.11 0.53 -0.19 0.10
FUTL 0.01 0.00 0.41 -0.26 0.11 0.22 0.17 1.53 -0.36 0.26
CHIN -0.05 -0.02 0.94 -1.00 0.43 0.03 0.05 1.00 -1.00 0.31
NITL -0.10 -0.10 0.25 -0.26 0.09 0.23 0.16 1.52 -0.24 0.23
CACL 0.68 0.66 1.57 0.18 0.27 1.19 1.14 3.84 0.20 0.42
RESIZE -3.72 -3.76 -2.25 -4.67 0.48 -3.10 -3.13 -1.35 -4.70 0.66
LEXRETURN -0.05 -0.02 0.57 -1.00 0.22 0.18 0.15 1.46 -1.77 0.33
LAGSIGMA 0.10 0.03 0.68 0.00 0.13 0.12 0.10 0.74 0.00 0.09
TLMTA 0.77 0.79 0.97 0.15 0.15 0.50 0.52 0.97 0.06 0.21
Table 2. Pearson correlation coeffcient
X
1
X
2
X
3
X
4
X
5
O
L
S
I
Z
E
T
L
T
A
W
C
T
A
C
L
C
A
N
I
T
A
F
U
T
L
C
H
I
N
N
I
T
L
C
A
C
L
R
E
S
I
Z
E
L
E
X
R
E
T
U
R
N
L
A
G
S
I
G
M
A
X2 0.74
X3 0.42 0.60
X4 0.18 0.26 0.54
X5 0.17 0.15 0.29 0.01*
OLSIZE -0.04* 0.15 0.09 -0.08 -0.15
TLTA -0.80 -0.86 -0.56 -0.35 -0.07 -0.07
2011 339
This table presents Pearson correlation coeffcient. The coeffcients that are marked with asterisk (*), are not signifcant but others are signifcant at
the 5% level or lower.
The results of Pearson correlation coeffcients show
that the most of coeffcients between research variables are
signifcant at the 5% level or better.
Model estimation
The regression results of models and In-sample and
out-of-sample prediction metrics to compare the models
are presented in Table 3. The estimation results of Altman
(1968) model indicate that X1 (-0.12), X2 (-0.48) and X3
(-0.52) have a negative and signifcant relationship with
dependent variable (Failing) at the 1% level and X4 (0.01)
has a positive relation with dependent variable at the 5%
level but X5 has no signifcant relationship with dependent
variable. The F-statistic of this model (468.16) is signifcant
at the 1% level.
Altman (1968)
model
Ohlson (1980) model
Zemijewsky (1984)
model
Shumway (2001)
model
Combined model
Panel A: Model estimation results
Intercept 0.18** -12.87** -8.00** -16.94** -2.22**
X1 -0.12** -0.95
X2 -0.48** -13.40**
X3 -0.52**
X4 0.01*
X5 0.00
OLSIZE -0.14
TLTA 14.58** 8.21** 17.75**
WCTA -2.92
CLCA -0.60
OENEG -0.17 1.73**
NITA -1.40
FUTL -2.60
INTWO 3.79** 0.25
Table 3: The results of regression models
X
1
X
2
X
3
X
4
X
5
O
L
S
I
Z
E
T
L
T
A
W
C
T
A
C
L
C
A
N
I
T
A
F
U
T
L
C
H
I
N
N
I
T
L
C
A
C
L
R
E
S
I
Z
E
L
E
X
R
E
T
U
R
N
L
A
G
S
I
G
M
A
WCTA 0.31 -0.12 -0.11 -0.24 0.26 -0.17 0.24
CLCA -0.90 -0.61 -0.37 -0.18 -0.23 0.08 0.67 -0.39
NITA 0.51 0.68 0.91 0.55 0.20 0.12 -0.68 -0.18 -0.44
FUTL 0.19 0.36 0.61 0.57 0.12 0.02* -0.43 -0.32 -0.20 0.56
CHIN 0.06 0.09 0.29 0.16 0.09 0.01* -0.07 0.00* -0.07 0.33 0.10
NITL 0.39 0.52 0.81 0.64 0.13 0.04* -0.59 -0.23 -0.37 0.86 0.69 0.28
CACL 0.76 0.44 0.35 0.29 0.16 -0.15 -0.57 0.31 -0.77 0.39 0.28 0.07 0.51
RESIZE 0.09 0.28 0.45 0.43 -0.07 0.76 -0.26 -0.27 -0.04* 0.48 0.30 0.10 0.40 -0.01*
LEXRETURN 0.10 0.16 0.27 0.21 0.01* 0.15 -0.09 0.04* -0.09 0.29 0.03* 0.09 0.20 0.06 0.34
LAGSIGMA 0.10 0.05 0.05 0.02* 0.03* 0.05 -0.06 0.03* -0.10 0.07 -0.04* 0.06 0.01* 0.04* 0.11 0.09
TLMTA -0.34 -0.41 -0.67 -0.79 -0.07 0.10 0.54 0.28 0.30 -0.71 -0.56 -0.16 -0.73 -0.37 -0.51 -0.28 -0.08
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
340 Business Intelligence Journal July
Altman (1968)
model
Ohlson (1980) model
Zemijewsky (1984)
model
Shumway (2001)
model
Combined model
CHIN -0.25
NITL -3.21* -9.88* -0.91
CACL -0.03 0.01
RESIZE 0.01
LEXRETURN -1.25
LAGSIGMA -0.60
TLMTA
Panel B: In-sample prediction power metric
Adj. R
2
51.25 Pse. R
2
83.21 72.71 76.07 87.83
F stat. 468.16** LR stat. 904.61** 1272.92** 1083.25** 1384.16**
Voung Z 11.48** 2.21* 4.81** 3.22**
Panel C: Out-of-sample prediction power metrics
RMSE 0.21 0.14 0.16 0.14 0.12
MAE 0.14 0.03 0.06 0.05 0.03
MAPE 4.96 1.64 2.69 2.05 1.41
*, ** signifcant at the 5% and 1%, respectively
In the Ohlson (1980) model, TLTA (14.58) and the
dummy variable of INTWO (3.79) are signifcant at the
1% level and in Zemijewsky (1984) model TLTA (8.21)
and NITL (-3.21) are signifcant at the 1% and 5% level,
respectively. The results of Shumway (2001) model
indicates that TLTA (17.75) and NITL (-9.88) are signifcant
at the 1% and 5% level respectively. Finally, our presented
model show that X2 (-13.40) and OENG (1.73) are both
signifcant at the 1% level.
The results of likelihood ratio (LR) for Ohlson model
(904.61), Zemijewsky model (1272.92), Shumway model
(1083.25) and our presented combined model (1384.16)
show that all models are signifcant at 1% level, generally.
The In-sample prediction power (adjusted R
2
) of our
combined model (87.83%) is higher than that of Altman
model (51.25%), Ohlson model (83.21%), Zemijewsky
model (72.71%) and Shumway model (76.07%). the results
of Voung (1989) test show that the differences between the
prediction power of combined model and Altman model
(11.48), Ohlson model (2.21), Zemijewsky model (4.81)
and Shumway model (3.22) are all signifcant at the 1%
level.
The results show that the root mean squared error
(RMSE) of our combined model (0.12) is lower than that
of Altman model (0.21), Ohlson model (0.14), Zemijewsky
model (0.16) and Shumway model (0.14). The results
also indicate that the mean absolute error (mean absolute
percent error) of our combined model, 0.03 (1.41) is lower
than (or equal to) that of Altman model, 0.14 (4.96), Ohlson
model, 0.03 (1.64), Zemijewsky model, 0.06 (2.69) and
Shumway model, 0.05 (2.05). Thus, the results indicate that
the In-sample and out-of-sample prediction power of our
presented model in predicting failed frms is higher than that
of other previous models. Thus, our logit type bankruptcy
prediction model signifcantly outperforms other models.
Conclusion
In this paper, we examine the predictive power of a
number of bankruptcy prediction models. The models use
a range of different independent variables (accounting
information and market and frm-characteristic data) and
a range of different econometric specifcations (multiple-
discriminate analysis, logit ?, probit models). We fnd that
frms are more likely to experience bankruptcy if they
have relatively lower earnings before interest and tax to
total assets, a larger decline in net income, relatively low
working capital to total assets, or high total liabilities to
total assets.
We compare the empirical performance of a range of
bankruptcy prediction models using a series of in-sample
and out-of-sample performance metrics. Across all of these
metrics, both in-sample and out-of-sample, we fnd that, our
combined model signifcantly outperforms models from the
extant literature.
2011 341
References
Agarwal, V., and Taffer, R. (2008), "Comparing the
performance of market-based and accounting-
based bankruptcy prediction models", Journal of
Banking & Finance 32, PP: 1541-1551.
Altman, E. I. (1968), "Financial ratios, discriminant
analysis and the prediction of corporate bankruptcy",
Journal of Finance 23, PP: 589-609.
Beaver, W. H. (1966), "Financial ratios as predictors
of failure", Journal of Accounting Research 4, PP:
71-111.
Beaver, W. H., McNichols, M.F., and Jung-Wu, R.
(2005), "Have fnancial statements become less
informative? Evidence from the ability of fnancial
ratios to predict bankruptcy", Review of Accounting
Studies 10, PP: 93-122.
Begley, J., Ming, J., and Watts, S. (1996), "Bankruptcy
classifcation errors in the 1980s: an empirical
analysis of Altman's and Ohlson's models", Review
of Accounting Studies 1, PP: 267-284.
Burgstahler, D., James, J., and Noreen, E. (1989),
"Changes in the probability of bankruptcy and equity
value", Journal of Accounting and Economics 11,
PP: 207-224.
Campbell, J. Y., Hilscher, J., and Szilagyi, J. (2008),
"In search of distress risk", Journal of Finance 63,
PP: 2899-2939.
Chava, S., and Jarrow, R. A. (2004), "Bankruptcy
prediction with industry effects", Review of
Finance 8, PP: 537-569.
Denis, D. J., Diane, K., and Denis, A. S. (1997),
"Agency problems, equity ownership, and corporate
diversifcation", Journal of Finance 52, PP: 135-
160.
Dichev, I. D. (1998), "Is the risk of bankruptcy a
systematic risk?", Journal of Finance 53, PP: 1131-
1147.
Etemadi, H., Anvari Rostami, A. A., and Dehkordi,
H. F. (2009), "A genetic programming model for
bankruptcy prediction: empirical evidence from
Iran", Expert Systems with Applications March,
36, PP: 3199-3207.
Francis, J. (1990), "Corporate compliance with debt
covenants", Journal of Accounting and Research
28, PP: 326-347.
Hillegeist, S. A., Keating, E., Donald P., Cram, J. G.,
and Lunstedt. (2004), "Assessing the probability of
bankruptcy", Review of Accounting Studies 9, PP:
5-34.
Komijani, A., and Saadatfar, J. (2007), "Application
of "artifcial neural network models", in predicting
economic bankruptcy of the registered corporations
in stock market", Journal of Economic Essays 3,
PP: 11-44.
Lyandres, E., and Alexei, Z. (2007), "Investment
opportunities and bankruptcies prediction.
Working Paper, Rice University, Houston. Ohlson,
James, 1980. Financial ratios and the probabilistic
prediction of bankruptcy", Journal of Accounting
Research 18, PP: 109-131.
Makian S.N.A.D., Almodaresi S.M.T., and Karimi
Takalou S. (2010), "A comparison among artifcial
neural network, discriminant analysis and logistic
regression techniques for bankruptcy: a case study
of Kerman's frms", The Economic Research 10,
PP: 141-161.
Mehrani, S., Mehrani, K., Monsef, Y., and Karami,
Gh. R. (2005), " Implication Zmijewski and Shirata
bankruptcy prediction models in Tehran stock
exchange", The Iranian Accounting and Auditing
Review 12, PP: 105-131.
Rahnamai Roud Poshti, F., Alikhani, R., and
Maranjouri, M. (2009), "Applicational investigation
of Altman and Fulmer bankruptcy prediction
models in Tehran stock exchange", The Iranian
Accounting and Auditing Review 16, PP: 19-34.
Rose, D. (1992), "Bankruptcy risk, frm-specifc
managerial human capital, and diversifcation",
Review of Industrial Organization 7, PP: 65-73.
Shumway, T. (2001), "Forecasting bankruptcy more
accurately: a simple hazard model", Journal of
Business 74, PP: 101-124.
Stone, M. (1991), "Firm fnancial stress and pension
plan continuation/replacement decisions", Journal
of Accounting and Public Policy 3, PP: 175-206.
Vassalou, M., and Xing, Y. (2004), "Default risk in
equity returns", Journal of Finance 59, PP: 831-868.
Vuong, Q. H. (1989), "Likelihood ratio tests for
model selection and non-nested hypotheses",
Econometrica 57, PP: 307-333.
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
342 Business Intelligence Journal July
Wu, Y., Gaunt, C., and Gray, s. (2010), "A comparison of
alternative bankruptcy prediction models", Journal
of Contemporary Accounting and Economics 6,
PP: 34-45.
Zmijewski, M. E. (1984), "Methodological issues
related to the estimation of fnancial distress
prediction models", Journal of Accounting Research
22 (Suppl.), PP: 59-82.
2011 343
STRATEGIC INFLUENCE OF PROMOTIONAL MIX ON
ORGANISATION SALE TURNOVER IN THE FACE OF STRONG
COMPETITORS
Adebisi Sunday. A (Ph.D)
Department of Business Administration, Faculty of Management sciences
University of Ado Ekiti, P.M.B 5363, Ekiti State Nigeria.
Email: [email protected]
Babatunde Bayode .O
Department of Business Administration, College of Management and Social Sciences,
Osun State University, P.M.B 2008, Okuku, Osun State, Nigeria
Email: [email protected]
Abstract
This paper aim at study strategic infuence of promotional mix on organisation sale turnover in manufacturing organisation.
The research data were gathered through the use of secondary data and primary data, secondary data included 6years annual report
comprising the sales turnover (2005-2009) and questionnaire which is an instrument of primary data collection. The questionnaires
were administered to the workers of 7up Company and some customers in Solebo Estate in Lagos. The researcher adopted the simple
percentage and regression model for the analysis of the collected data.. The result of the fndings revealed that strategic promotional
mix infuences the sale turnover with little 25% while other variable not included in the variable tested takes the larger 75% that will
rapidly lead to organisation growth. Since promotional mix constitute few % of variable that can push an organisation to the highest
level, therefore other factors of marketing mix such as product development, effective pricing, distribution of right quality and quantity
to the consumers should be appropriately considered. Key words: Strategic promotional mix, marketing mix, regression model, push
and pull strategy.
It is not enough for a business to have good products
sold at attractive prices. To generate sales and profts, the
benefts of products have to be communicated to customers.
In marketing, this is commonly known as “promotion”.
Although promotion is not done only for these factors but
for other such as to build brand loyalty, to reminds and
reassure costumers, to launch a new product and maybe
to defend market share by responding to competitors’
campaigns with their own advertising A business’ total
marketing communications programme is called the
“promotional mix” and consists of a blend of advertising,
personal selling, sales promotion, public relation stool sand
direct marketing.
The organization has to convey the message about the
product on offer to its consumers. This helps in sustaining a
perennial demand for the product and in suitably positioning
it among the target audience. The process of communicating
the message is called promotion. It infuences the purchase
decision of the consumer. The different channels available
to the organization for communicating the message
constitute the promotion mix. It includes advertising, sales
promotion, and public relations. The paper examines the
different elements of the promotion mix and discusses the
issues involved therein. It also delves into the advantages of
using the Internet as a selling medium.
It is very important for every product to be promoted,
that is to say it needs to be drawn to the attention of the
market place and it’s beneft be identifed. The aim of an
organization promotional strategy is to bring existing and
potential customers to a state of relative awareness of the
organization’s product and a not just that but also to a state
of adoption.
The promotional mixes (sales promotion, publicity,
personal selling, advertising, public relation) have a stage
at which it will be most effective. Advertising and publicity
are suitable for all stages while the remaining mix can be
effective from stage three.
Sunday A. A., Bayode O. B. - Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors.
Adebisi Sunday A., Babatunde Bayode O.
Business Intelligence Journal - July, 2011 Vol.4 No.2
344 Business Intelligence Journal July
Literature review
Cole (1996) defnes promotional mix as “the means use
in bringing customers from a state of relative unawareness
to a state of actively adopting the product”. It means of
communicating with individuals, groups, or organizations
to directly or indirectly facilitate exchange of informing and
persuading one or more audience to accept an organization’s
product.
Ross (2001) sees promotional mix as “the total marketing
communication programme of a particular product”.
Adebisi (2006) defned promotional mix as “any
marketing effort whose function is to inform or persuades
actual or potential consumers about the merit a product
possess for the purpose of inducing a consumer to either
start buying or continue to purchases the frm’s product.”
Elements of Promotional Mix
Every product needs to be drawn to the attention of
the target market, and its beneft identifed. The principal
methods are;
• Advertising
• Personal Selling
• Sales promotion
• Publicity
• Public relation
The aim of an organization’s promotional strategy
is to bring existing or potential from a state of relative
unawareness of the organization’s product to a state of
actively adopting them. Several stages of customer’s
behaviours have been identifed. This has been described
in several ways, but in summary can be stated as follows;
Stages Behaviours
One Unawareness of product
Two Awareness of product
Three Interest in product
Four Desire for product
Five Conviction about value of product
Six Adoption/purchases of product
Source: Ross (2001) Stages of Product Awareness
Advertising
Advertising is the process of communication, persuasive
information about a product to the markets by means of the
written and spoken word. There are fve principal media of
advertising as follows; The press, commercial television,
direct mail, commercial radio and outdoor.
Objective of Advertising
A company’s objective for embarking on advertisement
can therefore be on any of the following objectives
enumerated below;
i. To introduce a new product or service; here
advertisement attempt to present to the prospective buyers
a new product or service and this usually near a costly and
dramatic launching of a new product or service.
ii. To expand the market to new buyers; advertisement
is done to introduce a product to new buyers who might
fnd interest or usefulness of it.
iii. To announce modifcation; a product that is already
in the market might want to be given a new face, and then
there is need for advertisement to highlight to the consumers
that modifcations had been done.
iv. To announce a price change; advertisement can
be done if the organization’s product or service price
is increased or decreased so that the consumer might be
informed.
v. To introduce a new packaging; packaging
identifcation at the point of sales is always important and
is a reason why packages are shown in advertisement.
vi. To stimulate sales promotion; advertising maybe
make special offer when business environment is static. It
could be a gift of cup or biro etc at the purchase of such
product.
vii. To educate consumers; the educational
advertisement is necessary when a commodity, service or
an offer needs careful explanation.
viii. To maintain sales; customers product which had
been in the market use advertisement to maintain sales.
ix. To challenge competition; the purpose of some
advertisement are to challenge the competitors in the
presence of buyers.
2011 345
Personal Selling
This is the process by which the seller sells to the
consumer face to face.The personal selling consists of a
selling process, which is illustrated below.Personal selling
is the most expensive form of promotion. Company that
use more of personal selling are said to be adopting push
strategy while that of advertising are using pull strategy.
Sales Promotion
Sales promotion activities are a form of indirect
advertisement, designed to stimulate sales mainly by the use
of incentives; Free sample, Twin-pack bargain, Temporary
price reduction, Special discount bonus.
Publicity
Publicity differs from other promotional mix in that it is
costless most of the time. Publicity according to Cole (1996)
is “news about the organization or its products reported in
the press”. Publicity sometimes cost but its cost is always
related with advertisement. Publicity is a very necessary
tool because it creates the good will of an organization.
Use of publicity
Publicity when properly managed by the Public Relation
Offcer of an organization can serve the following purposes;
it can be used to attract public attention, it can also be used
to maintain public visibility and used for the provision of
information to the public. Publicity often takes the form of
news released or press conferences, appearance or event
sponsorship.
Public Relation
Public relation is another form of promotion. It is the
means by which the organization related or communicates
with the environment. Public relation is aimed at better
customer relations and immediate feedback.
Personal Selling Advertisement Sales Promotion Direct Marketing Public Relation
A
d
v
a
n
t
a
g
e
s
Permit measurement of
efectiveness.
Elicit an immediate
response.
Tailors the message to ft
the customer.
Personal contact
Reaches a large group
of potential buyers for a
relative low price.
Allows strict control over the
fnal message.
Can be adapted to either
mass audience or specifc
audience segment.
Produces an immediate
consumer response.
Attracts attention and
creates product awareness.
Allows easy measurement
of result.
Provides short time sales
increase.
Generates an immediate
response.
Covers a wide audience with
targeted advertising.
Allows complete, customized,
personal message.
Produces measurable results.
Creates a positive attitude
towards a product or
company.
Enhances credibility of a
product or company.
D
i
s
a
d
v
a
n
t
a
g
e
s
Relies almost exclusively
upon the ability of the
sales person.
Involve high cost per
contact.
Does not permit total
accurate measurement of
result.
Usually cannot close sales
Is non-personal in nature.
Is difcult to diferentiate
from eforts.
Sufers from image problem.
Involves a high cost per reader.
Depends on quality and
accuracy of mailing lists.
May not permit accurate
measurement of efect on
sales.
Involves much efort directed
towards non-marketing
oriented goals.
Promotional Mix Strategy
Marketing managers may choose between two
alternative strategies to use when promoting their product,
which are; Push strategy and Pull strategy.
a. Push strategy: When a market uses the pull promotion,
it means that the product involves “pushing” through
distribution channel till it gets to the fnal consumers.
The strategy involves the producer directing his
marketing activities towards channels members to
induce them to bring the product or promote the product
to the fnal consumers. One major promotional mix use
in this strategy is advertising and sales promotion.
b. Pull strategy: In this strategy the produce the fnal
consumers to induce them to buy the product. If the
strategy is effective, the consumers demand for the
product from channel members (middlemen). This is
the most used strategy.
In this strategy, consumer’s demand pulls the product
through the channel.
The two strategies can be applied simultaneously.
However the B2C (business to consumer) use more of pull
strategy while the B2B (business to business) use more of
the push strategy.
Sunday A. A., Bayode O. B. - Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors.
Adebisi Sunday A., Babatunde Bayode O.
Business Intelligence Journal - July, 2011 Vol.4 No.2
346 Business Intelligence Journal July
Push strategy versus Pull strategy
Producer’s marketing Resellers’ Marketing
Activities Activities
Producer Middlemen Consumers
Push Strategy
Source: Cole (1996) “Push Strategy”.
Demand Demand
Producer Middlemen Consumers
Producer’s marketing activities.
Pull Strategy
Source: Cole (1996) “Pull Strategy”.
Developing an optimal Promotional Mix
The blending of all mix of promotion (personal selling,
advertisement, public relation, sales promotion and
publicity) is what leads to the achievement of marketing
objectives. Qualitative measures can be used to determine
the effectiveness of a mix components in a given market
segment. The choice of a proper mix of promotion elements
present one of the most diffcult tasks for a marketer.
Steps to developing optimal Promotional Mix
An organization must make sure that is promotional mix
is effective, if it will beneft from it. The following are the
steps involves in having an optimal promotional mix.
Step 1; identify the audience: a marketing promotion
starts with clear target audiences in mind. The audience may
be potential buyers or current users, it may be individuals,
group, special public or the general public. The target
audiences are one of the major determinants of what mix
is to be used?
Step 2; determine the needs: after having identifed the
target audience, the next thing to do is to know what their
basic needs are and is the product relevant for the audience?
Step 3; determine the promotional objectives: the
promotional activities should have a target to be meant
which is referred to as objective. Is it to increase sales? Or
is it to just create awareness? It should be clearly stated.
Step 4; choose the mix: after the expected response is
determined, then the marketer can look critically at all mix
and identify which can best satisfy their objectives.
For a promotional mix to lead to purchase, it must have
pass through some stages which is illustrated in the diagram
below:
2011 347
The buying readiness stages
Brand Ignorance Awareness Knowledge
Conviction Purchase Liking Preference
Source: Cole (1996) “The Buying Readiness Stage”
Measuring The Effectiveness Of Promotional
Mix
If an organization cannot measure it’s effectiveness in
terms of promotion, it cannot say precisely if the activities
had been successful. For an organization to measure the
effectiveness of it promotional activities, this can be done
in either of the following ways;
• Direct sales result; this method reveals the sales revenue
for each amount input into promotion. That is, it
measures the rate of sales to the expense on promotion.
• Indirect evaluation; this method focus on quantitatiable
indicators of effectiveness. For instance, the
effectiveness is measured based on the organization
study of the number of audience that actually heard
about the product during the promotional activities.
• Returns method; this method is the work of Professor
Don Schutz. He said promotional effectiveness should
be measured based on the returns of the period of
promotion. What is the proft like during promotion
and when there is no promotion?
• Direct response method; this method is concentrating
on having a way of getting response from the targeted
audience and this response should be used to measure
the effectiveness of promotion.
Methodology
Methodology is a vital process of carrying out empirical
study. It forms the background in which the procedures
employed in carrying out a research are designed. It is the
procedure that follows a step after one another of which
data gather for a research is being analyzed”.
Population of the Study
The case study of this organization has many plants in
Nigeria in places like Ibadan, Ijora, Kano, Ikeja and so on.
The researcher therefore picked the Ikeja plant as the case
study. Marketing department personnel were sampled in the
organization since it is related with the topic under study.
The number of employees in the marketing department is
forty two as given by Mr. Wemimo the sales manager.
Sampling Procedure
Sampling method is the study of a fragment of the entire
population when it is not feasible to carry out the study on
the entire population. The study used the random sampling
method because it allows equal chance for every element
of the population and therefore the sample size was thirty
while the consumers were ffty.
Data Collection Instrument
This research work is both qualitative and quantitative
in nature, and as such data were collected via the use of
both primary data and secondary data.
Regression Method of Analysis
Adebisi (2006) defned regression as “the statistical
method of predicting and determining the probable value
of dependent (?) given the value of independent (?).The
hypothesis gathered for this study will be tested at 5% level
of signifcance.
The regression formular is:
? =a + b? + ui
Sunday A. A., Bayode O. B. - Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors.
Adebisi Sunday A., Babatunde Bayode O.
Business Intelligence Journal - July, 2011 Vol.4 No.2
348 Business Intelligence Journal July
Where;
? = dependent variable
? = independent variable
a = intercept
b = slope
ui = error term.
The regression model will build on,
Hypothesis One
? (dependent variable) sales turnover
? (independent variable)Strategic promotional mix.
Hypothesis Two
? (dependent variable) market share and growth
? (independent variable)Strategic promotional mix
Table 1.
Years Turnover
Selling and Distribution
Expenses
2004 14937371 2931311
2005 17346662 2207731
2006 22071731 4557690
2007 27309123 5576791
2008 30572218 6364557
2009 34022650 8230830
Source: 2003-2009 7up Nig Plc statement of income and expenditure
bulletin,
Hypothesis one
Ho- Strategic promotional mix does not infuence sale
turnover rate positively in the face of full competition.
H1- Strategic promotional mix infuence sale turnover
rate positively in the face of full competition.
For this hypothesis the six years fnancial report of
7up was used and 45% of selling expenses was taken as
promotional mix.
Years Turnover
Selling and
Distribution
Expenses
% taken as
promotional
expenses
2004 14937371 2931311 1319089.95
2005 17346662 2207731 993478.95
2006 22071731 4557690 2140960.5
2007 27309123 5576791 2509555.95
2008 30572218 6364557 2864050.65
2009 34022650 8230830 3212302.50
Source: Data analysis 2010.
Model B Std. Error T Sig. t R R
2
F
C
o
n
s
t
a
n
t
1.2E +08 8.6E + 07 1.416 .230
.173
a
.030 .123
S
t
r
a
t
e
g
i
c
P
r
o
m
o
t
i
o
n
a
l
m
i
x
-6.752 19.268
-.350 .744
Source: Data analysis 2010
Regression analysis showing the strategic infuence of
promotional mix on sales turnover.
?=f(?)
? = a + b? + ui
=0.000000012+6.752 ? + ui
Std. error = (0.00000086) (19.268)
t =(1.416)(.350)
The estimated regression model is given as:
The result of the regression analysis shows that the
regression coeffcient (R) is .173
a
. It implies that there is
positive and strong infuence between strategic promotional
mix and sales turnover in 7up manufacturing company. The
result also reveals that the coeffcient of determinant (R
2
) is
.030, it connotes that about 3% variation in sales turnover
could be explained by promotional mix. The remaining 97%
were largely due to other variables outside the regression
model that also affects sales turnover.
Testing the effect of independent variable ( strategic
promotional mix) on dependent variable (sales turnover)
2011 349
Regression analysis showing the impact of promotional mix on
sales volume.
Model B
Std.
Error
T Sig. t R R
2
F
C
o
n
s
t
a
n
t
2.050 3.898 .526 .692
.500
a
.250 .333
S
t
r
a
t
e
g
i
c
P
r
o
m
o
t
i
o
n
a
l
m
i
x
.500 .866 .577 .667
Source: Data analysis 2010
?=f(?)
? = a + b? + ui
=2.050+.500? + ui
Std.error= (3.898)(.866)
t=(.526)(.577)
The result of the regression analysis shows that the
correlation coeffcient (R) is .500
a
. It implies that there is
positive and strong infuence between strategic promotional
the result shows that t-value is .350 at 0.05 level of
signifcant (t=.350: p
IIU Press and Research Centre Business Intelligence Journal
Business Intelligence Journal
Business Intelligence Journal
July, 2011 Vol.4 No.2
Volume 4 - Number 2 - July 2011 - Semiannual Publicaton
The Business Intelligence Journal (BIJ) is published by the Business Intelligence Service of London, UK (BIS) in collabo-
raton with the European Business School (Cambridge, UK) and the Business Management and Economics Depart-
ment at the School of Doctoral Studies of the European Union (Brussels, Belgium), as semiannual open access content
publicaton.
Editorial Note
202
Words from the Board of Editors 203
Profle of Authors Included in this Number
204
Information for Contributors
206
Articles Banking Proftability Determinants
209
James W. Scott, José Carlos Arias
The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
231
Walter Gerard Amedzro St-Hilaire
E-Banking Patronage in Nigeria: An Exploratory Study of Gender Difference
243
Asikhia, Olalekan
Foreign Direct Investment and its Effects on the Nigerian Economy
253
Omankhanlen Alex Ehimare
Artifcial Learning and Support Vector Machines: Default Risk Prediction
262
Fabio Franch
Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
273
Carlos Acosta-Calzado
Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
293
Monireh Panahi, Mohammad Pakeniat
Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
303
Akinyele, S.T.
Intuitive Managerial Decision Making in Malaysia and the United States
312
Isola Oluwabusuyi
Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants
in Nigeria
320
Olu Ojo, I. A. Irefn
Relationship between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
327
Nadia Sajjad Hafza, Syed Sohaib Shah, Khalid Zaman, Humera Jamsheed
Comparing Bankruptcy Prediction Models in Iran
335
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
Continued on back cover
© Copyright 2011: IIU Press and Research Centre A.C.
This is an open access content publication recognized by the Scholarly Publishing and Academic Resources Coalition
of the European Union (SPARC Europe) and distributed worldwide by the Directory of Open Access Journals
(DOAJ) of the Lund University Libraries (Lund, Sweden). Printed in Canada.
ISBN: 978-1-4251-8179-6
ISSN: 1918-2325
Vol. 4 No. 2
July 2011
Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors 343
Continued From Front
Cover
Adebisi Sunday A. , Babatunde Bayode O.
The Practicability of Activity-Based Costing (ABC) in the Nigerian Retail Banks 351
Ahmed Audu Maiyaki
Social Support as Mental Health Improver for Managerial Women in the Organizational Work
Environment
355
Jocelyn Sackey, Mohammed-Aminu Sanda
Case Studies Three Case Studies On Bank Nagari Padang (Indonesia) Headquarters And Main Branch 362
Heryanto
Organizational Communication as an Important Factor of Company Success: Case Study of Bosnia And
Herzegovina
390
Kenan Spaho
Short
Communications
A Short Communication on - How a Leading Power Distribution Company Effectively Tracks business
areas like Safety, Finance and Operation for Region and Business wise for evaluating their KPI’s - Using
BusinessObjects Xcelsius Dashboards.
394
Rakesh Tej Kumar Kalahasthi, Benson Hilbert
Modeling Strategies for Financial Hedging 402
José Carlos Arias
Announcements 405
2011 Business Intelligence Journal 202
Business Intelligence Journal - July, 2011 Vol.4 No.2
SPARCEurope
DIRECTORY OF
JOURNALS
OPEN ACCESS
Business Intelligence Journal by Business Intelligence Service is licensed under a Creative Commons Attribution 2.0 UK: England & Wales License.
Further tips for using the supplied HTML and RDF are here:http://creativecommons.org/learn/technology/usingmarkup
In collaboration with the European Business School, Cambridge, UK and the School
of Doctoral Studies of the European Union based in Brussels, Belgium, the Business
Intelligence Service of London, UK (BIS) publishes the Business Intelligence Journal
(BIJ) as a semi-annual scientific and academic open access journal, which includes research analysis and inquiry into issues of importance to
the international business community. Articles within the BIJ examine emerging trends and concerns, among others, in the areas of general
management, business law, public responsibility and ethics, marketing theory and applications, business finance and investment, general business
research, business and economics education, production/operations management, organizational behavior and theory, strategic management
policy, management organization, statistics and econometrics, personnel development and industrial relations, technology and innovation, case
studies and management information systems. The goal of the BIJ is to broaden the knowledge of business professionals and academicians by
including valuable insight to business-related information, by fostering cutting-edge research work and by providing open access to original
research, information and ideas contained within the journal’s pages. All articles in the BIJ are peered reviewed; the BIJ has been granted the Seal
of the Scholarly Publishing and Academic Resources Coalition of the European Union (SPARC Europe). The BIJ is an open access publication
distributed worldwide by the Directory of Open Access Journal (DOAJ) of the Lund University Libraries, Sweden for the Business Intelligence
Research Centre of London.
EDITORIAL NOTE
European Business School,
School of Graduate Studies
Cambridge, UK.
European Business School,
Singapore.
Brussels, Belgium
The Oxford Association of
Management
Ibkan Consultants
Investment Bankers
School of Doctoral Studies
(European Union) Isles Internationale
Université Brussels, Belgium.
Business Intelligence Service
London, UK
Lund University Libraries Sweden.
Seal of the Scholarly Publishing
and Academic Resources Coalition
(Granted to the Business Intelligence
Journal on the 20
th
day of August, 2008)
Editorial Board Editorial Department
Reviewers Coordinators
Editorial Design
Publisher
Associate Editors:
Dr. Katherinna Kurlenko
European Union Analogue Standards Certification
Committee, Brussels, Belgium
Prof. José Carlos Arias
Business Intelligence Research Centre of London, UK
Dr. Jünger Albinger
School of Doctoral Studies of the European Union,
Brussels, Belgium
Prof. Eric Palme
European Business School, Cambridge, UK
Dr. Zeng Lisheng
European Business School, Singapore
Dr. Mathew Bock
The Cambridge Association of Managers, Cambridge, UK
Dr. Henk Haegeman
The Oxford Association of Management, Oxford, UK
Michael Summers
Susan G. Boots
Martin A. Miller
Kenneth C. Michaels
Anita Peters
Roger Puig
Robert Miller
Pablo Gamez Olivo
Eric Veloz
IIU Press and Research Centre, A.C.
ISSN 1918 2325http://www.saycocorporativo.com/saycouk/BIJ/journals.
html
The Business Intelligence Journal is an Open Access journal
Printed in Canada
© Copyrights: IIU Press and Research Centre, A.C.
The Business Intelligence Journal (BIJ)
Published by the Business Intelligence Service, London, UK (BIS) for the Business Intelligence Research Centre
145 St. John Street, 2nd Floor,London EC1V 4PY, United Kingdom
[email protected]
Editor: Dr. Anne Surrey
Managing Editor: Dr. Paul Stenzel
Accounting and Finance: Prof. Ira Joubert
Human Resources: Prof. Beverly Lanting
Marketing: Prof. Frans Cooper
Operations and Production: Prof. Guy Le Roy
Information and Knowledge Management: Prof. Mitsuaki Uno
Leadership and Corporate Policy: Prof. Tui Unterthiner
Change, Conflict and Crisis: Prof. Takako Iwago
Mathematical and Quantitative Methods: Prof. Maria Nicklen
Microeconomics: Prof. Carlo Grunewald
Macroeconomics and Monetary Economics: Prof. Stephen L. Freinberg
International Economics: Prof. Mitchell Alvarez
Economic Thought and Methodology: Prof. Vincent Haidinger
Business Intelligence Journal
2011 Business Intelligence Journal 203
Business Intelligence Journal - July, 2011 Vol.4 No.2
The Business Intelligence Journal has come a long way since its frst issue appearance on July 2008 as it reaches the last
number of its fourth Volume with this July 2011 edition.
Because of obvious needs of higher quality and low budget, at the beginning the Business Intelligence Journal’s collaboration
were admitted basically when authors’ origin was the Business Intelligence Centre of London, the European Business School
of Cambridge or the School of Doctoral Studies of the European Union in Brussels, Belgium.
Though this certainly ensured our publication’s quality of content at a very reasonable cost on reviewing process, it also
denied the opportunity to authors emerging from many other sources around the world, which research and content quality
was at least as good as that of our originally chosen authors’ work.
With this number, the Business Intelligence Journal closes its 4th year of publishing, having received, since 2007 over 782
articles submitted from authors with intention to become admitted. Along these years, seven numbers in four volumes have
been published, containing only 60 admitted of the 782 research articles, case studies and book reviews, presented by 123
authors and coauthors from 27 different countries and fve continents.
Along these years we have also increased our printing copies production from 800 to 4,200, meant to grow physical
distribution to suit demand at 386 Universities and 108 Libraries in Europe, Asia, Africa, Oceania and the Americas.
All research articles published during this four years period in the Business Intelligence Journal derive from original
research and have been certifed at the highest quality rank for cutting edge research by the European Union Analogue
Standards Certifcation Committee (EUASC) in Brussels, being granted the Golden Seal at this certifcation process; has
also been qualifed as highest quality of scientifc and academic content by the Directory of Open Access Journals (DOAJ)
in the Lund University Libraries of Sweden, who also works as exclusive media distribution tool for its content worldwide;
and has been granted Highest Quality of Open Access Scholarly information by The Scholarly Publishing and Academic
Coalition of Europe (SPARC Europe), at the Bodleian Library in Oxford, UK, granting our Journal with the SPARC Europe
SEAL.
In time, our format has been slightly modifying to gain space and to suit our authors, reviewers and readers needs better. Its
appearance will also experience mandatory changes as we reach our ffth volume on 2012 in order to better adapt it towards
new market tendencies.
So, may this communication be useful to properly close our Business Intelligence Journal frst development stage and to
prevent our readers, distributors, reviewers and authors on its new look starting 2012, without disregard of our traditional
and strict policy of internal and peer reviewing process, also demanding of highest quality cutting edge scientifc and
academic content.
Most cordially,
The Editorial Board
Words from the Board of Editors
Business Intelligence Journal
Profile of authors included in this number
204 Business Intelligence Journal July
Business Intelligence Journal - July, 2011 Vol.4 No.2
Article 1: Banking Proftability Determinants
Author: 1 - James W. Scott – (PhD) Researcher at the Business Intelligence Research Centre of hosted at the Business Intelligence
Service, London, UK.
email: [email protected]
2 - José Carlos Arias – (PhD) In Science by the school Robert de Sorbon, Paris, France, Professor of Management Science at
the European Business School, Cambridge, UK.
email: [email protected]
Article 2: The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Author: Walter Gerard Amedzro St-Hilaire – (PhD) Researcher on Strategy and Governance of Public Organizations in HEC
Montréal, Researcher for the Research Center on the Governance of Natural Resources.
email: [email protected]
Article 3: E-Banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Author: Asikhia, Olalekan – Senior Lecturer, Department of Business Studies, College of Business and Social Sciences, Covenant
University, Ota, Canaan land.
email: [email protected]
Article 4: Foreign Direct Investment and its Effects on the Nigerian Economy
Author: Omankhanlen Alex Ehimare – Lecturer at the Department of Banking and Finance, Covenant University, Ota, Ogun State,
Nigeria.
email: [email protected]
Article 5: Artifcial Learning and Support Vector Machines: Default Risk Prediction
Author: Fabio Franch – (PhD) from West Virginia University and studied at University of Trento, where he successfully completed his
studies (BA and MA) in 2007. He was also an exchange student at the Abo Akademi in Turku, Finland. His interests are machine
learning and the dynamic analysis of economic phenomena.
e-mail: [email protected]
Article 6: Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Author: Carlos Acosta Calzado – (MBA) BS in Industrial Engineer from ITESM, MBA with focus on Finance and Economics from
NYU Stern. He is currently partner at Ibkan Consultants, Investment Banking.
e-mail: [email protected]
Article 7: Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Author: 1 - Monireh Panahi – (PhD) candidate in Strategic Management in Shahid Beheshti University. Marketing consultant; author
of articles on marketing and strategic management in Iranian journals. School of Management & Accounting, Shahid Beheshti
University, Daneshjoo Blvd., Evin,
email: [email protected]
2 - Mohammad Pakeniat – Researcher in SRRC, Sharif University of Technology. MBA from School of Management &
Economics, Sharif University of Technology. Business consultant and researcher in the felds of strategic management in
developing countries and service marketing. Shahid Rezaii Research Center, Sharif University of Technology, Azadi Avenue,
Tehran, Iran
email: [email protected]
Article 8: Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Author: Akinyele, S.T. – (PhD) School of Business, Covenant University,Ota-Nigeria.
email: [email protected]
Article 9: Intuitive Managerial Decision Making in Malaysia and the United States
Author: Isola Oluwabusuyi – (MBA, MA, DBA) School of Business and Accounting Brown Mackie College, Atlanta.
email: [email protected]
2011 Business Intelligence Journal 205
Business Intelligence Journal - July, 2011 Vol.4 No.2
In order to make contact with any of the Authors referred to above, please forward your request to: [email protected],
including BIJ’s edition (BIJ Volume 4, Number 2, July 2011), article’s and author’s names with your requirement. BIJ’s Editor will be
glad to submit your requests or inquiries before authors.
Article 10: Franchising And Organizational Performance: Empirical Investigation Of Selected Fast Food Restaurants In Nigeria
Author: 1 - Olu Ojo – Department Of Business Administration College of Management and Social Sciences Osun State University P.
M. B. 2008, Okuku Osun State, Nigeria
email: [email protected]
2 - I. A. Irefn – Technology Planning and Development Unit Obafemi Awolowo University Ile-Ife, Osun State, Nigeria.
email: [email protected]
Article 11: Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Author: 1 - Khalid Zaman – Assistant Professor, Department of Management Sciences, COMSATS Institute of Information Technology,
Abbottabad, Pakistan.
email: [email protected], [email protected]
2 - Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed – MS Scholar, Department of Management Sciences,
COMSATS Institute of Information Technology, Abbottabad, Pakistan.
Article 12: Comparing Bankruptcy Prediction Models In Iran
Author: 1 - Ali Ebrahimi Kordlar – Faculty of management, Tehran university, Tehran, Iran.
2 - Nader Nikbakht – Faculty of management, Tehran University, Tehran, Iran.
email: [email protected]
Article 13: Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors
Author: 1 - Adebisi Sunday A. – (PhD) Department of Business Administration, Faculty of Management sciences
University of Ado Ekiti, P.M.B 5363, Ekiti State Nigeria.
email: [email protected]
2 - Babatunde Bayode O. – Department of Business Administration, College of Management and Social Sciences, Osun State
University, P.M.B 2008, Okuku, Osun State, Nigeria.
email: [email protected]
Article 14: The Practicability of Activity-Based Costing (ABC) in the Nigerian Retail Banks
Author: Ahmed Audu Maiyaki – Department of Business Administration Bayero University Kano, Nigeria.
email: [email protected]
Article 15: Social Support as Mental Health Improver for Managerial Women in the Organizational Work Environment
Author: 1 - Jocelyn Sackey, MPhil – Department of Business Administration, Technology and Social Sciences
Luleå University of Technology.
2 - Mohammed-Aminu Sanda – (PhD) Division of Industrial Work Environment, Department of Business Administration,
Technology and Social Sciences, Luleå University of Technology.
email: [email protected]
Case Studies 1: Three Case Studies On Bank Nagari Padang (Indonesia) Headquarters And Main Branch
Author: Heryanto – (PhD) Chief of RfD of the Chamber of Commerce and Industry; Professor of Mangement Science in Indonesia.
email: [email protected]
Case Studies 2: Organizational Communication As An Important Factor Of Company Success: Case Study Of Bosnia And Herzegovina
Author: Kenan Spaho – Center for Techological and Economical Development. Sarajevo, Bosnia and Herzegovina.
email: [email protected]
Short Communi-
cations 1:
A Short Communication on - How a Leading Power Distribution Company Effectively Tracks Business Areas Like Safety,
Finance and Operation for Region and Business Wise for Evaluating their Kpi’s - Using Businessobjects Xcelsius Dashboards.
Author: 1 - Rakesh Tej Kumar Kalahasthi – SAP BI Practice, Bangalore, India.
email: [email protected]
2 - Benson Hilbert – SAP BI Practice, Bangalore, India.
email: [email protected]
Short Communi-
cations 2:
Modeling Strategies for Financial Hedging
Author: José Carlos Arias – (PhD) In Science by the school Robert de Sorbon, Paris, France, Professor of Management Science at the
European Business School, Cambridge, UK.
email: [email protected]
Business Intelligence Journal - July, 2011 Vol.4 No.2
206 Business Intelligence Journal July
INFORMATION FOR CONTRIBUTORS
Electronic submission of manuscripts is strongly
encouraged, provided that the text, tables, and fgures
are included in a single Microsoft Word fle (preferably
in Times New Roman, 12 size font)
Submit manuscript as e-mail attachment to the
BIJ Editorial Offce at: edit.bij@saycocorporativo.
com. A manuscript number will be mailed to the
corresponding author within the following 7 days.
The cover letter should include the corresponding
author’s full address and telephone/fax numbers and
should be in an e-mail message sent to the Editor,
with the fle, whose name should begin with the frst
author’s surname, as an attachment. The authors may
also suggest two to four reviewers for the manuscript
(BIJ may designate other reviewers).
BIJ will only accept manuscripts submitted as
e-mail attachments.
Article Types
Three types of manuscripts may be submitted:
Regular Articles: These should describe new and
carefully confrmed fndings, and research methods
should be given in suffcient detail for others to verify
the work. The length of a full paper should be the
minimum required to describe and interpret the work
clearly.
Short Communications: A Short Communication
is suitable for recording the results of complete
small investigations or giving details of new models,
innovative methods or techniques. The style of main
sections need not conform to that of full-length papers.
Short communications are 2 to 4 printed pages (about
6 to 12 manuscript pages) in length.
Reviews: Submissions of reviews and perspectives
covering topics of current interest are welcome and
encouraged. Reviews should be concise and no longer
than 4-6 printed pages (about 12 to 18 manuscript
pages). Reviews manuscripts are also peer-reviewed.
Review Process
All manuscripts are reviewed by an editor and
members of the Editorial Board or qualifed outside
reviewers. Decisions will be made as rapidly as
possible, and the journal strives to return reviewers’
comments to authors within 3 weeks. The editorial
board will re-review manuscripts that are accepted
pending revision. It is the goal of the BIJ to publish
manuscripts within the following BIJ edition after
submission.
Regular Articles
All portions of the manuscript must be typed
double-spaced and all pages numbered starting from
the title page.
The Title should be a brief phrase describing the
contents of the paper. The Title Page should include
the authors’ full names and affliations, the name of
the corresponding author along with phone, fax and
e-mail information. Present addresses of authors
should appear as a footnote.
The Abstract should be informative and completely
self-explanatory, briefy present the topic, state the
scope of the work, indicate signifcant data, and point
out major fndings and conclusions. The Abstract
should be 100 to 200 words in length. Complete
sentences, active verbs, and the third person should
be used, and the abstract should be written in the past
tense. Standard nomenclature
Should be used and abbreviations should be
avoided. No literature should be cited.
Following the abstract, about 3 to 10 key words that
will provide indexing references to should be listed.
A list of non-standard Abbreviations should be
added. In general, non-standard abbreviations should
be used only when the full term is very long and used
often. Each abbreviation should be spelled out and
introduced in parentheses the frst time it is used in
the text.
The Introduction should provide a clear statement
of the problem, the relevant literature on the subject,
and the proposed approach or solution. It should be
understandable to colleagues from a broad range of
disciplines.
Materials and methods should be complete enough
to allow possible replication of the research. However,
only truly new research methods should be described in
Business Intelligence Journal - July, 2011 Vol.4 No.2
2011 Business Intelligence Journal 207
detail; previously published methods should be cited,
and important modifcations of published methods
should be mentioned briefy. Capitalize trade names
and include the manufacturer’s name and address.
Subheadings should be used. Methods in general use
need not be described in detail.
Results should be presented with clarity and
precision. The results should be written in the past
tense when describing author’s fndings. Previously
published fndings should be written in the present
tense. Results should be explained, but largely without
referring to the literature. Discussion, speculation and
detailed interpretation of data should not be included
in the Results but should be put into the Discussion
section.
The Discussion should interpret the fndings in view
of the results obtained in this and in past studies on the
topic. State the conclusions in a few sentences at the
end of the paper. The Results and Discussion sections
can include subheadings, and when appropriate, both
sections can be combined.
The Acknowledgments of people, grants, funds,
etc should be brief.
Tables should be kept to a minimum and be
designed to be as simple as possible. Tables are to be
typed double-spaced throughout, including headings
and footnotes. Each table should be on a separate
page, numbered consecutively in Arabic numerals and
supplied with a heading and a legend. Tables should
be self-explanatory without reference to the text. The
details of the research methods should preferably be
described in the legend instead of in the text. The same
data should not be presented in both table and graph
form or repeated in the text.
Figure legends should be typed in numerical order
on a separate sheet. Graphics should be prepared using
applications capable of generating high resolution
GIF, TIFF, JPEG or PowerPoint before pasting in
the Microsoft Word manuscript fle. Tables should
be prepared in Microsoft Word. Use Arabic numerals
to designate fgures and upper case letters for their
parts (Figure 1). Begin each legend with a title and
include suffcient description so that the fgure
is understandable without reading the text of the
manuscript. Information given in legends should not
be repeated in the text.
References: In the text, a reference identifed by
means of an author’s name should be followed by the
date of the reference in parentheses. When there are
more than two authors, only the frst author’s name
should be mentioned, followed by ‘et al’. In the
event that an author cited has had two or more works
published during the same year, the reference, both in
the text and in the reference list, should be identifed
by a lower case letter like ‘a’ and ‘b’ after the date to
distinguish the works.
Examples:
Smith (2000), Wang et al. (2003), (Kelebeni, 1983),
(Usman and Smith, 1992), (Chege, 1998; Chukwura,
1987a,b; Tijani, 1993, 1995), (Kumasi et al., 2001)
References should be listed at the end of the paper
in alphabetical order. Articles in preparation or articles
submitted for publication, unpublished observations,
personal communications, etc. should not be included
in the reference list but should only be mentioned in
the article text (e.g., A. Kingori, University of Nairobi,
Kenya, personal communication). Journal names are
abbreviated according to Chemical Abstracts. Authors
are fully responsible for the accuracy of the references.
Examples:
Papadogonas TA (2007). The fnancial performance of
large and small frms: evidence from Greece. Int. J.
Financ. Serv. Manage. 2(1/2): 14 – 20.
Mihiotis AN, Konidaris NF (2007). Internal auditing:
an essential tool for adding value and improving
the operations of fnancial institutions and
organizations. Int. J. Financ. Serv. Manage. 2(1/2):
75 – 81.
Gurau C (2006). Multi-channel banking in Romania:
a comparative study of the strategic approach
adopted by domestic and foreign banks Afr. J.
Financ. Servic. Manage. 1(4): 381 – 399.
Yoon CY,Leem CS (2004).Development of an
evaluation system of personal e-business
competency and maturity levels Int. J. Electron.
Bus. 2(4): 404 – 437.
Business Intelligence Journal - July, 2011 Vol.4 No.2
208 Business Intelligence Journal July
Short Communications
Short Communications are limited to a maximum
of two fgures and one table. They should present
a complete study that is more limited in scope
than is found in full-length papers. The items of
manuscript preparation listed above apply to Short
Communications with the following differences: (1)
Abstracts are limited to 100 words; (2) instead of a
separate Materials and Methods section, research
methods may be incorporated into Figure Legends and
Table footnotes; (3) Results and Discussion should be
combined into a single section.
Proofs and Reprints
Electronic proofs will be sent (e-mail attachment)
to the corresponding author as a PDF fle. Page proofs
are considered to be the fnal version of the manuscript.
With the exception of typographical or minor clerical
errors, no changes will be made in the manuscript at
the proof stage. Because BIJ will be published online
without access restrictions, authors will have electronic
access to the full text (PDF) of the article.
Authors can download the PDF fle from which
they can print unlimited copies of their articles.
Copyright
Submission of a manuscript implies: that the work
described has not been published before (except in the
form of an abstract or as part of a published lecture, or
thesis) that it is not under consideration for publication
elsewhere; that if and when the manuscript is accepted
for publication, the authors agree to automatic transfer
of the copyright to the publisher.
Costs for Authors
Revision, edition and publishing costs will be
totally paid by the IIU Press & Research Centre A.C.
and authors’ sole contribution will be providing BIS
with their invaluable work.
Publication Decisions
Decisions by the editor of all submitted manuscripts
refect the recommendations of members of the
Editorial Board and other qualifed reviewers using
a “blind” review process. Reviewers’ comments
are made available to authors. Manuscripts that
are inappropriate or insuffciently developed may
be returned to the authors without formal review
for submission to a more suitable journal or for
resubmission to BIJ following further development.
Manuscripts submitted will be judged primarily
on their substantive content, though writing style,
structure and length will also be considered. Poor
presentation is suffcient reason for the rejection of
a manuscript. Manuscripts should also be written as
concisely and simply as possible, without sacrifcing
clarity or meaningfulness of exposition. Manuscripts
will be evaluated by the editor when frst received,
on their contribution-to-length-ratio, meaning that
manuscripts with strong contributions will be assigned
more pages than those making narrower contributions.
Papers intended to make very extensive contributions
(over 35 double-space pages, using one inch margins
and Times New Roman twelve-pitch font) will, at
discretion of the editor, be allotted additional space.
Authors are expected to get and use feedback from
colleagues prior to submitting a manuscript for formal
review.
2011 209
BANKING PROFITABILITY DETERMINANTS
James W. Scott (PhD)
Researcher at the Business Intelligence Research Centre of hosted at the Business Intelligence Service, London, UK
2nd Floor, 147 St. John Street, London, UK
Email: [email protected]
José Carlos Arias (PhD, DBA)
Professor of Management Science at the European Business School, Cambridge, UK,
2nd Floor, 147 St. John Street, London, UK
Email: [email protected]
Abstract
Notwithsanding the enormously complex and dynamic nature of the environment in which they compete, there is a growing body
of evidence that suggests it is possible to discern relevant indicators of profitability for the banking industry today. The purpose of this
study was to develop an appropriate econometric model whereby the primary determinants of profitability of the top five bank holding
companies in the United States could be examined and understood. To accomplish this purpose, an econometric model based on internal
aspects of the banking organizations as they related to their return on assets and external aspects of the environment in which they
compete as measured by growth in GDP was developed based on guidance provided by economists and industry experts to determine
the impact of the external national economy of these five leading banks according to their size as measured by total assets. A critical
review of the relevant peer-reviewed, scholarly and organizational literature is followed by an analysis of the statistical data for these
bank holding companies using the econometric model. A summary of the research and salient conclusions are provided in the concluding
chapter. Key words: Finance, Economics, Banking, Profits, Profitability Measurements.
The banking industry in general has experienced some
profound changes in recent decades, as innovations in
technology and the inexorable forces driving globalization
continue to create both opportunities for growth and
challenges for banking managers to remain proftable in
this increasingly competitive environment. Most of the
studies concerning bank proftability to date, including
Short (1979), Bourke (1989), Molyneux and Thornton
(1992), Demirguc-Kunt and Huizinga (2000) and Goddard,
Molyneux, and Wilson (2004b), have employed different
linear models to estimate the impact of various factors
that could be signifcant in terms of explaining profts.
According to Athanasoglou and his colleagues (2005),
these studies were seminal in demonstrating the feasibility
of conducting a meaningful analysis of the determinants
of bank proftability, but some of the methods used by
these studies failed to take into account the robust and
dynamic nature of the economic environment in which they
competed. Moreover, the studies to date have primarily
considered determinants of proftability at the bank and/
or industry level, with the choice of variables used lacking
internal consistency in some instances; in addition, there
has been a dearth of research concerning the potential
infuence of the macroeconomic environment, due in
part to the small time dimension of the panels used in the
estimation (Athanasoglou et al., 2005).
Other factors that have constrained the research to date
include the fact that the econometric methodology used
in the study was not adequately described and/or failed to
account for some features of bank profts which suggest
that the estimates obtained by these studies may have
been biased or inconsistent (Athanasoglou et al., 2005).
Nevertheless, studies have amply demonstrated the ability
of models to better understand what fuels proftability in
fnancial services organizations, and this represents the
purpose of the study which is described further below.
The general purpose of this study was to develop an
appropriate econometric model whereby the primary
determinants of proftability of the top fve bank holding
companies in the United States could be examined and
understood. The specifc purpose of this study was to
determine the impact of the external environment in which
the fve leading bank holding companies in the United
States (as of June 2007) as measured by real growth in GDP
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
210 Business Intelligence Journal July
compared to internal aspects of the respective organizations
as measured by their return on assets as an indicator of
proftability.
This study reviewed the peer-reviewed, scholarly
and organizational literature as it applied to the banking
industry in general and to the United States in particular
to identify recent trends, determinants of proftability for
fnancial services organizations, and to provide the requisite
background to developing a straight-forward econometric
model that could be used to analyze bank-level data for the
nation’s fve leading bank holding companies as measured
by total assets.
According to Goddard and his colleagues (2004a), in
spite of the growing body of research into determinants
of banking proftability, there remains a paucity of studies
that have investigated the specifc relationship between
organizational size and its impact on proftability. These
authors report that, “Previous studies of the dynamics of
growth on the one hand, and proft on the other, have in
the main developed separately, and followed contrasting
empirical methodologies. Nevertheless, there are several
theoretical arguments to suggest that these two performance
indicators are closely related. [However], few researchers
have tested for empirical relationships between growth and
proft directly” (Goddard et al., 2004a p. 1069). Moreover,
as Cover (1999) emphasizes, the need for identifying
determinants of proftability in the banking industry has
never been greater: “As banks move into the twenty-frst
century, they must focus more than ever before on creating
new streams of revenue in order to shareholder value.
Crucial to this effort is the need to assess and analyze the
proftability of the bank’s current customers, relationships,
services, and products. It is only through such analyses that
banks can determine which customers to fght for, which
customer relationships to expand, and which prospective
customers to pursue” (p. 78).
The rationale that guided this study was that the larger
the bank holding company, the more impervious it would
be to downturns in the external environment as measured
by real growth in GDP. Larger fnancial services companies
enjoy a wide range of advantage by virtue of economies of
scale, accumulated tacit knowledge and expertise as well
as the resources needed to continue their expansions into
foreign markets where barriers to entry may be prohibitive
for smaller players in the market. For example, according to
Goddard, Molyneux and Wilson (2004), “Previous studies
of the determinants of concentration have proposed various
explanations as to why some frms grow and attain large
size. These include economies of scale or scope, effciency
gains attained through size, the adoption of entry-deterring
strategies, or the exercise of other forms of market power”
(p. 1069). Likewise, as noted above, a number of studies
have demonstrated the feasibility of using bank-level data
to investigate proftability determinants. In addition, the
research to date also suggests that banking frms are one of
the best places to test for the effects of various proftability
factors such as technological innovation. In this regard,
Berger and Deyoung (2006) emphasize that, “Banks
have embraced substantial advances in both physical and
fnancial technologies during the past two decades, and
the broader industry category of which banking is a part,
Depository and Nondepository Financial Institutions, is
the most information technology-intensive industry in the
United States” (p. 1483).
This study used a fve-chapter format to achieve the
above-stated research goals. The frst chapter introduced
the topics under consideration and provided a statement
of the importance of the study, its importance and scope,
as well as its supporting rationale. The second chapter of
the study consists of a critical review of the relevant peer-
reviewed, scholarly and organizational literature. The third
chapter more fully describes the econometric methodology
used to analyze the proftability of the top fve banks in the
United States today, and chapter four presents the analysis
of the data. Finally, chapter fve provides a summary of the
research, salient conclusions and recommendations.
Review of Related Literature
Background and Overview.
On the one hand, the banking industry today enjoys a
number of advantages compared to past years that would
appear to contribute to their ability to generate profts.
According to Berger and Deyoung (2006), the banking
industry in the United States has been in a constant process
of geographic expansion in recent years, both within
nations and across nations. These authors report that, “At
one time, nearly all customers were served by locally
based institutions. In contrast, it is now much more likely
that the bank or branch providing services is owned by an
organization headquartered a substantial distance away,
perhaps in another state, region, or nation” (p. 1483).
As an example, these authors note that between 1985
and 1998 the distance between the largest bank and the
other affliate banks in U.S. multibank holding companies
(MBHCs) increased by over 50 percent on average, from
123.35 to 188.91 miles, as a number of MBHCs acquired
2011 211
banks in other states and regions (Berger & Deyoung, 2006).
Moreover, the banking industry, like any other industry, will
experience potential diseconomies to geographic expansion
in the form of agency costs associated with monitoring
junior managers in a distant locale; however, innovations
in information processing and telecommunications may
lessen these agency costs by improving the ability of senior
managers located at the organization’s headquarters to
monitor and communicate with staff at distant subsidiaries
(Berger & Deyoung, 2006).
In the modern banking industry, technologies such as
ATM networks and transactional Internet websites allow
banks to interact more effciently with their customers
regardless of geographic proximity; furthermore, recent
innovations in fnancial technologies provide the capacity
to provide these services using long-distance interfaces with
customers. According to Berger and Deyoung, “Greater
use of quantitative methods in applied fnance, such as
credit scoring, may allow banks to extend credit without
geographic proximity to the borrower by ‘hardening’ their
credit information” (p. 1483). Likewise, new product mixes
of fnancial engineering, such as derivative contracts, may
provide banks of all sizes to unbundle, repackage, or
hedge risks at lower costs without regard to the geographic
proximity to the other party (Berger and Deyoung, 2006).
These fnancial innovations may also provide senior banking
managers with the ability monitor the decisions made by loan
offcers and managers at distant affliate banks more easily,
and to evaluate and manage the contributions of individual
affliate banks to the organization’s overall returns and risk
more effciently as well (Berger & Deyoung, 2006).
On the other hand, these same trends have introduced
some additional constraints on the banking industry as
they seek to maintain their existing market share and grow
their companies along multinational lines. In this regard,
Saunders and Walter (1998) report that, “Financial systems
that are deemed ineffcient or incomplete are characterized
by a limited range of fnancial services and obsolescent
fnancial processes. Both static and dynamic effciency are
of obvious importance from the standpoint of national and
global resource allocation, not only within the fnancial
services industry itself but also as it effects users of
fnancial services” (p. 19). In other words, because fnancial
services can be regarded as being “inputs” to the overall
production process of a country, the level of national output
and income -- as well as its rate of economic growth -- are
directly affected by the effciency characteristics of the
fnancial services sector (Saunders and Walter, 1998).
A “retarded” fnancial services industry, in this sense,
can represent a major impediment to a nation’s overall
real economic performance. Such retardation represents
a burden on the fnal consumers of fnancial services and
potentially reduces the level of private and social welfare. It
also represents a burden on producers, by raising their cost
structures and diminishing their competitive performance in
domestic and global markets (Saunders and Walter, 1998).
Therefore, any such retarded fnancial services industry
skews the trends involved in the resource allocation in the
national economy.
Financial system ineffciencies can be traced to a number
of factors:
1. Regulations that prevent fnancial frms from complete
access to alternative sources of funding or the full range
of borrowers and issuers;
2. Taxation imposed at various stages of the fnancial
intermediation process, including securities transfer
taxes, transactions taxes, etc.;
3. Lack of competition that reduces incentives to cut
intermediation costs and promote innovation;
4. Lack of market discipline imposed on owners and
managers of fnancial intermediaries, leading to poor
risk management, agency problems, and increased
costs; and,
5. Existence of major information imperfections among
contracting parties (Saunders and Walter, 1998).
National fnancial systems that are statically and/
or dynamically ineffcient tend to be disintermediated.
Borrowers or issuers in a position to do so seek foreign
markets or offshore markets that offer lower costs or a
more suitable range of products. Investors likewise seek
markets abroad that offer higher rates of return or improved
opportunities to construct more effcient portfolios.
These types of systems can be described as being
either “uncompetitive” or “unattractive” as venues for
fnancial intermediation in the context of global markets;
nevertheless, individual fnancial services institutions may
be able to cross-subsidize foreign activities from abnormal
profts earned domestically for a period of time (Saunders
and Walter, 1998).
The delivery of fnancial services can be conceptualized
as a market structure that combines three principal
dimensions in the delivery of fnancial services in terms of
the clients served (C), the geographic arenas where business
is done (A), and the products supplied (P) (Saunders and
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
212 Business Intelligence Journal July
Figure ___. The Client-Arena-Product (C-A-P) Matrix.
Source: Saunders and Walters, 1998 at p. 20.
The inherent attractiveness of each cell to the suppliers
of fnancial services is inextricably related to the size of
the prospective risk-adjusted returns that can be extracted
from it; in addition, the durability of returns generated will
depend on the ability of new players to enter the cell and the
development of substitute products over time (Saunders and
Walters, 1998). Based in large part on recent innovations in
technology and fnancial deregulation, competitors in the
fnancial services industry are confronted with expanded
potential access to each dimension of the C-A-P opportunity
set (Saunders and Walters, 1998)
To achieve maximum proftability, these authors
recommend that fnancial services companies allocate their
available resources to those C-A-P cells in Figure 1 that
promise to provide them with the highest risk-adjusted
returns. According to these authors, though, “In order to do
this, they will also have to allocate capital, costs, returns, and
risks appropriately across cells. Beyond this, however, the
economics of supplying fnancial services internationally
is jointly subject to economies of scale and economies
of scope. The existence of both types of economies have
strategic implications for players in the industry; indeed,
economies of scale suggest an emphasis on deepening the
activities of individual frms within a cell, or across cells
in the product dimension” (emphasis added) (Saunders and
Walter, 1998 p. 21).
Moreover, economies of scope indicate there has been
an effort made to broaden fnancial services activities
across cells; in other words, a bank can produce a given
level of output in a given cell more inexpensively or market
it more effectively than institutions that are less active
across such multiple cells; however, this depends on the
benefts and costs of linking cells together in a coherent
fashion (Saunders and Walter, 1998).
In this regard, regulation of fnancial services can be
viewed as having an important infuence in terms of:
1. Accessibility of geographical arenas;
2. Accessibility of individual client groups by players
originating in different sectors of the fnancial services
business; and,
3. Substitutability among fnancial products in meeting
personal, corporate, or government fnancial needs
(Saunders and Walter, 1998).
Financial intermediaries are clearly sensitive to
incremental competition in C-A-P cells as illustrated in
Figure __ above, particularly where economic entry barriers
are limited (Saunders and Walter, 1998). Furthermore,
“Market penetration by competitors can erode indigenous
players' returns and raises protectionist motivations.
Given the economic interests involved, banks and other
fnancial institutions are in an excellent position to convert
them into political power in order to achieve protection
against potential rivals. They are often exceedingly well
connected politically, and their lobbying power motivated
by protectionist drives can be awesome” (Saunders and
Walter, 1998 p. 21).
Competitive distortions in the fnancial services industry
take the form of entry barriers and operating restrictions. In
terms of Figure 1 above, entry barriers tend to restrict the
movement of fnancial services frms in the lateral "arenas"
dimension of the matrix; a fnancial services organization
that is locked out of a particular national market does not
enjoy the same level of lateral opportunity set that excludes
the relevant segment of "client" and "product" cells
(Saunders and Walter, 1998).
Assuming that a fnancial services organization has
successfully gained access to a particular arena, there are
a number of operating restrictions that carry the potential
to constrain either the depth of service it can supply
to a particular cell (e.g., lending limits, staffng limits,
restrictions on physical location) or in the feasible set of
cells within the tranche (e.g., limits on services banks or
securities frms are allowed to supply and the client groups
Walter, 1998). In this regard, Figure 1 below illustrates
these dimensions in the form of a matrix of C X A X P cells
wherein individual cell characteristics can be analyzed in
terms of conventional competitive structure criteria.
2011 213
Table 1.Banking Structure in the United States.
Year
No. of
Banks
Total Assets
($ Billion)
No. of Banks
w/Assets>$10
Billion
% of Assets Held
by Large Banks
2000 8,315 $6.2 82 70%
1980 14,769 $1.9 18 34%
Source: FDIC cited in Gup, 2003 at p. 2.
Figure 2. No. of Banks in the United States: 2000 versus 1980
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
No. of Banks
2000 1980
Source: Based on tabular data cited in Gup, 2003 at p. 2.
$0.00
$1.00
$2.00
$3.00
$4.00
$5.00
$6.00
$7.00
Total Assets ($ billion)
2000 1980
Figure 3. Total Assets of Banks in the United States (in $
billion): 2000 versus 1980.
Source: Based on tabular data cited in Gup, 2003 at p. 2.
they are allowed to serve). Operating restrictions in turn
can be subclassifed in terms of whether they place limits
on the kinds of fnancial services that may be sold locally
(Type A) or the specifc client-groups that may be served
(Type B). According to these authors, “Operating limits
may severely reduce proftability associated with the arena
concerned, while creating signifcant excess returns for the
protected industry; regulators may also tolerate a certain
amount of anticompetitive, cartel-like behavior on the part
of domestic fnancial institutions” (Saunders and Walter,
1998 p. 21).
Economies of scope and scale may be signifcantly
constrained by entry and operating restrictions in a
particular market, indicating the importance of the impact
of competitive distortions on horizontal integration in
the fnancial services industry. Under universal banking
structures the entire C-A-P matrix is available to fnancial
institutions in terms of their competitive positioning and
execution, providing the potential -- absent anticompetitive
behavior -- for maximum static and dynamic effciency
in the fnancial system. Restrictions on entry by "ft and
proper" players, regardless of whether they are domestic
or international, as well as limits on the business that may
be done and the clients that may be served, represent a
signifcant threat for eroding the domestic and international
performance of fnancial institutions and of the national
economy as a whole (Sanders and Walters, 1998 p. 22).
As noted above, consolidation by acquisition or
otherwise in the banking industry in the United States from
1980 to 2000 is one of the most signifcant changes to affect
the industry in recent years (Gup, 2003). In their recent
study, “Competition from Large, Multimarket Firms and the
Performance of Small, Single-Market Firms: Evidence from
the Banking Industry,” Berger and his colleagues (2007)
report that, “Over the last two decades, retail banking in
the U.S. has changed dramatically. Large banks that branch
across multiple local markets have signifcantly increased
their share of local markets, and small banks that operate
in a single market have experienced substantially reduced
local market shares” (p. 331). In 1982, large, multimarket
banks--banks with gross total assets (GTA) over $1 billion
(real 1994 dollars) with branch offces in more than one
local market held 23 percent of local deposits in U.S.
metropolitan markets; however, by 2000, this fgures had
almost tripled to 65 percent (Berger et al., 2007). During this
same time period, shares in these markets of small, single-
market banks (GTA [less than or equal to] $1 billion, offces
in only one market) decreased by approximately two-thirds
from 60 percent to 19 percent (Berger et al., 2007).
As shown in Table 1 below, the number of banks
decreased by more than 6,000, while the percentage of total
assets held by the largest banks doubled; as a result, 82 large
banks representing less than 1 percent of the total number
of banks now hold more than two-thirds of all bank assets:
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
214 Business Intelligence Journal July
Figure 4. No. of Banks in the United States with Assets Greater
than $10 billion: 2000 versus 1980.
0
20
40
60
80
100
No. of Banks w/Assets > $10 Billion
2000 1980
Source: Based on tabular data cited in Gup, 2003 at p. 2.
Figure 5. Percentage of Assets Held by Large Banks in the
United States: 2000 versus 1980.
0%
10%
20%
30%
40%
50%
60%
70%
% of Assets Held by Large Banks
2000 1980
Source: Based on tabular data cited in Gup, 2003 at p. 2.
According to Toyne and Tripp (1998), reductions
in regulatory restrictions on interstate bank mergers
and increased proftability in the banking industry have
resulted in higher levels of merger and acquisition activity
in the United States in recent years. The trend towards
consolidation in the banking industry, though, is certainly
not unique to the United States. For example, a study by the
central banks of the Group of Ten Nations (G-10) identifed
a high level of concentration in all of the 13 countries they
investigated (Gup, 2003). These consolidations have been
the result of the following:
1. Deregulation of geographic markets;
2. Shifts in fnancial technology (e.g., securitization and
derivatives);
3. Innovations in communications technology;
4. Changes in information technology;
5. The desire to achieve economies of scale;
6. High stock prices used as currency in mergers, and,
7. Other factors (Gup, 2003).
In general, the consolidation involves large banks
within one country, cross-border bank consolidations, and
consolidations between banks and other types of fnancial
institutions (Gup, 2003). Some of the “other factors”
identifed by Gup (2003) above include the need to respond
to new competition from smaller bank that are able to
market higher proft products more effciently than their
larger counterparts. These and others determinants of bank
performance are discussed further below.
Determinants of Bank Performance.
While virtually all of the studies reviewed herein
emphasize the need for yet more studies, there has been
a growing body of evidence concerning the ability of
researchers to identify accurate determinants of bank
performance in recent years. Most of the studies on the
determinants of bank’s interest margin and proftability
have focused whether on a particular country (Berger,
1995; Guru et al., 2002; Barajas et al., 2001; Ben Naceur
and Goaied, 2001) or on a panel of countries (Abreu and
Mendes, 2002; Demerguç-Kunt and Huizingha, 1999). The
empirical evidence for the United States has largely been
compiled by Berger (1995), Neeley and Wheelock (1997)
and Angbazo (1997) (all cited in Ben Naceur, 2003).
One study by Berger (1995) investigated the relationship
between the return on equity and the capital asset ratio for
a sample of banks in the United States for the period 1983-
1992; using the Granger causality model, this researcher
determined that the return of equity and capital to asset
ratio tend to be positively related. Similarly, Neeley and
Wheelock (1997) studied the proftability of a sample of
insured commercial banks in the US for the 1980-1995
period. These researchers determined that bank performance
is positively related to the annual percentage changes in
the state’s per capita income. Likewise, Anghazo (1997)
studied the determinants of bank net interest margins for a
representative sample of banks in the United States for the
period 1989-2003; the fndings of that study showed that
default risk, the opportunity cost of non-interest bearing
reserves, leverage and management effciency are also
2011 215
Author/Date/Title/
Publication
Key Findings Comments
Spindler, J. Andrew, Jonathan T. B. Howe, and
David F. Dedyo. (1990, May). “The Performance
of Internationally Active Banks and Securities
Firms Based on Conventional Competitiveness
Measures”. Federal Reserve Bank of New York,
Mimeo.
In terms of size, U.S. banks fell consistently throughout the
period, especially against their Japanese competitors. The same
approximate pattern emerged with respect to revenue growth.
(The authors note that while a signifcant component of this
“size efect” is attributable to the decline in the value of the dollar
against the yen and other major currencies over the period.) They
performed slightly below the sample mean on ROA, well below
their Swiss, British, and Japanese counterparts. The same was true
of ROE in comparison with Japanese and French banks. In terms
of productivity, U.S. banks fell into the middle of the range. On
capitalization, U.S. banks were high in the rankings on the frst
measure cited above, but well behind Japanese, German, and
Swiss banks on the second.
The authors report the results of a study by the Federal Reserve
Bank of New York attempted to assess the performance of 51
banks and securities frms based in various countries (the
United States, Canada, France, Germany, Japan, Switzerland,
and the United Kingdom) that were internationally active
during the second half of the 1980s. The performance
measures used include frm size (total assets and total revenue,
in real terms); proftability (return on assets [ROA] and return
on equity [ROE]); productivity (ratio of total revenue to total
non-interest expense); and capitalization (as measured by
the ratio of shareholder equity to total assets, and the ratio of
market capitalization to reported earnings).
Goddard, John, Phil Molyneux and John O.S.
Wilson. (2004). “Dynamics of Growth and
Proftability in Banking.” Journal of Money, Credit
& Banking, 36(6), 1069.
Empirical research concerning the dynamics of company
proftability is based on an account of the determinants of proft
that is an alternative to the essentially static Structure-Conduct-
Performance (SCP) paradigm; however, although the relevant
microtheory identifes SCP relationships applicable when markets
are in equilibrium, there is no certainty that a proft fgure observed
at any point in time represents an equilibrium value.
The hypotheses tested in the persistence of proft (POP)
literature are that entry and exit are sufciently free to
eliminate any abnormal proft quickly, and that all frms’ proft
rates tend to converge to the same long-run average value.
The alternative is that some frms possess special knowledge
or other advantages enabling them to prevent imitation or
block entry. If so abnormal proft may tend to persist from
year to year, and diferences in average proft rates may be
sustained indefnitely. Empirical tests of the POP hypothesis
in banking are few in number; however, recent studies have
presented extensive evidence of POP in U.S. banking.
Stiroh, Kevin J. (2004). “Diversifcation in Banking:
Is Noninterest Income the Answer?” Journal of
Money, Credit & Banking, 36(5), 853.
These researchers conclude that the banking industry is the
U.S. is steadily shifting away from traditional sources of revenue
such as loan making and toward activities that generate fee
income, service charges, trading revenue, and other types of
noninterest income; while noninterest income has always played
an important role in banking revenue, there is a clear trend that it
is gaining importance. By 2001, noninterest income represented
43% of net operating revenue (net interest income plus
noninterest income), an increase from just 25% in 1984. Also note
that, “banks with higher noninterest income shares have lower
proftability per unit of risk” (p. 853).
The shift toward noninterest income has contributed to
higher levels of bank revenue in recent years; there is also an
indication that it will lower the volatility of bank proft and
revenue and reduce risk.
Kaushik, Surendra K. and Raymond H. Lopez.
(1996) “Proftability of Credit Unions, Commercial
Banks and Savings Banks: A Comparative Analysis.”
American Economist, 40(1), 66.
Commercial banks and credit unions experienced growth in
assets faster than growth in loans during the 1990s; as a result,
their investment portfolios have been increasing in absolute and
relative size. “This tends to hold down proftability since margins
on loans are greater than those on investments” (p. 67).
Authors also determined that credit union loan portfolios have
grown more rapidly than either commercial banks’ or savings
institutions’ in recent years, and their net interest margins have
consistently been above banks.
Prendergast, C. (1993, March). “The Provision
of Incentives in Firms.” Journal of Economic
Literature, 7, 63.
Studies have shown that some incentive structures may result
in dysfunctional behavior; this may occur more frequently when
incentives within regulated fnancial services companies relate to
volume and create a clear bias towards writing business.
Bank managers may be rewarded by the volume of loans rather
than their risk-adjusted proftability; numerous instances of
bank distress have been related to inappropriate incentive
structures that created a bias in favor of the bank’s balance
sheet growth.
Grigorian, David A. and Vlad Manole. (2006).
“Determinants of Commercial Bank Performance
in Transition: An Application of Data Envelopment
Analysis.” Comparative Economic Studies, 48(3),
497.
One of the methods used in the literature to evaluate productivity
and performance of banks is the data envelopment analysis (DEA),
a non-parametric method that provides a framework in which it
is possible to account for a wide range of functions performed by
the banks.
This method compares relative performance of decision-
making units (DMU) (i.e., banks) by constructing a frontier
comprised of the most efcient DMUs and focusing on how
close other DMUs are to this frontier.
Hasan, Iftekhar and William C. Hunter. (1996).
“Management Efciency in Minority- and Women-
Owned Banks.” Economic Perspectives, 20(2), 20.
Studies comparing the economic performance of smaller minority-
and nonminority-owned banks have generally shown that the
minority-owned banks have tended to be smaller, somewhat less
proftable, and more expenditure prone than comparable groups
of nonminority banks.
Previous studies reported that smaller minority-owned banks
tended to operate with lower ratios of equity capital to assets,
to employ more conservative asset portfolio management
policies, and to post higher loan losses than their nonminority
peers.
Li, Shaomin, Yigang Pan and David K. Tse. (1999).
“The Impact of Order and Mode of Market Entry
on Proftability and Market Share.” Journal of
International Business Studies, 30(1), 81.
Large frms have more resources to invest in innovation, pursue
more aggressive expansion strategies, and perform better; in
addition, large frms beneft from economies of scale, scope, and
learning. “In short, large frms tend to perform better holding
other factors constant” (p. 81).
Financial services’ proftability and market share performance
are determined by a number of diferent factors and the size of
frms has long been of interest to business researchers.
Table 2.Recapitulation of Proftability Determinants in the Banking Industry.
all positively associated with bank interest spread (Ben
Naceur, 2003).
Table 2 below provides a recapitulation of comparable
studies of proftability indicators in the banking industry in
general, and in the United States in particular.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
216 Business Intelligence Journal July
Author/Date/Title/
Publication
Key Findings Comments
Cover, Jerry. (1999). “Proftability Analysis - A
Necessary Tool for Success in the 21st Century.”
ABA Banking Journal, 91(2), 78.
“Pareto’s Curve,” or the so-called 80/20 rule, holds that 80
percent of all business activity results from 20 percent of current
customers; however several recent studies reveal that in the
banking business, the ratio is even more extreme. One study found
that 15 percent of a bank’s customer base is responsible for 85
percent of its proftability. In the small business banking industry,
the ratio is even more pronounced with fewer than 10 percent of
a bank’s relationships produce 90 percent of its profts. In a typical
retail portfolio, 20 percent of accounts contribute profts equaling
200 percent of the overall return, while up to half of the accounts
generate losses.
“The real winners will be those banks who take advantage of
this window of opportunity. Clearly, it is a window, because
other banks and non-bank competitors are focusing on those
same customers” (p. 78).
Frieder, Larry A. (1991). “Determinants of Bank
Acquisition Premiums: Issues and Evidence.”
Contemporary Policy Issues, 9(2), 14.
This study found that both the internal factors of returns on equity
(ROE) of the target organization and the external factor of the
market growth rate of the target’s state (as measured by deposit
growth and projected population growth) were signifcant positive
determinants of a bank’s capacity for expansion and growth.
It is also possible to measure a fnancial services company’s
proftability by using either return on assets (ROA), ROE, the
ratio of operating earnings to assets, or the net interest spread
to assets.
Berger, Allen N., Astrid A. Dick, Lawrence
G. Goldberg and Lawrence J. White. (2007).
“Competition from Large, Multimarket Firms and
the Performance of Small, Single- M a r k e t
Firms: Evidence from the Banking Industry.”
Journal of Money, Credit & Banking, 39(2-3), 331.
Under the efciency hypothesis, technological progress in
the 1990s signifcantly improved the performance of large,
multimarket banks relative to small, single-market banks;
therefore, a greater presence of large, multimarket banks exerted
more competitive pressure and had more deleterious efects on
the performance of small, single-market banks in their markets in
the second period, 1991-2000, than in the frst period, 1982-90.
The more intense competition from large, multimarket banks in
the second time period may be manifested in decreased revenues
for small, single-market banks (e.g., lower fees or rates on loans,
lower fees on deposits) and/or increased expenses (e.g., higher
rates on deposits, additional expenses on advertising or quality to
retain customers).
Authors note that the relevant research on bank size and
performance in the U.S. includes studies of cost and revenue
performance, as well as the abilities of banks of diferent sizes
to provide retail services in which both large and small banks
compete, such as loans to small businesses and deposits.
Having established some of the important determinants
of proftability for the banking industry today, a review of
the leading bank holding companies in the U.S. today is in
order, and this review is provided below.
Review of Top Five Bank Holding Companies in the
United States Today.
Rank Institution Name Location Total Assets
1 Citigroup Inc. New York, NY $2,220,866,000
2 Bank of America
Corporation
Charlotte, NC $1,535,684,280
3 JPMorgan Chase & Co. New York, NY $1,458,042,000
4 Wachovia Corporation Charlotte, NC $719,922,000
5 Wells Fargo &
Company*
San Francisco,
CA
$539,865,000
Note: As of June 30, 2007, Taunus Corporation was listed as the 5
th
largest bank holding company by the Federal Reserve; however, in
2003, Taunus changed its organization from a Bank Holding Company
to a Financial Holding Company – Domestic (seehttp://www. ffec.
gov/nicpubweb/ nicweb/InstitutionHistory.aspx?parID_RSSD=
2816906&parDT_END =99991231), and was therefore was replaced
with the next top BHC, Wells Fargo & Company, for the purposes of this
analysis; a list of the top ffty bank holding companies in the U.S. as of
June 2007 is provided at Appendix A.
Source: Federal Reserve National Information Center, 2007.
Table 3. United States' Largest Bank Holding Companies (as of
June 30, 2007).
Figure 6. United States' Largest Bank Holding Companies (as
of June 30, 2007).
$0
$500,000,000
$1,000,000,000
$1,500,000,000
$2,000,000,000
$2,500,000,000
Total Assets
Citigroup BAC JPMorgan Wachovia Wells Fargo & Co.
Source: Based on tabular data from Federal Reserve National
Information Center, 2007.
The top fve banking holding companies in the United
States today by consolidated assets are, respectively,
Citigroup Inc., Bank of America Corporation, JPMorgan
Chase & Co., Wachovia Corporation and Wells Fargo &
Company. These companies’ respective rank in the banking
industry, city and state of their headquarters, and total, are
shown in Table 3 and Figure 6 below.
2011 217
Overview of Top Five Bank Holding Companies in the
United States (as of June 2007).
An overview and brief description of the top fve bank
holding companies in the United States as of June 2007 is
provided below, followed by a recapitulation of management
summaries from each bank holding company’s most
recent Form 10-Q fling with the Securities and Exchange
Commission (SEC).
Citigroup Inc.
Citigroup, Inc. (hereinafter “Citigroup” or alternatively,
“the bank”) was established in 1812 and is headquartered
New York City (Citigroup, 2007). Citigroup provides a wide
range of fnancial services to its domestic and international
customers through the following business segments.
Business Segment Description
Global Consumer segment This segment features banking, lending,
insurance, and investment services; as of
March 31, 2007, this business unit boasted
a network of 8,140 branches, about 19,100
automated teller machines, 708 automated
lending machines, and the Internet.
Markets and Banking This segment is responsible for the
provision of various investment and
commercial banking services and
products; these services consist of
investment banking and advisory services,
debt and equity trading, institutional
brokerage, foreign exchange, structured
products, derivatives, and lending
(Citigroup, 2007). The bank also provides
cash management and trade fnance for
corporations and fnancial institutions;
custody and fund services to insurance
companies and pension funds; clearing
services to intermediaries; and depository
and agency/trust services to multinational
corporations and governments (Citigroup,
2007).
Global Wealth Management This segment of the bank ofers investment
advice, fnancial planning, and brokerage
services to afuent individuals, companies,
and non-proft organizations (Citigroup,
2007). This segment also provides wealth
management services for its high net
worth clients, including investment
management, such as investment funds
management, capital markets solutions,
trust, fduciary, and custody services;
investment fnance that comprises credit
services, such as real estate fnancing,
commitments, and letters of credit; and
banking services, which consist of deposit,
checking, and savings accounts, as well
as cash management and other banking
services (Citigroup, 2007).
Alternative Investments The bank’s Alternative Investments
segment manages products across fve
asset classes, such as private equity, hedge
funds, real estate, structured products, and
managed futures (Citigroup, 2007).
Bank of America Corp.
Bank of America Corporation was established in 1874
and is currently headquartered in Charlotte, North Carolina
(Bank of America, 2007). Bank of America Corporation
competes today through its banking and non-banking
subsidiaries as a provider of fnancial services and products
throughout the United States and in selected international
markets (hereinafter “the bank” or alternatively, “the
bank” or “Bank of America”). Through its banking and
nonbanking subsidiaries, the bank and its subsidiaries
provide a diverse range of fnancial services and products
throughout the U.S. and in selected international markets
(Bank of America, 2007). As of June 30, 2007, the Bank of
America operated its banking activities under two primary
charters: (a) Bank of America, National Association (Bank
of America, N.A.) and (b) FIA Card Services, N.A. (Form
10-Q, 2007).
The bank competes through three business segments as
shown in Table 5 below.
Table 5. Bank of American Business Segments.
Business Segment Description
Global Corporate and
Investment Banking
This segment provides commercial and
corporate bank loans, indirect consumer
loans, commitment facilities, real estate
lending products, and leasing and asset-
based lending products for banking
clients, middle market commercial clients,
multinational corporate clients, public
and private developers, homebuilders,
and commercial real estate frms; advisory
services, fnancing, and related products for
institutional investor clients in support of
their investing and trading activities; debt
and equity underwriting, merger-related
advisory services, and risk management
solutions; and treasury management, trade
fnance, foreign exchange, short-term credit
facilities, and short-term investing options for
correspondent banks, commercial real estate
frms, and governments.
Global Consumer and
Small Business Banking
This segment provides savings accounts,
money market savings accounts, certifcate
of deposits, individual retirement accounts,
and regular and interest-checking accounts;
consumer cards, business cards, debit cards,
international cards, and merchant services;
mortgage products for home purchase and
refnancing needs; insurance services; and
lines of credit and home equity loans.
Global Wealth and
Investment Management.
This segment features wealth management
and retail brokerage services, as well as asset
management services, including mutual
funds, liquidity strategies, and separate
accounts.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
218 Business Intelligence Journal July
As of July 30, 2007, the company operated approximately
5,700 retail banking offces and 17,000 automated teller
machines (Bank of America, 2007).
J. P. Morgan Chase & Company.
The company was established in 1823 and is currently
headquartered in New York, New York (JPMorgan, 2007).
According to its company profle, JPMorgan Chase & Co.
(hereinafter “the company” or alternatively, “the bank”),
through its subsidiaries, provides a range of fnancial
services worldwide (JP Morgan, 2007). The bank currently
operates through six segments as shown in Table 6 below.
Table 6. JP Morgan Business Segments.
Business Segment Description
Investment Bank This segment ofers investment banking
products and services, such as advising on
corporate strategy and structure, capital raising
in equity and debt markets, risk management,
market-making in cash securities and derivative
instruments, and research. It serves corporations,
fnancial institutions, governments, and
institutional investors.
Retail Financial Services This segment provides regional banking
services, including consumer and business
banking, home equity lending, and education
lending, as well as ofers mortgage banking and
auto fnance services.
Card Services This segment issues credit cards, and general-
purpose cards to individual consumers, small
businesses, and partner organizations, including
cards issued with AARP, Amazon, Continental
Airlines, Marriott, Southwest Airlines, Sony,
United Airlines, and Walt Disney Company
brands.
Commercial Banking This segment provides lending, treasury services,
investment banking, and asset management
services to corporations, municipalities, fnancial
institutions, and not-for-proft entities.
Treasury and Securities
Services
This segment provides transaction, investment,
and information services to institutional clients.
It also ofers custodian services and cash
management solutions, including trade fnance
and logistics solutions, wholesale card products,
and liquidity management services.
Asset Management This segment provides investment and wealth
management services to institutions, retail
investors, and high-net-worth individuals.
It ofers global investment management
services; trust, estate, and banking services; and
retirement services (JPMorgan, 2007).
Wachovia Corp.
The parent company, Wachovia Corporation, engages in
capital management, the general bank, wealth management,
and the corporate and investment bank businesses.
Wachovia (hereinafter “the company” or alternatively, “the
bank”), provides a range of commercial and retail banking,
as well as a range of trust services through its full-service
banking offces in the U.S. (Wachovia, 2007).
The bank provides its customers with a full range of
checking, savings, check card, foreign currency, annuities,
life insurance, brokerage account transfers, CAP accounts,
individual retirement accounts, credit cards, home equity,
mortgage, hazard and food insurance, escrow, taxes, private
mortgage insurance, education loans, online services, online
banking, online bill pay, and online brokerage services
(Wachovia, 2007). Wachovia was established in 1879 and
is currently headquartered in Charlotte, North Carolina
(Wachovia, 2007).
The company also provides various other fnancial
services, including mortgage banking, investment banking,
estate planning, investment advisory, asset management,
credit and debit card products, trust services, charitable
services, mortgage banking, asset-based lending, leasing,
insurance, and international and securities brokerage
services. In addition, it engages in equity and debt
underwriting activities, private equity investment activities,
derivative securities activities, investment and wealth
management advisory businesses, and brokerage activities.
As of June 1, 2006, Wachovia operated 3,159 offces in 16
states, as well as operated 40 offces internationally. The
company reports assets of approximately $541.8 billion, as
of June 1, 2006 (Wachovia, 2007).
On a fnal note, according to the company’s most recent
Form 10-Q fling:
In the second quarter of 2007, we announced a
realignment of some of our businesses and other corporate
functions. This included the combination of the General
Bank’s private advisory group into the Wealth Management
businesses and the General Bank’s commercial real
estate business into the Corporate and Investment Bank’s
investment banking line of business, which is expected to
occur by the end of the year. We are still evaluating how
these realignments may affect our segment reporting for
future periods; however, we expect Wealth Management
will remain a separate reporting segment. (Form 10-Q, p. 4)
Wells Fargo & Company.
Wells Fargo & Company was established in 1929 and is
currently headquartered in San Francisco, California (Wells
Fargo, 2007). According to its company profle, Wells Fargo
& Company (hereinafter “the company” or alternatively,
“the bank”), through its subsidiaries, provides banking and
fnancial products and services in the United States. The
bank operates in three segments as shown in Table 7 below.
2011 219
Table 7. Wells Fargo & Company Business Segments.
Business Segment Description
Community Banking This segment provides a comprehensive
group of deposit products, including checking
accounts, savings deposits, market rate
accounts, individual retirement accounts, time
deposits, and debit cards; this segment’s loan
portfolio includes: lines of credit; equity lines
and loans; equipment and transportation loans,
including recreational vehicle and marine;
education loans; origination and purchase
of residential mortgage loans; servicing of
mortgage loans; and credit cards (Wells Fargo,
2007). This segment also provides receivables
and inventory fnancing, equipment leases, real
estate fnancing, small business administration
fnancing, venture capital fnancing, cash
management, payroll services, retirement plans,
health savings accounts, and credit and debit
card processing services.
Wholesale Banking This segment provides commercial, corporate,
and real estate banking products and services
in the United States. This segment’s products
include traditional commercial loans and lines
of credit, letters of credit, asset-based lending,
equipment leasing, mezzanine fnancing, high-
yield debt, international trade facilities, foreign
exchange services, treasury management,
investment management, institutional
fxed income and equity sales, interest rate,
commodity and equity risk management,
online/electronic products, insurance, and
investment banking services (Wells Fargo, 2007).
Wells Fargo Financial This segment consists of consumer fnance and
auto fnance operations. It also provides credit
cards and lease, and other commercial fnancing
services; as of March 24, 2007, this segment
provided its services through approximately
6000 branches (Wells Fargo, 2007).
Table 8. Recapitulation of Citigroup’s Management Summary from Most Recent Form 10-Qs.
Date of Filing Organization Key Performance Metric Highlights Comments
August 3, 2007 Citigroup Income from continuing operations rose 18% to $6.226 billion and was the highest
ever recorded by the bank. Diluted EPS from continuing operations was also up
18%. Revenues were a record $26.6 billion, an increase of 20% over the previous
year to date, led by Markets & Banking, up 33%. The bank’s international operations
recorded revenue growth of 34% in the quarter, with International Consumer up
16%, International Markets & Banking up 50%, and International Global Wealth
Management more than doubling. U.S. Consumer revenues grew 3%, while
Alternative Investments revenues increased an impressive 77%.
Acquisitions represented approximately 4% of the
bank’s revenue growth.
August 7, 2007 Bank of America At June 30, 2007, the Corporation had $1.5 trillion in assets and approximately
196 thousand full-time equivalent employees. Net interest income on a FTE basis
decreased $145 million to $8.8 billion and $588 million to $17.4 billion for the three
and six months ended June 30, 2007 compared to the same periods in 2006. The
primary drivers of the decreases were the impact of the divestitures of certain foreign
operations in 2006 and the frst quarter of 2007, increased hedge costs, higher
cost of deposits, spread compression, reduced benefts from purchase accounting
adjustments and the negative impact of the adoption of FSP 13-2. These decreases
were partially ofset by a higher contribution from market-based activity, higher
levels of consumer and commercial domestic loans and increased ALM portfolio
levels.
In July 2007, the Corporation completed the acquisition
of U.S. Trust Corporation (U.S. Trust) for $3.3 billion in
cash.
The company reports that the competitive environment
varies for its far-fung operations. For example, the
bank’s most recent Form 10-Q fling reports that, “The
fnancial services industry is highly competitive. Our
subsidiaries compete with fnancial services providers,
such as banks, savings and loan associations, credit
unions, fnance companies, mortgage banking companies,
insurance companies, and money market and mutual fund
companies” (March 1, 2007 p. 3). Beyond this broad range
of competitors, the bank’s subsidiaries were also confronted
with a number of challenges on some new fronts as well:
“They also face increased competition from nonbank
institutions such as brokerage houses and insurance
companies, as well as from fnancial services subsidiaries
of commercial and manufacturing companies. Many of
these competitors enjoy fewer regulatory constraints and
some may have lower cost structures” (Form 10-Q, March
1, 2007 p. 3).
Recapitulation of Management Summaries from Most
Recent Form 10-Qs.
A recapitulation of the respective fve leading bank
holding companies’ management summaries from their
most recent Form 10-Q flings with the SEC and comments
are provided in Table 8 below.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
220 Business Intelligence Journal July
Date of Filing Organization Key Performance Metric Highlights Comments
August 9, 2007 JP Morgan The company reports $1.5 trillion in assets, $119.2 billion in stockholders’ equity and
operations worldwide. The company reported 2007 second-quarter Net income of
$4.2 billion, or $1.20 per share, compared with Net income of $3.5 billion, or $0.99
per share, for the second quarter of 2006. Return on common equity for the quarter
was 14% compared with 13% in the prior year.
Net income for the frst six months of 2007 was $9.0 billion, or $2.55 per share,
compared with $6.6 billion, or $1.85 per share, in the comparable period last year.
Return on common equity was 16% for the frst six months of 2007 compared with
12% for the prior-year period.
The second quarter of 2007 economic environment
was a contributing factor to the performance of the
Firm and each of its businesses. The overall economic
expansion, strong level of capital markets activity and
positive performance in equity markets helped to
drive new business volume and organic growth within
each of the company’s wholesale businesses; however,
weakness in the housing markets resulted in increased
losses in Retail Financial Services, causing in an increase
in provision related to the home equity portfolio.
July 30, 2007 Wachovia
Corporation
Wachovia’s net income in the frst six months of 2007 was $4.6 billion, up 29 percent
from the frst six months of 2006, and diluted earnings per common share were up 7
percent to $2.42. After-tax net merger-related and restructuring expenses amounted
to 1 cent per share in the frst six months of 2007 and 4 cents per share in the same
period of 2006. Tax-equivalent net interest income increased 24 percent in the frst
six months of 2007 from the frst six months of 2006, refecting a larger balance sheet.
The net interest margin declined 22 basis points to 2.97 percent, primarily due to
growth in lower-spread consumer and commercial loans, a shift in deposits toward
lower-spread categories, the impact of acquisitions and the efect of an inverted
yield curve.
Wachovia and certain of its subsidiaries are currently
involved in a number of judicial, regulatory and
arbitration proceedings concerning matters arising
from the conduct of its business activities; these
proceedings include actions brought against Wachovia
and/or its subsidiaries with respect to transactions in
which Wachovia and/or its subsidiaries acted as banker,
lender, underwriter, fnancial advisor or broker or in
activities related thereto. The actual costs of resolving
legal claims may be substantially higher or lower than
the amounts reserved for those claims.
March 1, 2007 Wells Fargo At December 31, 2006, the company reports assets of $482 billion, loans of $319
billion, deposits of $310 billion and stockholders’ equity of $46 billion; based on
total assets, WFC is the ffth largest bank holding company in the United States. At
December 31, 2006, the bank had 158,000 active, full-time equivalent team members.
The fling reports that the bank enjoyed record earnings in 2006 with record diluted
earnings per share of $2.49, record net income of $8.5 billion, both up 11%, and
exceptional, broad-based performance across the company’s 80+ businesses. The
report also emphasizes that, “Over the past twenty years, our annual compound
growth rate in earnings per share was 14% and our annual compound growth rate
in revenue was 12%. Our total annual compound stockholder return of 14% the past
fve years was more than double the S&P 500 — and at 15% almost double for the
past ten years. We far out-paced the S&P 500 the past 15 and 20 years with total
annual compound shareholder returns of 18% and 21%, respectively — periods with
almost every economic cycle and economic condition a fnancial institution can
experience” (p. 34).
All common share and per share disclosures in this
Report refect the two-for-one stock split in the form of
a 100% stock dividend distributed on August 11, 2006.
Sources: As indicated.
Stock Performance of Top Five Bank Holding
Companies in the U.S.: Past Five Years to Date.
A comparison of the top fve bank holding companies’
stock performance for the past fve years to date is provided
in Figure 7 below.
Figure 7. Stock Performance of Top Five Bank Holding
Companies in the U.S.: Past Five Years to Date
Source: Yahoo! Finance, 2007.
Key:
C = Citigroup Inc.
BAC = Bank of America
JPM = JPMorgan Chase & Co.
WB = Wachovia
WFC = Wells Fargo & Company
Methodology
Description of the Study Approach
This study used a mixed methodology to achieve the
above-stated research purpose. The frst component of the
research methodology consisted of a comprehensive review
of the relevant peer-reviewed, scholarly and organizational
literature. This approach is congruent with various social
researchers who report that such a literature review is the
natural place to begin such an endeavor (Neuman, 2003;
Wood and Ellis, 2003). According to one authority, “Both
the opinions of experts in the feld and other research
studies are of interest. Such reading is referred to as a
review of the literature” (Fraenkel and Wallen, 2001 p.
48). Similarly, Gratton and Jones (2003) emphasize that
2011 221
Likewise, Silverman (2005, p. 300) suggests that a
literature review should aim to answer the following
questions:
1. What do we know about the topic?
2. What do we have to say critically about what is already
known?
3. Has anyone else ever done anything exactly the same?
4. Has anyone else done anything that is related?
5. Where does your work ft in with what has gone before?
6. Why is your research worth doing in the light of what
has already been done?
It was with the above goals and guidance that the review
of the relevant peer-reviewed, scholarly and organizational
data was conducted. Both university and public libraries
were consulted for this purpose, as well as a range of
online resources including EBSCO, Questia, Edgar Online,
Yahoo! Finance and others.
The econometric approach used in the study is also
congruent with a number of economic researchers who
emphasize that it is common practice for economists to
describe or discuss a theory in terms of an equation or a
set of equations; moreover, even elementary economics
textbooks present postulated relationships between
economic variables in an algebraic form, and suggest
inferences by mathematical manipulations (Charemza,
Deadman & Elgar, 1997).
Therefore, extending the process to include assignation
of quantitative measures to these relationships represents a
logical step. According to Charemza and his colleagues, “The
most widely used tool of economists to determine empirical
forms of theoretical constructs is that of econometrics.
The likely originator of the term 'econometrics' defned
it as '...the unifcation of economic theory, statistics, and
mathematics...' Much of the early empirical work in
economics (in other words, prior to 1940) was concerned
with the measurement of demand elasticities, and the
representation of the business cycle” (p. 1).
To a great extent, these trends were refective of the
activity of economists in developing theory in these areas,
and the increasing availability of reasonably long runs
of statistical data on agricultural commodities, foreign
trade and various industries (Charemza et al., 1997).
Subsequent developments of national income accounting
in conjunction with Keynesian economic theory further
created new opportunities for the econometric analysis
of macroeconomic series, including complete models of
economies; these estimated macroeconomic models could
be used for economic policy purposes, such as forecasting
or simulation (Charemza et al., 1997).
Regardless of the objectives being considered, it is
clear that in order to undertake econometric analysis, the
following must be used:
1. A relevant economic theory;
2. Statistical data;
3. A method that allows for the expression of the economic
theory using the statistical data (in practice, a theory
of estimation stemming from econometric theory)
(six properties that would be considered desirable
in an estimated model: (a) relevance, (b) simplicity,
(c) theoretical plausibility, (d) explanatory ability, (e)
a critical reviewing of the timely literature is an essential
component of almost any type of research endeavor: “No
matter how original you think the research question may
be,” they advise, “it is almost certain that your work will
be building on the work of others. It is here that the review
of such existing work is important. A literature review is
the background to the research, where it is important to
demonstrate a clear understanding of the relevant theories
and concepts, the results of past research into the area, the
types of methodologies and research designs employed in
such research, and areas where the literature is defcient”
(p. 51).
In addition, Wood and Ellis (2003) identifed the
following as important outcomes of a well conducted
literature review:
1. It helps describe a topic of interest and refne either
research questions or directions in which to look;
2. It presents a clear description and evaluation of the
theories and concepts that have informed research into
the topic of interest;
3. It clarifes the relationship to previous research and
highlights where new research may contribute by
identifying research possibilities which have been
overlooked so far in the literature;
4. It provides insights into the topic of interest that are
both methodological and substantive;
5. It demonstrates powers of critical analysis by, for
instance, exposing taken for granted assumptions
underpinning previous research and identifying
the possibilities of replacing them with alternative
assumptions;
6. It justifes any new research through a coherent critique
of what has gone before and demonstrates why new
research is both timely and important.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
222 Business Intelligence Journal July
accuracy of coeffcients and (f) forecasting ability; a
“good” model will display all of these properties to
some degree, but the existence of a potentially large
number of theoretically plausible models which also
satisfy some or all of the criteria makes the model
choice problem a nontrivial one in practice;
4. A “know-how,” which guides how to apply the
estimation theory to the statistical data, and how to
decide whether this application has been successful; in
this case, the requisite “know hows” relate to methods
of defning “good” models, and in fnding them
(Charemza et al., 1997, p. 3).
Based on the foregoing guidance, this study used a
weighted average approach to determining the impact of
the external national economy on the fve leading bank’s
proftability according to their total assets. The econometric
model described further in Chapter 4 below that was
developed for this study was deemed to have satisfed all of
these criteria to some extent.
Data-Gathering Method and Database of
Study
Beyond the critical review of the relevant peer-reviewed,
scholarly and organizational literature concerning
proftability determinants in the banking industry today, this
study also used bank-level data for the top fve banks in the
United States for last fve years for its econometric analysis,
as well as fve years’ of data from various international
fnancial statistics for the U.S. banking industry in general.
As noted above, these data were collected from reliable
online resources such as the U.S. Federal Reserve System’s
National Information Center, Free Edgar, Yahoo! Finance
and comparable organizational and governmental Web
sites.
Data Analysis
Econometric Model.
In their study, “Proftability of Credit Unions,
Commercial Banks and Savings Banks: A Comparative
Analysis,” Kaushik and Lopez (1996) report that,
“Proftability is the measure of both performance of each of
the industries and the degree of competition among them”
(p. 66). On the one hand, a straightforward approach to
defning competitiveness at any industry level is in terms
of the percentage of market shares held domestically and
internationally (Saunders and Walter, 1998). While this is
fairly simple task when it comes to industries such as steel
or automobiles that provide comparatively homogeneous
goods and services, such an application to the fnancial
services industry is far more complicated because of
the wide range of mixes and national settings involved
(Saunders and Walter, 1998).
Moreover, defning competitiveness in these terms may
not be especially refective of an individual organization’s
proftability, which depends as well on a variety of market
structure and other cost factors such as scale and scope
economies (Saunders and Walter, 1998). On the other hand,
using market share is a commonly used indicator of the
outcome of the competitive process itself because in free
markets, companies that enjoy the lowest cost structure or
feature highest product quality generally gain market share
compared to those with a higher cost structure or lower
product quality (Saunders and Walter, 1998).
In terms of applying this method to the fnancial services
industry, the picture becomes more complex. For instance,
Saunders and Walter emphasize that, “The fnancial services
industry sells perhaps 50 more or less distinct services to
perhaps 20 more or less distinct client groups. They range
from credit card loans to lower middle income households
all the way to swap-driven repackaged synthetic securities
engineered for the largest multinational companies. Some
of these services are highly internationalized, indeed
globalized, while others are individually sold or mass-
marketed domestically” (p. 24). Further complicating any
such across-the-board analysis is the paucity of market share
data; in some cases, this information is reasonably available;
however, in other cases, there are no available market share
data whatsoever (Saunders and Walter, 1998). Moreover,
some information is also obscured or contaminated by
exchange rate changes that take place during the periods
of time that data are available to researchers (Saunders and
Walter, 1998).
Based on the foregoing constraints, a better indicator
of proftability was required for the purposes of this
investigation. As noted above, the empirical test used in
this study is concerned with the determinants proftability
of these fve leading bank holding companies in the United
States as of June 2007. For this purpose, the measure of
proftability of each bank was defned as the return on assets
(ROA); the ROA is a ratio that is calculated by dividing
the net income over total assets. The macro-economic
variable GDP per capita growth was also used in the model
as estimated by the CIA World Factbook for 2006 and the
International Monetary Fund for the years 2004 and 2005.
2011 223
The GDP per capita growth rate was expected to have
a positive impact on bank’s performance according to the
well documented literature on the association between
economic growth and fnancial sector performance by
virtue of economies of scale that accrue to larger banking
organizations whose assets run into the billions. This is
congruent with the fndings of Berger et al. (2007) who
report, “Early research on bank cost scale economies
using data on U.S. banks from the 1980s generally fnds
very little scale economies or diseconomies except at very
small sizes, typically well under $1 billion in assets. Later
research suggests that there may be more extensive cost
scale economies in the 1990s, with average costs declining
up to asset sizes of $25 billion or more” (p. 331).
The rationale in support of this model was found
extensively throughout the body of evidence concerning
relevant proftability indicators for the banking industry
that clearly demonstrated economies of scale for larger
fnancial institutions, particularly as they compete across
international borders, notwithstanding any substantive
changes in the external operating environment to the
contrary. To gauge the soundness of this rationale, the
degree of proftability for each bank holding company
was determined by using a return on investment weighted
average approach. In this regard, Li, Pan and Tse (1999)
report that, “Proftability is measured by returns on assets
(ROA)” (p. 81). According to the editors of MoneyTerms,
“A weighted average is more heavily infuenced by some
of the numbers it is calculated from than others. It is
calculated by multiplying each number by a weight, adding
these together and then dividing the total by the sum of the
weights” (Weighted Average, 2007 p. 3). The formula for
this model is shown below:
ROA (net income/total assets) * GDP per capita growth
/ (ROA*GDP per capita growth/ROA+GDP per capita
growth)
The return on assets calculations internal aspect for the
fve respective banks compared with the GDP per capital
growth external factor rate is provided in Tables 9 through
13 below, as well as graphic representations of the data in
the accompanying fgures; the percentage differential is
also shown.
Table 9. Bank of America: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 21,133,000 16,465,000 14,143,000
Total Assets 1,459,737,000 1,291,803,000 1,110,457,000
Return on Assets
(ROA)
1.45% 1.27% 1.27%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 1.00% 0.95% 0.98%
Percentage
Diference ROA-
Weighted Avg.
0.45% 0.33% 0.30%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 10. JP Morgan: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 14,444,000 8,483,000 4,466,000
Total Assets 1,351,520,000 1,198,942,000 1,157,248,000
Return on Assets
(ROA)
1.07% 0.71% 0.39%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 0.80% 0.59% 0.35%
Percentage
Diference ROA-
Weighted Avg.
0.27% 0.11% 0.03%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
224 Business Intelligence Journal July
Table 11. Citigroup: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 21,538,000 24,589,000 17,046,000
Total Assets 1,884,318,000 1,494,037,000 1,484,101,000
Return on Assets
(ROA)
1.14% 1.65% 1.15%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 0.84% 1.14% 0.90%
Percentage
Diference ROA-
Weighted Avg.
0.30% 0.51% 0.25%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 12. Wachovia: December 31-December 2006.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 7,791,000 6,643,000 5,214,000
Total Assets 707,121,000 520,755,000 493,324,000
Return on
Assets (ROA)
1.10% 1.28% 1.06%
Per Capita
Growth Rate*
3.20% 3.70% 4.20%
Weighted
Average
0.82% 0.95% 0.84%
Percentage
Diference
ROA-Weighted
Avg.
0.28% 0.33% 0.21%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 13. Wells Fargo & Co.: December 31-December 2006
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Net Income 8,482,000 7,671,000 7,014,000
Total Assets 481,996,000 481,741,000 427,849,000
Return on Assets
(ROA)
1.76% 1.59% 1.64%
Per Capita Growth
Rate*
3.20% 3.70% 4.20%
Weighted Average 1.14% 1.11% 1.18%
Percentage
Diference ROA-
Weighted Avg.
0.62% 0.48% 0.46%
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
2004 2005 2006
Per Capita Growth Rate*
Weighted Average
Percentage Difference ROA-Weighted Avg.
Table 14. Comparison of Percentage Differences in ROA-
Weighted Averages.
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Bank of America 0.45% 0.33% 0.30%
JP Morgan 0.27% 0.11% 0.03%
Citigroup 0.30% 0.51% 0.25%
Wachovia 0.28% 0.33% 0.21%
Wells Fargo 0.62% 0.48% 0.46%
0.00%
0.10%
0.20%
0.30%
0.40%
0.50%
0.60%
0.70%
2004 2005 2006
Bank of America JP Morgan Citigroup Wachovia Wells Fargo
2011 225
Summary and Conclusions
Summary
The purpose of this study was to develop an appropriate
econometric model whereby the primary determinants of
proftability of the top fve bank holding companies in the
United States could be examined and understood. To this
end, the research showed that proftability determinants for
the banking industry include positive relationship between
the return of equity and capital to asset ratio as well as the
annual percentage changes in the external per capita income.
There was also a virtual consensus identifed concerning
the effect that the internal factor of size as measured by
an organization’s total assets had on its ability to compete
more effectively, even in times of economic downturns.
This relationship was clearly identifed in the data
analysis, wherein all fve of the leading bank holding
companies in the United States (as of June 2007) enjoyed
increases in their weighted averages of return on assets
despite decreases in the GDP per capita rate. While there
were some fuctuations and variations identifed between
the respective organizations in spite of their relative sizes,
these differences could be accounted for by virtue of specifc
economic initiatives in place during these isolated points in
time that are not accounted for in the analytical model.
Some of the constraints identifed during the research
process included the fact that the majority of research to
date has investigated determinants of proftability at the
bank and/or industry level, with the choice of variables
used failing to provide internal consistency in some cases
and a paucity of timely research concerning the potential
infuence of the macroeconomic environment in which
fnancial services companies compete. There were also
a broad range of unique fscal activities that took place
during the time period examined that clearly infuenced a
given bank’s proftability, either positively or negatively,
over the short-term while failing to provide any indication
of its potential long-term impact on the bank’s proftability.
In the fnal analysis, it would appear that the industry
analysts and experts were absolutely correct in their
assertions that although it is possible to develop a model that
can provide researchers with an indication of the relative
importance of internal and external factors on a company’s
proftability, the analysis is rife with opportunities to miss
important yet unforeseeable infuences that may contribute
to changes in short-term proftability while leaving long-term
proftability unaffected. Likewise, there are unforeseeable
factors concerning innovations in technology that relate
both to a fnancial services company’s internal and external
environment that will depend on the effectiveness of the
bank’s management to add value or not. Responsiveness
to the external economic environment and potential threats
to far-fung operations based on terrorist activities abroad
represent yet more factors that may be highly signifcant,
but are diffcult to model accurately.
Conclusions
The studies on the determinants of bank’s performance
in the United States in recent years have shown some
mixed results, with some researchers fnding that little
cost saving can be achieved by increasing the size of the
banking frm and others report signifcant scale economies
for banks whose asset size extends well into the billion
range such as those investigated herein. In addition to being
extremely diffcult to measure on an international level,
determinants of proftability may be skewed by an inability
to obtain accurate and timely data and by an unbalanced
competitive environment in which a given bank may be
forced to compete with smaller players that can market
high proft products more effciently. In fact, it would be
reasonable to conclude that the determinants of proftability
can be discerned at a given point in time as they relate to
a specifc set of factors, but there is much more involved
in the analysis than a straight-forward modeling approach
can address. Nevertheless, the research was absolutely
consistent in emphasizing the need to gauge proftability
and to identify what revenue sources provide fnancial
services companies with the biggest “bang for the buck”
so they can focus their energies where they will do the
most good both in the short term and for the long haul. In
conclusion, analyzing proftability in the fnancial services
sector requires constructing a comprehensive picture of a
wide range of relevant performance metrics. This analysis
also requires a signifcant amount of judiciousness and
recognition of their limitations when interpreting them.
References
Abreu, M. and V. Mendes. (2002). “Commercial bank
interest margins and proftability: evidence from
E.U countries.” Porto Working paper series in Ben
Naceur, 2003.
Angbazo, L. (1997). “Commercial bank net interest
margins, default risk, interest-rate risk, and
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
226 Business Intelligence Journal July
off-balance sheet banking.” Journal of Banking and
Finance, 21, 55-87 in Ben Naceur, 2003.
Athanasoglou, Panayiotis P., Sophocles N. Brissimis
and Matthaios D. Delis. (2005, June). “ Ba nk-
Specifc, Industry-Specifc and Macroeconomic
Determinants of Bank Proftability.” MPRA
Paper 153, University Library of Munich,
Germany, revised 2006. [Online]. Available:http://www.bankofgreece.gr/publications/pdf/
Paper200525.pdf.
Bank of American Form 10-Q. (2007, August 7). Edgar
Online. [Online]. Available: ht t p: / / yahoo.
brand.edgar-online.com/ fetchFilingFrameset.
aspx?dcn=0001193125-07-
175583&Type=HTML.
Bank of America. (2007). Yahoo! Finance Company
Profiles. [Online]. Available: http:// finance.
yahoo.com/q/pr?s=BAC.
Barajas, A., R. Steiner, and N. Salazar. 1999. “Interest
spreads in banking in Colombia 1974- 1 9 9 6 . ”
IMF Staff Papers, 46, 196-224.
Ben Naceur, Samy. (2003, October). “The
Determinants of the Tunisian Banking Industry
Proftability: Panel Evidence.” Université Libre de
Tunis: Department of Finance.
Ben Naceur Samy and M. Goaied. (2001). “The
determinants of the Tunisian deposit banks’
performance.” Applied Financial Economics, 11,
317-19 in Ben Naceur, 2003.
Berger, Allen N. (1995). “The relationship between
capital and earnings in banking.” Journal o f
Money, Credit and Banking, 27, 404-31 in Ben
Naceur, 2003.
Berger, Allen N. and R. Deyoung. (2006). Technological
progress and the geographic expansion of the
banking industry. Journal of Money, Credit &
Banking, 38(6), 1483.
Berger, Allen N., Astrid A. Dick, Lawrence G.
Goldberg and Lawrence J. White. (2007).
“Competition from Large, Multimarket Firms and
the Performance of Small, Single- Market Firms:
Evidence from the Banking Industry.” Journal of
Money, Credit & Banking, 39(2-3), 331.
Bourke, P. (1989). “Concentration and other
determinants of bank proftability in Europe, North
America and Australia.” Journal of Banking and
Finance 13, 65-79 in Athanasoglou et a l . ,
2005.
Brealey, Richard A., Alastair Clark, Charles Goodhart,
Juliette Healey, Glenn Hoggarth et al. ( 2001) .
Financial Stability and Central Banks: A Global
Perspective. London: Routledge.
Charemza, W. W. and D. F. Deadman. (1997). New
Directions in Econometric Practice: General
to Specifc Modelling, Cointegration, and Vector
Autoregression. Lyme, NH: Edward Elgar.
Citigroup, Inc. (2007). Yahoo! Finance Company
Profles. [Online]. Available:http://fnance.
yahoo.com/q/pr?s=C.
Cover, Jerry. (1999). “Proftability Analysis
- A Necessary Tool for Success in the 21st
Century.” ABA Banking Journal, 91(2), 78.
Demirguc-Kunt, A. and H. Huizinga. (2000). Financial
Structure and Bank Profitability. World B a n k
Mimeo in Athanasoglou et al., 2005
Fraenkel, J. R. and N. E. Wallen. (2001). Educational
Research: A Guide to the Process.
Mahwah, NJ: Lawrence Erlbaum Associates.
Frieder, Larry A. (1991). “Determinants of Bank
Acquisition Premiums: Issues and Evidence.”
Contemporary Policy Issues, 9(2), 14.
Goddard, J., P. Molyneux and J. O. S. Wilson. (2004a).
“Dynamics of Growth and Profitability i n
Banking.” Journal of Money, Credit & Banking,
36(6), 1069.
--- (2004b). “The proftability of European banks: a
cross-sectional and dynamic panel
analysis.” Manchester School, 72(3), 363-381.
Gratton, C. and I. Jones. (2003). Research Methods for
Sport Studies. New York: Routledge.
Grigorian, David A. and Vlad Manole. (2006).
“Determinants of Commercial Bank Performance
in Transition: An Application of Data Envelopment
Analysis.” Comparative Economic Studies,
48(3), 497.
Gup, B. E. (2003). The Future of Banking. Westport,
CT: Quorum Books.
Guru B., J. Staunton and Balashanmugam. 2002.
“Determinants of commercial bank proftability
in Malaysia.” University Multimedia working
papers in Ben Naceur, 2003.
JP Morgan. (2007). Yahoo! Finance Company Profles.
[Online]. Available: http:// finance.yahoo.
com/q/pr?s=JPM.
Kaushik, Surendra K. and Raymond H. Lopez. (1996)
“Profitability of Credit Unions, Co mme r c i a l
2011 227
Banks and Savings Banks: A Comparative
Analysis.” American Economist, 40(1), 66.
Li, Shaomin, Yigang Pan and David K. Tse. (1999).
“The Impact of Order and Mode of Market
Entry on Proftability and Market Share.” Journal
of International Business Studies, 30(1), 81.
Mogel, L. (2002). Making It in Public Relations:
An Insider’s Guide to Career Opportunities.
Mahwah, NJ: Lawrence Erlbaum Associates.
Molyneux, P. and J. Thornton. (1992). “Determinants
of European bank profitability: A note. Jour nal
of Banking and Finance, 16, 1173-1178 in
Athanasoglou et al., 2005.
“Weighted Average.” (2007). MoneyTerms.
Co. [Online]. Available: moneyterms.co.uk/
glossary-p/.
Neeley, M. C. and D. C. Wheelock. (1997, March/
April). “Why Does Bank Performance Vary
Across States?” Federal Reserve Bank of St. Louis
Review, 79(2), 27-40.
Neuman, W. Lawrence. (2003). Social Research
Methods: Qualitative and Quantitative Approaches,
5th ed. New York: Allyn & Bacon.
Prendergast, C. (1993, March). “The Provision of
Incentives in Firms.” Journal of Economic
Literature, 7, 63 in Brealey et al., 2001.
Saunders, Anthony and Ingo Walter. (1998). Universal
Banking in the United States: What C o u l d
We Gain? What Could We Lose? New York:
Oxford University Press.
Short, B. K. (1979). “The relation between commercial
bank profit rates and banking concent rat i on
in Canada, Western Europe and Japan.” Journal of
Banking and Finance 3, 209-219.
Silverman, D. (2005). Doing Qualitative Research: A
Practical Handbook (2nd ed.). London: Sage.
Spindler, J. Andrew, Jonathan T. B. Howe, and David
F. Dedyo. (1990, May). “The Performance
of Internationally Active Banks and Securities Firms
Based on Conventional C o mp e t i t i v e n e s s
Measures”. Federal Reserve Bank of New York,
Mimeo in Saunders and Walter, 1998.
Stiroh, Kevin J. (2004). “Diversifcation in Banking:
Is Noninterest Income the Answer?” Jour nal
of Money, Credit & Banking, 36(5), 853.
Toyne, Michael F. and James D. Tripp. (1998).
“Interstate Bank Mergers and Their Impact on
Shareholder Returns: Evidence from the 1990s.”
Quarterly Journal of Business and Economics,
37(4), 48.
Wachovia Corp. (2007). Yahoo! Finance Company
Profles. [Online]. Available:http://fnance.
yahoo.com/q/pr?s=WB.
Wells Fargo & Co. (2007). Yahoo! Finance Company
Profiles. [Online]. Available: http:// finance.
yahoo.com/q/pr?s=WFC.
Wood, G. D. and R. C. Ellis. (2003). “Risk management
practices of leading UK cost cons ul t ant s . ”
Engineering, Construction and Architectural
Management, 10(4), 254-62.
Yahoo! Finance. (2007). [Online]. Available: http://
fnance.yahoo.com/.
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
228 Business Intelligence Journal July
United States' Largest Bank Holding Companies (as of June 2007)
Appendix A
Rank Institution Name (RSSD ID) Location Total Assets
1 CITIGROUP INC. (1951350) NEW YORK, NY $2,220,866,000
2 BANK OF AMERICA CORPORATION (1073757) CHARLOTTE, NC $1,535,684,280
3 JPMORGAN CHASE & CO. (1039502) NEW YORK, NY $1,458,042,000
4 WACHOVIA CORPORATION (1073551) CHARLOTTE, NC $719,922,000
5 TAUNUS CORPORATION (2816906) NEW YORK, NY $579,062,000
6 WELLS FARGO & COMPANY (1120754) SAN FRANCISCO, CA $539,865,000
7 HSBC NORTH AMERICA HOLDINGS INC. (3232316) PROSPECT HEIGHTS, IL $483,630,057
8 U.S. BANCORP (1119794) MINNEAPOLIS, MN $222,530,000
9 SUNTRUST BANKS, INC. (1131787) ATLANTA, GA $180,314,372
10 ABN AMRO NORTH AMERICA HOLDING COMPANY (1379552) CHICAGO, IL $160,341,966
11 CITIZENS FINANCIAL GROUP, INC. (1132449) PROVIDENCE, RI $159,392,731
12 CAPITAL ONE FINANCIAL CORPORATION (2277860) MCLEAN, VA $145,937,957
13 NATIONAL CITY CORPORATION (1069125) CLEVELAND, OH $140,648,168
14 REGIONS FINANCIAL CORPORATION (3242838) BIRMINGHAM, AL $137,624,205
15 BB&T CORPORATION (1074156) WINSTON-SALEM, NC $127,577,050
16 BANK OF NEW YORK COMPANY, INC., THE (1033470) NEW YORK, NY $126,457,000
17 PNC FINANCIAL SERVICES GROUP, INC., THE (1069778) PITTSBURGH, PA $125,736,711
18 STATE STREET CORPORATION (1111435) BOSTON, MA $112,345,777
19 FIFTH THIRD BANCORP (1070345) CINCINNATI, OH $101,389,721
20 KEYCORP (1068025) CLEVELAND, OH $93,490,903
21 BANCWEST CORPORATION (1025608) HONOLULU, HI $70,661,335
22 HARRIS FINANCIAL CORP. (1245415) WILMINGTON, DE $64,475,903
23 NORTHERN TRUST CORPORATION (1199611) CHICAGO, IL $59,609,734
24 COMERICA INCORPORATED (1199844) DETROIT, MI $58,945,727
25 MARSHALL & ILSLEY CORPORATION (1199497) MILWAUKEE, WI $58,327,527
26 M&T BANK CORPORATION (1037003) BUFFALO, NY $57,869,069
27 UNIONBANCAL CORPORATION (1378434) SAN FRANCISCO, CA $53,173,833
28 CHARLES SCHWAB CORPORATION, THE (1026632) SAN FRANCISCO, CA $49,003,812
29 ZIONS BANCORPORATION (1027004) SALT LAKE CITY, UT $48,703,130
30 COMMERCE BANCORP, INC. (1117679) CHERRY HILL, NJ $48,231,325
31 POPULAR, INC. (1129382) SAN JUAN, PR $46,985,000
32 MELLON FINANCIAL CORPORATION (1068762) PITTSBURGH, PA $43,389,057
33 TD BANKNORTH INC. (1249196) PORTLAND, ME $42,981,084
34 FIRST HORIZON NATIONAL CORPORATION (1094640) MEMPHIS, TN $38,395,825
35 HUNTINGTON BANCSHARES INCORPORATED (1068191) COLUMBUS, OH $36,422,081
36 COMPASS BANCSHARES, INC. (1078529) BIRMINGHAM, AL $34,938,942
37 SYNOVUS FINANCIAL CORP. (1078846) COLUMBUS, GA $33,295,823
38 NEW YORK COMMUNITY BANCORP, INC. (2132932) WESTBURY, NY $29,638,404
39 RBC CENTURA BANKS, INC. (1826056) RALEIGH, NC $25,374,678
40 COLONIAL BANCGROUP, INC., THE (1080465) MONTGOMERY, AL $23,823,484
41 ASSOCIATED BANC-CORP (1199563) GREEN BAY, WI $20,849,531
2011 229
Rank Institution Name (RSSD ID) Location Total Assets
42 BOK FINANCIAL CORPORATION (1883693) TULSA, OK $19,363,601
43 W HOLDING COMPANY, INC. (2801546) MAYAGUEZ, PR $17,894,049
44 FIRST BANCORP (2744894) SAN JUAN, PR $17,596,317
45 WEBSTER FINANCIAL CORPORATION (1145476) WATERBURY, CT $16,964,451
46 SKY FINANCIAL GROUP, INC. (1071203) BOWLING GREEN, OH $16,807,287
47 FIRST CITIZENS BANCSHARES, INC. (1075612) RALEIGH, NC $16,012,041
48 COMMERCE BANCSHARES, INC. (1049341) KANSAS CITY, MO $15,531,107
49 NEW YORK PRIVATE BANK & TRUST CORPORATION (3212091) NEW YORK, NY $15,095,466
50 FULTON FINANCIAL CORPORATION (1117129) LANCASTER, PA $15,078,415
Appendix B
Excel Spreadsheet Results of Data Analysis
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Bank of America
Net Income 21,133,000 16,465,000 14,143,000
Total Assets 1,459,737,000 1,291,803,000 1,110,457,000
Return on Assets (ROA) 1.45% 1.27% 1.27%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 1.00% 0.95% 0.98%
Percentage Diference ROA-Weighted Avg. 0.45% 0.33% 0.30%
JPMorgan
Net Income 14,444,000 8,483,000 4,466,000
Total Assets 1,351,520,000 1,198,942,000 1,157,248,000
Return on Assets (ROA) 1.07% 0.71% 0.39%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 0.80% 0.59% 0.35%
Percentage Diference ROA-Weighted Avg. 0.27% 0.11% 0.03%
Citigroup
Net Income 21,538,000 24,589,000 17,046,000
Total Assets 1,884,318,000 1,494,037,000 1,484,101,000
Return on Assets (ROA) 1.14% 1.65% 1.15%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 0.84% 1.14% 0.90%
Percentage Diference ROA-Weighted Avg. 0.30% 0.51% 0.25%
Wachovia
Net Income 7,791,000 6,643,000 5,214,000
Total Assets 707,121,000 520,755,000 493,324,000
Return on Assets (ROA) 1.10% 1.28% 1.06%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 0.82% 0.95% 0.84%
Percentage Diference ROA-Weighted Avg. 0.28% 0.33% 0.21%
Wells Fargo & Co
Net Income 8,482,000 7,671,000 7,014,000
Scott J. W., Arias J. C. - Banking Proftability Determinants
James W. Scott, José Carlos Arias
Business Intelligence Journal - July, 2011 Vol.4 No.2
230 Business Intelligence Journal July
Period Ending 31-Dec-06 31-Dec-05 31-Dec-04
Total Assets 481,996,000 481,741,000 427,849,000
Return on Assets (ROA) 1.76% 1.59% 1.64%
Per Capita Growth Rate* 3.20% 3.70% 4.20%
Weighted Average 1.14% 1.11% 1.18%
Percentage Diference ROA-Weighted Avg. 0.62% 0.48% 0.46%
*Based on 2006 estimates from CIA World Factbook and bar graph data from International Monetary Fund
2011 231
THE STRATEGIC IMPACT OF THE BUSINESS DYNAMICS
IN EMERGING COUNTRIES ON CONTEMPORARY
PERSPECTIVES
Walter Gerard Amedzro St-Hilaire, Ph.D
Researcher on Strategy and Governance of Public Organizations in HEC Montréal
Researcher for the Research Center on the Governance of Natural Resources
Email: [email protected]
Abstract
The current competitive environment requires countries to make strategic choices regarding international trade which are not without
impact on their value regards to positioning. Therefore, it becomes important for business owners and for analysts to assess the impact
of policy decisions on global competitiveness and wonder if becoming a future “Giant” does not pass through a choice of sector for
export growth. Also how is the comparative advantage of the current Asian Dragon evaluated in the various sectors that underpin its
economy? In this paper, we propose a both strategic and economic analysis case of the new global competitive order based on the concept
of comparative advantage to fnd that China, by taking advantage of all its assets, has built a model which is typical while integrating
the new realities of contemporary global trade, occupying a prominent place in the global economic landscape. The present article goes
beyond this Asian dragon’s rather negative image to highlight its success in adapting to the reconceptualization of economic exchanges.
Key words: Strategy, competitiveness, added value, economic analysis, economic development.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Marking the 2000s by its ascension into world trade
1
,
and after surpassing, in 2008, the German level in
manufactured exports, China saw, in 2009, its share in total
world trade of goods exceeding the Germany
2
. From 1998
to 2008, the total Chinese share in world exports grew at
23% per year, twice the growth rate of total world exports.
If the trend continues, exports from China could represent
a quarter of the global total in 2018, surpassing the U.S.
in the 1950s that didn’t reach this fgure since totalizing
at the time, only 18% of world exports (as against 8% in
2009). However, the possibility of a decline of the Chinese
exports growth rate remains. According to IMF forecasts,
these exports are expected to represent 12% of world total
in 2010. Worth 9.6% in 2009, this share corresponded to
that of Japan at its peak, is 1986
3
. In addition, the IMF, in
a study published in 2009, indicated that a Chinese GDP
growth rate remaining at or above 8% and a growth still
heavily dependent on exports would mechanically raise this
dragon’s share in world exports to 17% in 2020. However,
the crisis of 2008-9 caused the share of exports in China’s
GDP to drop from 36% in 2007 to 24% in 2009.
In this regard, it is worth asking whether the future
growth of Chinese exports wouldn’t come from its exports
of high added value products (computers and vehicles)
rather than textile. How comparative advantage in this
country is evaluated the various sectors that underpin its
economy?
To address this issue of International Affairs, it will be
conducted a comparative analysis of this dragon with the
rest of the world by sector, and with the assumption that
it will, as regards the export growth, take the same way as
Japan that gradually exported products of increasingly high
added value. It will, in this article, be traced the concept
of “comparative advantage”, to then bring out a stark and
pertinent analysis of the results recorded by the Chinese
trade during the last decade.
Conceptual Framework
For this paper purposes, we are led to identify one
key concept necessary in understanding our theme: the
concept of comparative advantage. It is the ability of an
economic agent (individual, company, country) to create
a product or service for a lower opportunity cost than
1
Figures on China's place in global trade given below are from the article "China's
export prospects, Fear of the Dragon", The Economist, January 7, 2010.
2
In 1997, China was only ..... the 10th largest exporter! See the table of the main
exporters in 1997.
3
Since then, Japan’s share fell to 5%, partly because of the rising yen.
Business Intelligence Journal - July, 2011 Vol.4 No.2
232 Business Intelligence Journal July
that borne by other economic agents. For Paul Krugman,
it’s especially the ability to grasp the difference between
comparative advantage and absolute advantage that defnes
an economist
4
. If comparative advantage should not be
confused with absolute advantage, which is a special case of
comparative advantage, it should not be with the concepts
of free trade (international trade without tariffs and other
artifcial restrictions to trade), the gains from trade, or the
international division of labor which are distinct, though
related to that of comparative advantage. The latter is,
with returns to scale, the main determinant of international
trade
5
.
Although the idea of comparative advantage was exposed
by Robert TORRENS in 1815
6
, its authorship is attributed
to David Ricardo (1817), whose best known proof could be
found in his Principles of political economy and taxation.
Taking the example of England and Portugal, both of able
to produce wine and cloth, he shows that the comparison
of the production costs of wine to cloth (or cloth to wine)
in each country simultaneously reveals the comparative
advantage of each country. The ratio of production costs
is also known as relative cost, opportunity cost, or even
relative productivity. The two basic models of comparative
advantage are Heckscher-Ohlin’s and Ricardo’s.
The Heckscher-Ohlin-Samuelson model
Named after its three principal architects, the Swedish
economists Eli Heckscher Filip (1879-1952) and Bertil
Ohlin (1899-1979) and American economist Paul Samuelson
(1915-2009), the Heckscher-Ohlin-Samuelson model is
frequently referred to as the “HOS model”. But the name
4
" If there were an Economist’s Creed, it would surely contain the affrmations «
I understand the Principle of Comparative Advantage » and « I advocate Free
Trade ». For one hundred seventy years, the appreciation that international trade
benefts a country whether it is “fair” or not has been one of the touchstones of
professionalism in economics. Comparative advantage is not just an idea both
simple and profound: it is an idea that conficts directly with both stubborn
popular prejudices and powerful interests. This combination makes the defence
of free trade as close to a sacred tenet as any idea in economics” - Paul KRUG-
MAN, "Is Free Trade Passé?" by Paul R. KRUGMAN, Journal of Economic
Perspectives, vol.1, no.2, 1987, pp.131-44.
5
I like to begin classes on international trade by telling students that there are
two basic explanations of international trade. The frst is comparative advan-
tage, which says that countries trade to take advantage of their differences – a
concept that lay at the heart of Alan Deardorff’s beautiful, classic paper “The
general validity of the law of comparative advantage” (1980). The second is in-
creasing returns, which says that countries trade to take advantage of the inher-
ent advantages of specialization, which allows large-scale production – which
is what the “new trade theory” was all about" - Paul KRUGMAN, “Increasing
Returns in a Comparative Advantage World”, novembre 2009, 15 pages.
6
But the issue is controversial [see, for example, the article by Roy RUFFIN,
Debunking a Myth: Torrens on Comparative advantage]
of Wassily Leontief (1909-1999) should also be associated
with this model because the empirical test that he has done.
The study of this model requires, initially, understanding the
functioning of a “Heckscher-Ohlinian” economy in a closed
economy or “autarky”, and then comparing it to another
economy that is identical in all respects except for relative
factor endowments. This comparison aims to highlight the
direct dependence of the concept of comparative advantage
to both countries’ factor endowments: it is the heart of the
Heckscher-Ohlin “theorem”.
Indeed, stating that comparative advantage is determined
by differences in relative endowments between countries,
the HOS theorem is a direct consequence of the Rybczynski
theorem which allowed us to establish, under the HOS
model with two goods and two factors, that the higher the
capital-labor ratio of a country, the higher the supply curve
is shifted to the right. The proof of this theorem involves
the comparison of the autarkic equilibrium of two countries
with identical technologies and different production relative
factor endowments.
Let two countries, h, home country, and f, foreign
country when the country h has a capital-labor ratio higher
than the country f’s (k > k*)
7
. The h’s relative supply curve
is more right than f’s while their relative demand curves
merge, the two countries having the same utility function. It
follows that in absence of international trade, at equilibrium
of supply and demand of each country, the relative price of
good 1 is lower in country f than h while good 2 has lower
relative price in h than in f. In other words, the country h
has a comparative advantage in the production of good 2
while f harbors a comparative advantage in the production
of good 1.
It should be noted that, in this model, comparative
advantage is explained neither by differences in technology
(since, by defnition, production technologies are identical
from one country to another) nor by differences in taste
(preferences for goods are identical). The only reason why
relative prices before exchange are different lays in the fact
that h has a capital-labor ratio greater than f, which gives it
a comparative advantage in producing good 2. Conversely,
f has a higher labor-capital ratio than h, which gives it a
comparative advantage in producing good 1.
In sum, in our example, where k > k* and where, by
hypothesis
8
, production of good 2 requires relatively more
7
To distinguish the country f from the country h, an asterisk is added to the sym-
bols on the country f.
8
This hypothesis, like other HOS model assumptions is discussed in more details
athttp://www.mazerolle.fr/Economie-internationale/Glossaire-economie-inter-
nationale/Modele-2X2-standard-Hypotheses-centrales.pdf
2011 233
capital than labor, the Heckscher-Ohlin theorem can be
stated as follows: the country h, relatively abundant capital
(k > k*) has a comparative advantage in production of good
2 (since the production of good 2 uses capital relatively
intensively). As for the country f, relatively abundant
in labor, it has a comparative advantage in production
of good 1 (because the production of good 1 uses the
work of relatively intensively). It follows that at autarkic
equilibrium, the country h produces relatively more of
good 2 than f (conversely, the country f produces relatively
more of good 1 than h). Openness to trade will cause h
to specialize in production and export of good 2 and f’s
specialization in production and export of good 1. As this is
a model of just two countries, exports of h will go to f and
f’s exports will go to h.
Optimality of Free Trade in the Ricardo Model
On another hand, the RICARDO model bases itself
on usual concepts of microeconomic theory to show that
the specialization of each country in the production and
export of goods for which it has a comparative advantage,
improves well-being in each country. For this, it shows
that the international exchange based on the exploitation of
comparative advantage improves the purchasing power of
wages in both countries. Indeed, the comparison between
autarky and free trade shows the overall superiority of
the purchasing power of wages in conditions of free trade
relative regards to the purchasing power of wages in the
autarkic economy.
For example, RICARDO considers two countries,
namely England and Portugal, and two goods, wine and
cloth, from which he evaluates the purchasing power of the
hourly wage for these two products in autarkic economy,
then where free trade occurs. In England, the hourly wage
yields 0.01 hl of wine for free trade instead of 0.008 hl
in autarky. Regarding cloth, the purchasing power of
wages is the same in autarky and free trade (0.01 m2). In
Portugal, the hourly wage yields 0.0125 m2 of cloth in
free trade instead of 0.011 m2 in autarky. As regards wine,
the purchasing power of wages is the same in autarky and
free trade (0.0125 hl). We can therefore conclude that free
trade improves both the English and Portuguese overall
purchasing power of wages.
The Theory of International Trade
In regard to the model that was previously shown, it is
important to understand the concept of exchange, including
international exchange, distinct from that of comparative
advantage, though linked to it. Based on the concept of
exchange among economic agents, the positive theory of
international trade of David Ricardo, developed mainly
in the microeconomics feld, is rooted in the ideas of
neoclassical economists, particularly in the theory of general
equilibrium including consideration of the phenomenon
of exchange without production, called “pure trade” and
the exchange with production. Insofar as the exchange
becomes international, it fts into the broader framework
of international economic relations. International Economy
could be seen as an observable reality that should be
described from a statistical point of view and facts: the
economic exchanges between the nations are steadily
growing and being an essential component of globalization,
while as trade policy of countries. In this latter sense, it
is often the terms “international trade” or “international
trade”, which are used.
Methodology: Content Analysis Linked
to the Index of Competitiveness of the
World Economic Forum
The study of countries’ growth factors will be based
on an analysis of the competitiveness index published
annually by the World Economic Forum in Davos. This
index is a score assigned to different countries varying, in
2009, between 2.58 (Burundi) and 5.6 (Switzerland), from
which countries are ranked. This classifcation is meant to
refect the competitiveness of countries, which is evaluated
by 12 criteria (the 12 pillars of competitivity
9
) defned by
the WEF and also by the interaction of these criteria allow
in fact distinguishing three categories of economies.
The Three Stages of Development According
to the Wef
It must indeed be noted that the 12 factors or pillars
of competitiveness defned by the WEF interact and even
mutually reinforce themselves. For example, innovation
(12th pillar) is inconceivable in a world where powerful
institutions (Pillar 1) are absent, because starting innovation
requires that a protection by copyright IP. Similarly, if the
workforce is not educated (5th pillar), or the different
9
See the notice "Les 12 piliers de la compétitivité selon le World Economic Fo-
rum",http://www.mazerolle.fr/Economie-internationale/Glossaire-economie-
internationale/12-piliers-de-la-competitivite-selon-le-World-Economic-Forum.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
234 Business Intelligence Journal July
An Emerging Giant in Global
Competitiveness
According to the WEF latest report of (Global
Competitiveness Report 2009-10), this Asian giant’s Global
Competitiveness Index 2009-10 is 4.74 out of 7, giving it
the 29th position out of 133 countries, an honorable and
progressive position. Below is China’s ranking with more
details on each of the 12 pillars:
Represented by a so-called “spider” graph and
compared with the average performance of category 2
countries (economies whose competitiveness is determined
by effciency), this ranking shows that the dragons’ country
does as well or better, especially as regards the market size
and macroeconomic stability.
markets not effcient (pillars 6, 7 and 8) or there is no
infrastructure (pillar 2), innovation will not be possible
or will be very limited. From these interactions emerges a
classifcation of countries according to factors that infuence
their competitiveness. These economic classes defne the
countries’ development level, and by extension the role
of the 12 pillars. Thus, improving the competitiveness of
Chad does not go through the same channels as in Japan
or the United States, these countries not being at the same
stage of development. As the latter evolves, the level of
wage increases, and thus increases the income per head
of population, hence the need to improve productivity to
maintain the living standards. Three types of economies
can be distinguished, including economies whose
competitiveness is under the infuence of their resources in
production factors, those whose competitiveness is under
the infuence of effciency and those whose competitiveness
is infuenced by innovation.
• Economies whose competitiveness is determined by
their resource inputs. These are economies whose
competitiveness is sustained by their resources
in production factors, mainly unskilled labor and
natural resources. Companies in these countries
sell commodities, manufactured by unskilled labor
and therefore not well paid. The maintenance
and enhancement of competitiveness at this stage
of development essentially requires the proper
functioning of public and private institutions (pillar 1),
the effciency of infrastructure (pillar 2), the stability of
the macroeconomic environment (pillar 3) and healthy
workforces a basic education (pillar 4).
• Economies whose competitiveness is determined by
their effciency. As wages rise, as production methods
become more complex and product quality improves,
competitiveness is mainly governed by the workforce
qualifcation and vocational training (pillar 5), effcient
goods and services market (pillar 6), labor market
(Pillar 7) and capital markets (pillar 8), and by the
domestic and / or export market size (pillar 10) and
technological agility (pillar 9).
• Economies whose competitiveness is determined by
innovation. The economies that fall into this category
are those where wages are high, even very high, as well
as living standards, and where the main competitive
factors become innovation (pillar 12), which takes
the form of knowing how to make new and different
products, as well as a dense and advanced organization
intra and inter frms (pillar 11).
Result and Analysis
Still ranked by the WEF in 2008-9 in an intermediate
stage between stages 1 and 2, according to the GDP per
capita criterion (in current dollars not in PPP dollars), China
is now ranked among countries of group 2 with its GDP per
capita in current dollars over $ 3,000.
GDP threshold for membership in a stage of development
Development stage GDP per capita (in current dollars)
Stage 1: Economies whose
competitiveness is determined by
their resourse inputs
17000 dollars
2011 235
However, economic development does not take place
smoothly. In this regard, the chart below highlights the
most problematic aspects of business:
Competitive Advantage and Competitive
Disadvantage
The small square appearing next to the China’s rank
for each indicator shows a situation of either competitive
advantage (dark blue) or competitive disadvantage (light
gray). For determining the nature of competitiveness (the
color of the square), the WEF proceed in accordance to the
position of countries in the global CGI index. Indeed, those
classifed in the top 10 have a competitive advantage for an
indicator when it also has a rank lower than 10. Otherwise,
it is considered a competitive disadvantage. Furthermore,
countries at ranks higher than or equal to 11 (it’s the
case of the Asian dragon that is 29
th
) have a competitive
advantage for an indicator when its rank is lower than the
country’s global ranking, otherwise there is a competitive
disadvantage for the indicator. Indicators ranked below 29
have therefore a dark blue square to indicate the competitive
advantage of the dragons’ country.
Some Examples of Expanding Industries
If the Asian giant is an emerging economy, relatively
new to the world trading scene, it does however have
many assets to shine on it and make the longer-established
countries worry. Indeed, expanding in many sectors, the frst
country in the world in terms of market size is participating
in reshaping world trade, particularly by cutting itself a
special place.
The Aviation Industry
Until the late 2000s, with the U.S. McDONNELL
DOUGLAS that disappeared in 1997, and the Brazilian
EMBRAER, BOMBARDIER the Canadian, Russian and
Japanese Mitsubishi SUKHOI limited to regional aircraft
and having no real infuence on neither of the Toulouse and
Seattle giants’ strategies, airliners manufacturers competing
with Boeing and Airbus were rare. However, AVIC, the
newcomer from the land of dragons, supported by this
land’s 11th Five-Year Plan (2006-2010), could change the
deal, and make the two global giants worry, especially by
the planned launch of a medium-haul with 150 seats, thus
competing with the two giants much earlier than expected.
Next to AVIC, Bombardier, with its CSeries, also
becomes a potential competitor:
“Annoncé pour une capacité standard de 130 sièges, il
pourrait ainsi faire de l’ombre à l’A318, le plus petit des
Airbus, dès 2013. Sukhoï, avec son nouveau Superjet et
Embraer avec son E-190 affchent également leur ambition
sur ce segment de marché. Mais la principale menace
est le futur appareil chinois C919, dont le premier vol est
annoncé pour 2014. D’une capacité de 170 à 190 sièges, ce
futur rival de l’A320 et du B737 pourra lui aussi compter
sur le dernier cri en matière de réacteurs et d’équipements.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
236 Business Intelligence Journal July
Bien que très jeune, l’industrie aéronautique chinoise
a en effet tous les atouts. Outre ses centaines de milliers
d’ingénieurs, elle peut compter sur les moyens fnanciers
colossaux de l’Etat et sur un marché national en plein
boom, susceptible d’absorber quelque 2.800 avions neufs
d’ici à 2026. Aucun de ses concurrents n’a autant de cartes
en main. Bombardier l’a d’ailleurs bien compris, qui a
conclu un partenariat stratégique avec l’industrie chinoise,
lui offrant son expertise et son implantation internationale
pour le C919 en échange d’un coup de pouce fnancier
pour son Cseries”
10
. It should also be noted that with the
recent installation in Tianjin of the A320 frst assembly line
outside Europe, Airbus aims to take a part in the largest
market of world
11
.
Satellite Positioning System
In addition to the aviation industry, the Asian giant
cuts itself a place in the satellite space by launching the
building of its own system, called “Beidou”, whose release
is scheduled for 2015. Direct competitor of the European
Galileo project, it has more than 30 satellites, fve of
which have already been installed. Concerned, Europe,
which has fallen behind in the Galileo project, denounced
the fact that China is currently pre-empting the frequency
allocated to the European Union by the International
Telecommunication Union. Indeed, China has published
frequencies and signals that overwrap those of Galileo.
However, the strongest signal “crushing” the others, some
of Galileo signals could be jammed by the Chinese satellites
and made non-functional, especially the encrypted signals
reserved among others to Defense
12
.
High Speed Trains
In terms of land transport, land of dragons is setting
up travel at high speed. Indeed, although the French
Alstom denied it the transfer of its technology, the Chinese
TGV was nevertheless established, namely the Wuhan-
Guangzhou line (departure from the new Guangzhou
Railway Station), where circulate Nippon and German
13
See « La Chine s’affrme comme un acteur majeur de la grande vitesse ferrovi-
aire », Les Echos, June 21, 2010.
14
See “Deux oiseaux dans les télécoms chinoises”, by Philippe ESCANDE, Les
Echos, April 7, 2010.
designed TGV(derived from the Japanese Shinkansen and
the German Velaro) reaching peaks of 350 km / h (cons
320 maximum in France). He still benefted from the SNCF
expertise in station construction. Now, China intends to
sell its know-how on TGV to Saudi Arabia and also plans
to build, by 2020, 45,000 km of high-speed lines, widely
outdoing the French 1000km of high speed lines
13
.
Mobile Phone Industry
Regarding the mobile sector, it is in February 2010 that
the Asian giant passed the milestone of one billion users
with a mobile operator market controlled entirely by local
businesses, which do well on the OEM market too.
An article in “Les Echos” focuses on the two giants
China Telecom, ZTE and Huawei
14
that, like many others,
are typical of the rise of the Asian Dragon’s economy and,
more signifcantly, follows the curve of its evolution. In the
early 1980s, Ren Zhengfei, , a researcher in the laboratories
of the army in Beijing and founder of Huawei, and Hou
Weigui, ZTE’s founder, working in Xian in a research
center in electronics, adopted a similar strategy to win :
“Face à un marché balbutiant en Chine et des acteurs
européens comme Alcatel, Siemens, Nokia ou Ericsson,
qui trustent les relations avec les grandes sociétés d’État,
les deux nouveaux venus se sont lancés à l’assaut des
campagnes chinoises totalement sous-équipées, avec
une arme, l’engouement pour le téléphone mobile. Et un
argument massue, le prix. Quand le fournisseur européen
vendait son matériel de 200 à 300 dollars la ligne, Huawei
et ZTE séduisaient les PTT de troisième zone avec leurs
matériels trois fois moins chers”.
They then repeated this low cost strategy in Pakistan,
Burma and Africa, with this incredible feature, inspired by
the postwar Marshall Plan, i.e. fnance the purchase of their
own products with the help of the Chinese Government to
these countries, and now they are entering India, the old
Europe and the strong America. Indeed, something to give
cold sweat to Alcatel that, a decade ago, reigned supreme
but must now be content with the bare minimum:
“Ainsi, lors de son dernier appel d’offres en décembre
dernier pour son nouveau réseau au protocole Internet, le
premier opérateur du pays, China Mobile, a en quelque
sorte livré une photo du nouvel équilibre des forces. Selon
10
Bruno TREVIDIC, Le troisième avionneur mondial sera-t-il chinois ? Les
Echos, December 7, 2009.
11
Seehttp://www.mazerolle.fr/Economie-internationale/Glossaire-economie-in-
ternationale/IDE-francais-en-Chine-et- chinois-en-france.pdf
12
But Beijing was involved in the Galileo project to the tune of 65 million euros
and a portion of the work was entrusted to Chinese companies. Despite this
cooperation, the Chinese government has developed a competing system.
2011 237
15
The price-to-book ratio, or P/B ratio, is a fnancial ratio used to compare a com-
pany’s book value to its current market price. Book value is an accounting term
denoting the portion of the company held by the shareholders; in other words,
the company’s total tangible assets less its total liabilities.
16
Clive COOKSON, China scientists lead world in research growth, Financial
Times, January 25, 2010.
les analystes d’iSuppli, il aurait accordé 35% du marché
à ZTE, autant à Huawei, 10% à Alcatel- Lucent et 5 % à
Ericsson. On comprend mieux les problèmes d’Alcatel, qui
dix ans avant était le leader incontesté de ce marché”.
Banking
The fnancial sector does not fall behind, given the vast
expansion of banks in the country of dragons since the
2000s. As a proof, in 2010, four of the top fve banks in the
world are Chinese according to their book to market ratio
15
.
Scientific Research
Regarding the scientifc research, the Asian giant is
experiencing a continuous boom since the 1980s, 1990s and
2000s, ranking second in the world behind the United States,
according to the Financial Times
16
. Although the indicators
of this development are not necessarily very illuminating
and reliable, the newspaper bases itself on the research
articles published in scientifc journals criterion to make
this classifcation. Since 1981, the number of publications
has been multiplied by 64 reaching, in 2008, no fewer than
112,318 articles published by Chinese researchers (against
332,916 in the U.S.).
The Financial Times quoting James Wilsdon, director of
the Science Policy Centre of the Royal Society, shows three
factors thought to be behind this growth, namely the huge
public investment in education and research at all research
levels, in both schools and universities, the fast and well
organized circulation and dissemination of scientifc
information and the set up by the Chinese government of
very attractive opportunities for researchers in the Diaspora,
allowing them to teach in their country, while continuing to
spend part of their time abroad.
Steelmaking
Concerning heavy industry, steel production, considered
as an indicator of the power of a national industry, because
of the use of this material in the composition of a large
number of industrial products, brand the supremacy of the
Asian giant, by far the largest producer of steel. Moreover,
the Chinese Hebei Steel and Bao Steel are two of the fve
largest producers of steel in the world (see graphic).
Multiple Assets: Acquisitions, Counterfeiting,
Espionage and Authoritarian Planning ... But
Also Tourism Counterfeiting, the Cybersitter
Case Example
To reach the level of the greatest in areas where
technology is usually a hurdle, China does not hesitate
to use the counterfeit, fast and inexpensive, to build
its competitiveness in high technology industries. The
example of the Green Youth Escort Dam, complex
software tracking keywords to prevent any connections
to websites deemed politically sensitive
17
is typical of the
use of counterfeiting at the State’s highest level. In 2009,
imposed on all manufacturers of computers working on the
world’s largest market (since June 2009, the installation is
no longer mandatory but only recommended), this software
is actually derived from the piracy of a parental control
software belonging to American CYBERSITTER, based
in Santa Barbara, California. The company attacked, in
17
The Green Dam completes the censoring system already imposed on the Internet
in this country. It prevents access to websites reporting on the repression of the
Tiananmen place riots, the Tibet crisis or the Falungong organization activities.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
238 Business Intelligence Journal July
Taken not long ago as nasty polluters, as gravediggers
for the Copenhagen Summit of December 2009, the Asian
giant is now poised to overtake the objectives of the Summit.
In fact, while in France the carbon tax, is sacrifced for
economic growth, the land of dragons carved a very large
place on the wind turbine market. Furthermore, world’s
fourth solar panel manufacturer, the company Yingli Solar
produces good quality panels for a price 30% cheaper,
bought by the biggest companies in the world, including
GDF Suez, which has ordered 145,000 of them to build a
large solar power plant in southern France. We must note
that four of the top 10 solar panels makers are Chinese:
Suntech (2nd), Yingli (4th), Trina Solar (6th) and Solarium
(8th).
It should be noted, however, that the largest population
of the world remains one of the biggest polluters in the
world, although progressing in the right direction while
many others, such as France, although prone to criticism,
go down. And, indeed it was in June 2010 that Beijing
launched its subsidy plan for electric cars in fve test
cities
18
. However, it is likely that these subsidies are not
18
Dans un programme d’aides comparable à ceux expérimentés en Occident et au
Japon, le ministère des Finances a annoncé qu’il allait offrir aux habitants
des villes de Shanghai, Shenzhen, Hangzhou, Hefei et Changchun un rabais
allant jusqu’à 50000 yuans , soit 5800 euros lors de l’achat d’une automobile
hybride rechargeable et de60000 yuans (7200 euros) pour chaque inves-
tissement dans un véhicule tout électrique». This targeted strategy aims to help
local manufacturers, which are located differently in these fve cities: SAIC in
Shanghai, BYD is in Shenzhen, Geely is in Hangzhou, Chery is in Hefei and
FAW in Changchun.
suffcient for the demand of electric cars to take off if they
are not accompanied by massive public investment in
infrastructure adequate for their use (battery management
terminals or loading points, or battery exchange stands).
Goods and Services
Comprehensive and systematic data on world trade are
known about 18 months apart from the current date. In
2009, for example, fgures for 2008 are published around
the middle of the year in the World Trade Report 2009 of
the WTO. The frst data to be published are those of the
classifcation of countries according to their total exports
and imports. An early circulation of data is possible, but
it carries the risk of a wrong recycling of data from the
previous year exploiting the difference between the date of
publication and the time of data.
Goods International Trade In 2009
After studying the Chinese economy through its business
activities, it is important to consider it on a bigger feld:
International trade. It requires noting that the landscape of
global trade depends on how the EU-27 is processed in the
distribution of trade fows, because of the centrality of the
EU. Thus, should be considered the case in which the main
goods exporters and importers include the EU countries as
individual entities, then the case of the main goods exporters
and importers excluding intra-EU trade. In the table below,
each country of the EU-27 is treated separately. Several
interesting points emerge
19
:
• In 2009, China became the world’s largest exporter,
overtaking Germany.
• Three major powers are clearly distinguishable from
others, as regards both exports and imports: China,
Germany and the United States.
• Regarding China, it should be noted that a signifcant
portion of Chinese exports are actually produced by
Western and Japanese multinational frms operating in
the largest market in the world.
• The role of France in international trade, though
modest, is signifcant mainly as a consumer-importer
(6th exporter, importer 4th)
• The bulk of world exports and imports is the result of a
small number of countries.
19
Source: WTO, World Trade Report 2010
An extraordinary and unexpected development in the
environmental sector
Five keys to understand the Solar Sector
• Germany represented more than half of the world´s photovoltaic systems
in 2009.
• Down last year, the global market should have its growth back in 2010,
with an increase of the sales volume worth at least 40% according to
experts.
• The actual production capacities (around 17 Gig watts) are already able
to cover market´s needs till 2012-2013, according to PwC.
• U.S. and China are considered as the next growth relay of this sector.
• This sector employs 5500 people in France. According to the renewable
energies Union, it could have around 15000 employees the year 2012.
U.S. court, (the communist government of the country of
dragons totally controlling the local judiciary) seven PC
manufacturers and two local designers for hacking 3000
code lines. This has not been to slow the Asian giant, which
plans to export its product.
2011 239
• The total of world exports and imports do not coincide
due to signifcant statistical error margins.
Dealing instead with the EU as a whole and treats
the intra-European as intra-regional trade, such as trade
between California and Texas in the United States, or
between Guangdong and Sichuan in China, draws a quite
different picture
20
:
• The EU-25 then appears as the leading exporter (unless
you add China, Hong Kong and Taiwan).
• The U.S. remains the largest importer before the EU
and the Asian giant.
• The disappearance of trade between EU countries
“reduces” the total world trade and therefore increases
the share of major countries. For example, the dragons’
country’s share is 11.8% (excluding Hong Kong and
Taiwan).
• In both cases, we note that:
• Countries like India and Russia are also major emerging
economies; however, they remain far behind the world
three major goods exporters and importers.
• Japan remains a major exporting power, and it is more
apparent if we look at the chart excluding intra-EU trade,
but it was surpassed by China in a decade, refecting
the speed of the transformations characterizing global
economy.
The Dominance of Intra-Regional Over Inter-
Regional Trade
Like the EU, “the regional blocs” are very important in
the assessment of international trade, since, in general, the
largest share of world trade takes place within the blocks,
especially within the EU.
In addition to goods, there are, on the international
market, a sizeable fow of other types of properties, less
tangible than the frst, namely services. It is therefore
important to understand well the services international
trade, given their increasingly important role in an economy
more and more globalized. In the Balance of Payments
Statistics, the various parts of the current account are titled
respectively: Goods, Services (including services provided
or received by the government), Income (investment
income and wages) and Current transfers (operations
without consideration).
20
Idem
Not existing as such, the commercial services category,
defned by the WTO “as equal to services minus those
provided or received by public administrations”
21
, is
divided into three categories: transportation, travel and
other commercial services.
Transportation category. Covering all transport
services (sea, air and others, including land and inland water
transport, through space and pipeline) offered by residents
of an economy to those of another, it concerns transporting
passengers, goods carriage (freight), rental (charters) of
carriers with crew and other linked services.
Travel category. Covering goods and services purchased
for personal use by travelers - for health, educational or
other purposes - as well as those who travel for professional
reasons, it does not involve a particular kind of service,
but an assortment of goods and services “consumed” by
travelers. The most common entries in this category are
housing, food and beverages, entertainment, transportation
(within the economy visited), gifts and souvenirs.
Other Commercial Services
• Communication services: they include the
telecommunications, postal services and courier
services. Telecommunication services include the
transmission of sound, images or other information by
telephone, telex, telegram, radio and cable television
broadcasting, satellite, electronic mail, fax etc..,
including communications networks, teleconferencing
and support services, but not the value of the transmitted
information. Are also included mobile telephone
services, Internet services and basic online access
services, including the provision of Internet access;
• Building and public works services: they include work
performed on construction and installations projects by
employees of a company outside the economic territory
(the one-year rule used to determine the resident status
is to be applied with fexibility). Furthermore, the
goods used for these projects are included, involving a
tendency to overestimate these services.
• Insurance services related to various forms of insurance
to non-residents by resident insurance companies and
vice versa, as freight insurance, direct insurance (eg life
insurance) and Reinsurance;
• Computer and information services including data
services (services related to hardware and software and
21
Detailed classifcation source: WTO metadata,http://www.wto.org/french/res_f/
statis_f/its2007_f/its07_metadata_f.pdf pages 161-2.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
240 Business Intelligence Journal July
data processing services), the news agency services
(provision of information, pictures and articles to
the media) and other information services (services
database and web search);
• Royalties and license fees, including payment and
receipts pertaining to the export of non-fnancial
intangible assets and property rights such as patents,
copyrights, trademarks, industrial processes, and
franchises;
• Other services to enterprises, namely trade-related
services, operational leasing (leasing without
operators), and miscellaneous business, professional
and technical services such as legal, accounting and
management consulting, public relations services,
advertising, market research and opinion polling,
research and development services, architectural
services, engineering and other technical services,
agricultural services, mining and processing on site.
• Personal, cultural and recreational services subdivided
into two sub-categories:
• Audiovisual services covering services and commissions
relating to the production of flms, radio and television
broadcasts and musical recordings.
• Other cultural and recreational services which cover,
among other services associated with museums,
libraries, archives and other cultural activities, sports
and recreation.
When providing services to foreign countries is done
through the establishment of a subsidiary or a branch, it
is excluded from international trade fows measured by
the balance of payments, hence an under-estimation of the
importance of the commercial services provided by foreign
frms to a country. This is highlighted by data on foreign
direct investment in companies operating in the services
sector showing an absorption in this sector of at least 50-
55% of total foreign direct investment with, as a dominant
means of service providing, the establishment of foreign
subsidiaries.
In this regard, marking a major turning point in
international trade, the strategy initiated by Deng in
1978, generally considered to have borne fruit, despite
the signifcant imbalances and the major challenges
(environment, population, income disparities, and
geographic inequalities) to meet in the coming decades,
hard to explain. However, in an article in the Financial
Times, Gideon Rachman writes:
“The background to Deng Xiaoping’s liberalisation of
the Chinese economy in 1978 was a fscal and foreign
22
Bankruptcy could be good for America, Financial Times, January 11, 2010.
23
See Nicolas BAVEREZ, “La bataille des Normes”, Le Monde de l’économie,
April 20, 2010.
exchange crisis. Finding itself desperately short of cash,
the Chinese government was much more willing to embrace
heterodox economic ideas that promised to deliver faster
growth and higher revenues. The rest is history”
22
It would thus be the realization that, during a severe
crisis, the former communist regime was at its loss, which
made the land of dragons to change course in addition to
having a capable man in command.
In 2009, despite criticism of its recovery plan, this
country surprised the world by the effectiveness of its
response to the moderate decline in exports to the United
States and Europe, ensuring a record growth rate of 8.5%,
triggering a good recovery of its exports after a relatively
short period of very strong decrease.
Done in favor of domestic demand, but more towards
the infrastructure demand than in direction of consumer
demand, the recovery plan has given birth to a fear of
overcapacity. In particular, the generosity of banks in
granting loans to enterprises supported by the CPC (on
its request) has generated signifcant growth in the money
supply.
Moreover, the latter continues to grow with the purchase
of dollars from the products exported by the Bank of China
to maintain the fxed parity. However, there is doubt about
the “sterilization” of this currency (i.e. the withdrawal of an
equivalent amount of traffc to avoid excess liquidity) that
is not very likely, given the highly expansionary monetary
policy. The reserve requirement imposed on banks may
well have been raised (and is already among the highest in
the world, approximately 16%), this important liquidity, not
necessarily quite appropriately invested causes imbalances.
Since the entry of the Asian giant in the WTO in 2001
and especially since his claim as world’s top exporter
in 2009, thanks to the global crisis, some no longer
hesitate to speak of the “Beijing Consensus” to describe
the emergence of a new world economic order “fondé
sur la dissociation de la liberté politique et de la liberté
économique, la restriction de l’État de droit, le contrôle de
l’information et de la société, le pilotage du développement
et des investissements par la puissance publique, la gestion
discrétionnaire du crédit et du change”
23
. This term, coined
in 2004 by Joshua Cooper Ramo, is quoted in a recent book
by American scholar Stefan Halper, in its title: The Beijing
Consensus, How China’s Authoritarian Model Will
Dominate the Twenty-First Century.. However, in the land
2011 241
24
Title of an article published in « The Economist », May 6, 2010.
of dragons, where it appears that “The Beijing Consensus
is to keep quiet”
24
, one remains quite discreet about this
concept even hesitant to talk about “a Chinese development
model”, despite various publications such as”China Model:
A New Development Model from the Sixty Years of the
People’s Republic” (November 2009), or “China Model:
Experiences and Diffculties”(January 2010).
Conclusion
At the end of this article about the strategic impact of the
business dynamics in emerging countries on contemporary
perspectives and its questions about International Affairs,
it should be drawn some useful lessons. Indeed, China, the
largest market in the world, occupies a prominent place
in the global economic landscape. Booming, thanks to
the expansion of its various sectors, this Asian giant has
come to upset the world trade with a growing interest in
high technology. Taking advantage of all its assets (a
large population, a strong regulatory culture, etc...) and no
inhibitions (counterfeit) to have a place beside or in front
of the previously dominant economies, this has something
to be feared by those economies, as well in the aviation
industry as in Steelmaking and banking. Beyond its rather
negative image (communist, repressive and polluter), the
Asian dragon has built a model which is typical while
integrating the new realities of contemporary global trade.
While meeting the challenge of modernity, it succeeded in
adapting to the reconceptualization of economic exchanges
which are often defned today, so much broader than simple
barter of property against another, as envisaged in the theory
of exchange. Exchanges of this country with the world in fact
include not only goods and services but also international
capital fows, technology transfers, the movement of
ideas, economic migration, research and development as a
determinant of international competitiveness of enterprises,
etc...: from this point of view it is important not to isolate
this giant, frmly committed to shape its destiny, with too
unilateral visions, partial views but proved by the linear
reality of its winning evolution recursion, because this
country has specifc properties that exceed the global
business environment.
References
SMITH, A., 1776. Recherche sur la nature et les causes
de la richesse des nations. Paris : Gallimard, 1976.
BALASSA, B. 1963. « An Empirical Demonstration
of Classical Comparative Costs Theory», Review
of Economics and Statistics, volume 40, pages 231-
8.
Banque mondiale, Online Atlas of the Millenium
Development Goal,http://devdata.worldbank.org/
atlas-mdg/ CCAVES.
Richard E., FRANKEL, J. A. et JONES, R. W.,
Commerce et paiements internationaux, 9ème
édition.
Banque Mondiale, Rapport sur le développement
dans le monde. Publié chaque année vers le mois
d’octobre.
Banque Mondiale, site internet :http://www.
worldbank.org/
KAWECKI, B. et A. 2000. Économie Internationale.
Paris: Dunod, collection « Travaux dirigés », pages
25-65.
GRUBEL H. G. et LOYD, P.J., 1975, Intra-industry
Trade, the Theory and Measurement of International
Trade in Differentiated Products. Londres :
CIA World Factshttps://www.cia.gov/library/
publications/the-world-factbook/
Classement des 100 premières multinationales
et nombreuses informations utiles : http://
www. u n c t a d . o r g / Te mp l a t e s / St a r t Pa ge .
asp?intItemID=2068&lang=2 D
DEARDORFF, A.V., 1985. « Testing Trade Theories
and Predicting Trade Flows », in R. JONES et
P. KENEN, éditeurs, Handbook of Intemational
Economics, Amsterdam : North Holland. Volume
2, pages 467-518.
Conférence de Nations Unies pour le Commerce et
le Développement (CNUCED) : Publication d’un
rapport annuel sur l’Investissement International et
l’activité des frmes multinationales.
De MELO, J., et GRETHER, J-M., 1997. Commerce
International. Bruxelles : DeBoeck, collection
«Balises», pages 190-238.
Fond Monétaire International,http://www.imf.org/
external/french/index.htm FRIEDMAN,
Douglas A., Against the Tide: An Intellectual History
of Free Trade, 1996.
Amedzro W. G. - The Strategic Impact of the Business Dynamics in Emerging Countries on Contemporary Perspectives
Walter Gerard Amedzro St-Hilaire
Business Intelligence Journal - July, 2011 Vol.4 No.2
242 Business Intelligence Journal July
KRUGMAN, P., 1980. « Scale Economies, Product
Differentiation, and the Pattern of Trade ».
American Economic Review , 70, pages 950-959.
KRUGMAN, P., 1995. « Increasing Returns,
Imperfect Competition and the Positive Theory of
International Trade », in Handbook of International
Economics, volume 3, edited by GROSSMAN
Gene, M. et Kenneth ROGOFF. North Holland.
LANCASTER, K., 1980. « Intra-Industry Trade under
Perfect Monopolistic Competition », Journal of
International Economics, mai 1980, pages 151-75.
LASSUDRIE-DUCHENE B., 1971. « La demande de
différence et l’échange international » , Economies
et sociétés.
LEONTIEF, W., 1956. « Factor Proportions and the
Structure of American Trade: Further Theoretical
and Empirical Analysis», The Review of Economic
and Statistics, novembre, pages 386-- 407.
LINDER, S., 1961. An Essay on Trade and
Transformation. New York, John WILEY and Sons.
LEONTIEF, W., 1954. « Domestic Production and
Foreign Trade the American Capital Position Re-
examined», Economia Internationale, n°1, février,
pages 3-32.
Mac DOUGALL, G.D.A. 1951. «British and American
Exports : A Study Suggested by the Theory of
Comparative Costs, part 1 ». Economic Journal.
Volume 61, décembre, pages 697-724.
Mac DOUGALL, G.D.A. 1952. « British and
American Exports : A Study Suggested by the
Theory of Comparative Costs, part 2». Economic
Journal . Volume 62, septembre, pages 487-521.
HELPMAN, E. et KRUGMAN, P., 1985. Market
Structure and Foreign Trade. MIT Press.
Mathematica Eurostat, page d’accueil du portail
statistique de l’UE Eurostat, page d’accueil des
statistiques de l’UE sur le commerce international.
Mc KENZIE R., «Comparative Advantage», Vidéo
You tube,http://fr.youtube.com/watch?v=aLe_
LF622JI , Mars 2007. O
Organisation de Coopération et de Développement
Economique (OCDE), Portail des échanges
internationaux:http://www.oecd.org/topic/0,3373,
fr_2649_37431_1_1_1_1_37431,00.html
Organisation Mondiale du Commerce, site internet :http://www.wto.org/index.htm
KRUGMAN, P. et OBSTFELD, M., Economie
internationale, 7ème édition.
KENEN, P. B., 1965. « Nature , Capital and Trade »,
Journal of Political Economy, 73, octobre, pages
437-60.
Programme des Nations-Unies pour le Développement,
site internet :http://www.undp.org/french.
RICARDO, D., 1817. Des principes de l’économie
politique et de l’impôt. Paris Flammarion, 1977.
RYBCZYNSKI, T. M., 1955. « Factor Endowments
and Relative Commodity Prices», Economica,
pages 336--41.
SAMUELSON, P. A., 1948, « International Trade and
the Equalization of Factor Prices », The Economic
Journal , juin, pages 181-97.
STOLPER, W. F. et SAMUELSON, P. A., 1941. «
Protection and Real Wages», Review of Economic
Studies, 9, novembre, pages 58-73.
2011 243
E-BANKING PATRONAGE IN NIGERIA: AN EXPLORATORY
STUDY OF GENDER DIFFERENCE
Dr. Asikhia Olalekan
Senior Lecturer, Department of Business Studies
College of Business and Social Sciences,
Covenant University, Ota, Canaan land,
Km.10, Idiroko, P.M.B.1023, Ota,
Ogun state, Nigeria.
Email: [email protected]
Abstract
Although some studies had already been done concerning electronic banking yet there has been no study on gender differences
in e-banking patronage. The aim of this study is to determine if gender difference exist in electronic banking patronage in Nigeria.
The major attributes of electronic banking are used as measuring items using likert scale, which is subjected to factor and descriptive
statistical analysis. The study includes a survey on gender patronage of e-banking by individual customers in Abuja and Lagos, Nigeria
conducted in December, 2008. The results of the study indicate gender differences in e-banking patronage and also customer satisfaction
is found to be low in this respect which calls for improvement in service delivery in terms of timeliness and consistency in standards.
Key words: Electronic banking, Patronage, Gender differences, Nigeria.
There has been transformation of the Nigerian banking
industry after the consolidation exercise in 2005 by the
Central Bank of Nigeria (CBN). Akpan (2009) asserts that
maximizing returns and optimizing proftability became
the focus of banks and these can only be achieved through
enhanced patronages; that is increased customer base with
attendant satisfaction suffcient to consolidate loyalty.
The banks are therefore confronted with delivering their
services in the most effcient ways, using electronic means
to deliver additional products and services. Thus, managing
their assets of service delivery to customers became a major
objective.
Electronic banking is the effective deployment of
information technology by banks. Jun and Cai (2001)
identifes three dimensions of service quality in E-banking,
they are banking service product quality, customer service
quality and online systems quality. Bank service product
quality is related to product variety and the diverse features
of the service products. Customer service quality is related
to the differences between customer expectations of service
providers’ performance and their evaluation of the services
they received. Online systems quality is the quality that
the customer perceived as the end-users of the information
system. They further identify the various dimensions of
electronic banking that may be perceived differently by
male and female customers, which may likely affect their
level of patronage; this involved the level of perceived
risks, user input factors, price factors, service product
characteristics, and individual factors.
Recent literature has established that communication,
information sharing, distant in electronic means, and
level of perceived usefulness and innovativeness are
major setbacks in E-banking acceptance and patronage by
customers (Mukherjee and Nath, 2003; Rotchanakitumnuai
and Speece, 2004; and Arne and Horst, 2006) and gender
has been noted in the literature as a factor that affects
customer patronage. Thus studying the gender differences
in e-banking patronage is a useful gap in literature that this
paper intends to fll, since the listed factors that are presently
seen as setbacks for e-banking may be perceived differently
by male and female customers alike. The research addressing
the issue of gender difference in e-banking patronage could
help bankers to fnd better ways of communicating with
both sexes which will guide electronic banking marketing
mix decisions.
This research will contribute to the body of consumer
behaviour literature by examining the interrelationships
among observed variables and subsequently, a model of
interrelationships will be created by means of exploratory
factor analysis.
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
244 Business Intelligence Journal July
Literature review
Numerous studies have provided considerable evidence
that gender relates to customers’ perceptions, attitudes,
preferences and purchase decisions (Slyke, Comunale
&Belanger,2002; Mitchell & Walsh, 2004; Fischer &
Arnold,2004; Bakewell & Mitchell,2006; Kwan,Yeung
and Au, 2008; Wan Omar, Ali, Hussin and Rahim,2009).
Sproles (1985) identifed some items that affect customers’
cognitive and affective orientation towards shopping
activities. These items were later refned by Sproles
and Kendall(1986) and a scale consisting of eight mental
customer style characteristics emerged. These include;
perfectionistic high- quality conscious consumer, brand
conscious “price equals quality” consumer, novelty-fashion
conscious consumer; recreational, hedonistic consumer;
Price conscious “value –for-money” consumer; impulsive,
careless consumer; confused by overchoice consumer, and
fnally habitual, brand-loyal consumer. These characteristics
differ from male to female (Mitchell and Walsh, 2004).
Bakewell and Mitchell (2003) later investigated the
decision-making methods of adult female generation Y
consumers in the UK and discovered fve meaningful and
distinct decision making groups; “recreational quality
seekers “, recreational discount seekers”, “trend setting
loyals”, “shopping and fashion uninterested” and confused
time/money conserving”. Similarly in studying the male
decision making method they found that the male consumers
exhibited all the eight traits earlier outlined by Sproles and
Kendall(1986) and four new traits were identifed namely;
store loyal/low price seeking, time-energy conserving,
confused time restricted and store-promiscuity. This is a
pointer to the point that apart from the fact that some of the
purchase characteristics differ on the basis of sex, they are
also product specifc.
Mitchell and Walsh (2004) compared the decision-
making method/styles of male and female shoppers in
Germany, they found that male individuals were slightly
less likely to be perfectionists, less novelty and fashion
conscious, and less likely to be confused when making
purchases than their female counterparts. In a similar study
by Bakewell and Mitchell (2006) they established that nine
decision-making styles were common to both genders,
and three new male traits; store-loyal/low-price seeking,
confused time-restricted and store promiscuity, and three
female traits; bargain seeking, imperfectionism and store
loyal were also discovered.
Recently Hanzaee and Aghasibeig (2008) established
10-factor style for males and 11-factor styles for females,
nine factor styles were common to both genders, they
concluded that male and female customers differ in their
decision-making styles.
It is obvious that all these studies aligned with the fact
that males and females differ in their decision making styles,
noting that e-banking has its inherent features as discussed
below, could it be possible for males and females disparity
in patronage? This is a gap to be flled by this study.
By accessing banking services from any place and at any
moment, end users can beneft from increased convenience,
simplicity and fastness. Besides banks can reduce their
transaction cost as e-banking is fve times cheaper than
traditional banking ways. They can strengthen their core
business and broaden their customer scope by reaching
valuable customers, selling new fnancial e-services as an
attractive differentiating tool, (Ezio, 2008).
Jun and Cai (2001) identifed bank customers’
perceptions of service quality dimensions using quantitative
techniques. The authors’ conceptualized internet banking
service quality based on three quality perspectives; banking
service product quality, customer service quality and online
systems quality.
Perceived risk is considered an important risk attribute
that impacts on the consumer decision-making process when
buying a product or consuming some services (Mitchell,
1998). Electronic banking is a technology-enabled channel
and consumers’ perceive the use of electronic banking as a
risky decision because technology-enabled services exhibit
invasive technological, unfamiliar and indefnite stimuli
(Davidow, 1986). Therefore, when consumers decide to use
electronic banking, they are exposed to uncertainties such
as the availability, the compatibility, and the performance
of the complementary electronic banking channels (Sarin,
Sego and Chanvarasuth, 2003). The degree to which
individuals accommodate these uncertainties may have
gender implications.
Consumers perceive greater risks when buying
services than tangible goods (Zeithaml, 1981). Services
are perceived as riskier than products because services are
intangible, non-standardized, and regularly sold without
guarantees or warrantees. Consumers can hardly ever
return a service to the service provider since they have
already consumed it, and some services are so technical or
specialized that consumers possess neither the knowledge
nor the experience to evaluate whether they are satisfed,
even after they have consumed the service (Zeithaml,
1981).
Consumer Perceived risks identifed by literature of
electronic banking include;
2011 245
• fnancial risk
• performance risk
• physical risk
• social risk and
• Psychological risk.
Financial risk represents the fnancial loss in using
electronic banking, as consumers may perceive that
reversing a transaction, stopping a payment after discovering
an error, or a refund may not be possible. Performance risk
in electronic banking is less satisfying than non-electronic
banking, as consumers may perceive that electronic banking
cannot be used to complete a transaction when needed due
to the denial of access to their account. Physical risk in
electronic banking refers to possible injury when personal
information is accessed by a third party. Social risk refers to
the older generation who may object to the use of electronic
banking due to their perception that non-electronic banking
is personal and friendly. Psychological risk represents
consumer perceptions that the use of electronic banking
would reduce the self-image of them, or have a negative
effect on their perceived image from other consumers. Time
risk in electronic banking implies that it takes more time
to complete a banking transaction than a non-electronic
banking transaction. Sathye (1999) and Polatoglu &
Ekin(2001) found that the reliability dimension was an
important determinant for consumers who used electronic
banking. Furthermore, Sathye (1999) and Liao & Cheung,
(2002) found that reliability was positively related to the
use of electronic banking. They concluded that the more
secure the consumer perceived electronic banking to be; the
more likely they were to use electronic banking, this can
also vary with sex.
Previous studies have identifed that user input factors
are a function of control, enjoyment and intention to use
(Ng and Palmer, 1999). Control could be described as the
amount of effort and involvement required by consumers
in electronic banking. Enjoyment is the perceived
playfulness and intrinsic value consumers experience from
the utilization of electronic banking. The intention to use
is described as the level of resistance to change, which is
associated with consumers’ intention to change from non-
electronic banking to electronic banking. This may differ
with gender. Gerrard & Cunningham (2003) identifed
that consumers who were more fnancially innovative had
a higher probability of adopting electronic banking than
less fnancially innovative consumers. Similarly, Sathye
(1999) found that even when consumers were aware of the
availability of electronic banking, some consumers might
still not operate this type of banking due to consumers’ low
intention to use electronic banking.
Price factors suggest that perceived relative economic
advantages will motivate consumers to use electronic
banking (Sathye, 1999). For example, consumers using
electronic banking could lower the fxed and variable
costs that are associated with the banking process, due to
reductions in personal error and labour cost savings.
The Report(1997) indicated that for consumers to use
technologies, the price to use technologies needed to be
reasonable when compared to alternatives. Sathye (1999)
argued that, in the context of internet banking, two kinds of
price were accounted for; the normal costs associated with
internet activities, and the bank costs and charges. Polatoglu
and Ekin’s (2001) study further identifed that users of
electronic banking were considerably satisfed with the
cost saving factor through electronic banking. Contrarily,
Sathye (1999) identifed that the costs associated with
electronic banking, such as the cost of electronic banking
activities and bank charges, had a negative effect on
electronic banking adoption. Bakewell and Mitchell (2003)
had earlier found that females tend to be money conserving
than males: this may also have an impact on the level of
patronage of e-banking by males and females.
In general, additional defnite service features, service
specifcations, targets of a service, and the core service
comprised the service product characteristics. The service
product characteristics of electronic banking including:
consumers’ perception of a standard and consistence
service, the time saving feature of electronic banking, and
the absence of personal interactions, have been empirically
found to infuence consumers’ use of electronic banking
which may have gender implications, (Polatoglu & Ekin,
2001 and Karjaluoto, Mattila & Pento, 2002).
The electronic banking literature supports individual
factors such as knowledge (Sathye, 1999), consumer
resources, such as money and information reception and
processing capabilities (Karjaluoto, Mattila and Pento,
2002; Gerrard and Cunningham, 2003), and lifestyle
(Polatoglu and Ekin, 2001) as having impacts on consumers’
adoption of electronic banking. Knowledge refers to the
consumers’ responsiveness of each type of electronic
banking channel in the marketplace, their awareness of
the benefts associated with electronic banking, and their
knowledge of how to utilize electronic banking. The
consumer resource money refers to the ease of access of a
Personal Computer (PC) and the internet. The information
processing and processing capabilities resource is
concerned with consumers’ computer expertise, aptitude of
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
246 Business Intelligence Journal July
internet, and the comprehensibility of electronic banking.
Lifestyle refers to the social life in consumers’ banking
patterns, such as the consumers’ value, the independence
of the electronic banking process, or values relating to the
personal interactions associated with the non-electronic
banking process which may have gender alignment.
Consumers’ knowledge of electronic banking plays
an important role in their use of electronic banking.
Sathye (1999) and Polatoglu and Ekin (2001) empirically
supported the idea that consumer knowledge had an effect
on electronic banking adoption while Sathye (1999) found
that the lack of awareness about electronic banking and
its benefts, including the perception of it (being non-
user friendly) contribute to the non-adoption of electronic
banking. Furthermore, Polatoglu and Ekin (2001) stated
that the more knowledge and skills a consumer possessed
about electronic banking, the easier it was for the consumer
to utilize electronic banking. Colgate, Nguyen and Lee
(2003) coroborated the fact that when consumers made
decisions for different alternatives in the marketplace, the
awareness of the existing alternatives was a determinant
for consumers to stay with their current banking provider
and this is a function of the amount of knowledge they
already possess. The knowledge possessed may vary from
sex to sex depending on the level of commitment to such
endeavour.
Consumer resources also infuence the use of electronic
banking. Mols (1998), Sathye (1999) and Karjaluoto,
Mattila, and Pento’s (2002)’s studies showed that some
consumers lacked access to a personal computer (PC)
and this prohibited the adoption of electronic banking.
Studies have also shown that consumer resources including
computer profciency infuence the consumers’ employment
of electronic banking. Sathye (1999) demonstrated that
consumers described incomprehensibility as a reason for
not using electronic banking. Similarly, Karjaluoto, Mattila,
and Pento’s (2002) literature suggests’ that non-electronic
banking users considered electronic banking as diffcult
to use because they found computers diffcult to operate.
Gerrard and Cunningham (2003) found that consumers
who were non-adopters of electronic banking could be
differentiated by their lower computation profciency and
computer skills. Similarly, Karjaluoto, Mattila, and Pento’s
(2002) empirical results suggested that non-electronic
banking users considered electronic banking as diffcult
to use because they found computers diffcult to operate.
Gerrard and Cunningham (2003) found that consumers
who were non-adopters of electronic banking could be
differentiated by their lower computation profciency and
computer skills. Acquisition of skills may vary with interest
which differs in gender.
Al-Ashban & Burney (2001) identifed white-collar
consumers as being most likely to use electronic banking.
It can be postulated that occupation status (namely white-
collar) is positively related to the choice of electronic
banking. They further showed that as consumers increased
their educational qualifcation level, their adoption of
electronic banking would increase as well. Chan (1997)
established that income was the single most important
variable that infuenced a consumer’s use of a credit
card. Empirical fndings of income positively infuencing
adoption of electronic banking can be found in Al-Ashban
and Burney (2001) and Karjaluoto (2002) studies.
Bank management need to continuously assess the
customers’ decision- making process as well as the
formation of attitudes, preferences and satisfaction of
automated services. It is of little use for banks to attempt
to position an offering to the female gender (for example)
by emphasizing particular attributes that do not constitute
signifcant choice criteria (Delvin,2002).
The Nigerian banking industry has tried to take
advantage of the productivity and customer service gains
that e-banking tend to offer. In spite of the advantages of
these new technologies, many customers are unwilling
to adopt them (Asikhia, 2007). It is therefore important
to understand the customer preferences attitudes and
adaptations of these services particularly in relation with
adoption and such information could be used as marketing
tools to attract new clients and retain the existing clients.
Kolodinsky et al (2000) and Kolodinsky and Hogarth
(2001) studied the adoption of new innovations and services
in banking in US. They concluded that gender differences
existed for phone banking, electronic funds transfer and PC
banking.
It is pertinent to say that evidence abound that male and
female consumers or customers differ in decision making
styles and obvious perceptions about a product or service,
it is believed that e-banking customers in Nigeria may
also have certain distinctive gender characteristics and
alignment in their patronage. This study flls this gap by
studying the differences in decision-making of e-banking
patronage based on gender in the Nigerian context.
E – banking in Nigeria
The competition train in the banking industry has
shifted to technology and intensive delivery of services
which has created paradigm shift in banking services in
2011 247
Nigeria. Banks like Zenith Bank, and Guaranty Trust Bank
used technology as a competitive weapon and successfully
became one of the largest banks in Nigeria within four
years post-consolidation era. Banking service delivery was
re-engineered with the introduction of technology. Several
new products were introduced, for example we have
internet banking, Automated Teller Machines (ATMs),
phone banking/ cell banking, debit cards, credit cards etc.
Integrated Banking Solutions (IBS) was also introduced
in most banks. A central server replaced all branch level
and controlling offce level servers. Every transaction that
is taking place across any service point like a branch or
ATM is recorded in the server. This new operational method
helps customers to conduct normal banking transactions
like deposit/withdrawal of cash from account, account
transfers etc. from any branch of the twenty - fve banks in
Nigeria without the account being domiciled in that branch.
Automated Teller Machines (ATM) are equipped to
facilitate several transactions, customers can withdraw
from the ATM of a bank that they don’t have an account.
Other roles like account transfers, payment of bills, balance
enquiry are all possible through ATM. Over 25,000 ATMs
are currently in use in the country connected to Inter Switch
Network.
Electronic banking is gaining patronage where
information from central server is made accessible to the
account holder using a PIN or PASSWORD. Account
statements, account transfer facilities, bill payment
facilities, list of cheques to be cleared, loan facilities in
form of assets, vehicle and mortgage, international bank
transfer are some of the facilities given by the banks to their
customers. However, recent security threat to customers
savings have made banks to be more careful in their online
transactions, customers are alerted from time to time on the
secrecy of their password. All banks often communicate
relevant information through electronic mail (e-mail) or
short messages on phone to their customers. Banks also
have their websites to promote their products and services
as well as publicize information that is relevant to the public
about the bank.
Credit and debit cards are gaining patronage; many
Nigerian banks have started providing these services.
Master and VISA are the most preferred companies with
which banks make arrangements for providing credit/debit
cards to clients. Many banks are also having their own
debit cards on which the Nigerian currency could be loaded
for transactions depending on the nature of the account.
If the account is a domiciliary account, then the foreign
currency could be loaded on it. And it is essentially used
to operate the account that the customer is having with the
bank, this has become important in the light of the fact that
most Nigerian banks now have branches in other parts of
the world particularly in Britain and most West African
countries. Account balances are now being sent through
phone and alerts are also sent when any money is deposited,
transferred or withdrawn from an account. Virtual banks
are on the horizon in Nigeria.
Method
The research instrument was self-administered, it consist
of 29- items e-banking characteristics features of service
quality (5 items), perceived risk factors (5items), user input
factors(5 items), price and service product characteristics
factor(5 items), individual factors(5 items) and customer
satisfaction factors (4 items). All scales are measured on
a 7-point Likert –type scales ranging from 1(very strongly
disagree) to 7 (very strongly agree). The reliabilities of
the items ranged from 0.79 to 0.97 as shown in Table
1.0. Order effects were reduced by random ordering of
items, and the research instrument asked questions to also
draw information on the demographic background of the
respondents. The questionnaire was self-administered to a
non-probability sample of 750 male and 750 female making
a total of 1500 questionnaires that were distributed randomly
amongst bank customers in Lagos and Abuja because these
cities are believed to have a near representation of all tribes
and professional spheres in Nigeria (Nigeria Bureau of
Statistics, 2007). The concentration in these two cities is
likely to reduce error within the measurement model being
infated by situational factors inherent in diverse samples
from the thirty six states of the federation. The banks used
in each of the location were selected through convenience
sampling method; customers of banks close to researcher’s
location within these cities were used. A total of 375 male
and 375 female customers were sampled randomly from the
two locations (Abuja and Lagos) and the eventual spread
of the customers according to banks and geographical
location is presented in appendix 1. Of 1500 questionnaires
distributed, one thousand and sixty questionnaires (1060)
were deemed usable for data analysis, which is a 70.7
percent response rate. The sample comprised of 54.7
percent males and 45.3 percent females. The demographic
data of the respondents is represented in table 1.0.
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
248 Business Intelligence Journal July
Table 1: Demographic Data of Respondents
/N Items Male Female
1 Age: 18-25 150 220
26-35 290 140
36-45 110 80
46 and above 30 40
Total 580 480
2 Educational qualifcation:
WASCE 100 60
OND 30 60
B.sc/BA 330 340
M.sc/PhD 110 20
Others specify 10 -
3 Duration of Patronage:
0-2 years 210 100
3-5 years 250 280
6-8 years 70 80
9 and above 50 20
4 Type of account:
Current 340 200
Savings 240 280
Others specify - -
Source: Fieldwork by Author.
Analysis
Exploratory principal component analysis with a
varimax rotation was used to summarize the items into
an underlying set of male and female decision-making
factors because they are constructs of a number of directly
observable variables. Exploratory factor analysis helps in
clustering these variables into factors with unique underlying
variables and it also helps to validate that respondents are
able to distinguish between the various variables despite
similarity of items questioned (De Vaus,2002).
Kaiser-Meyer – Olkin measure of sampling adequacy
for both samples were higher than the acceptable limit of
0.5 (Hair et al, 2005) The measure for male is 0.875 and
for females is 0.675 and Barlett’s tests were signifcant
refecting the suitability of data for factor analysis.
To assess the internal consistency of each factor
group obtained, a reliability analysis was conducted by
calculating the Cronbach’s alpha for each factor (Table 3).
For consistency, it was decided that reliability should not be
lower than 0.4, the same level used by Sproles and Kendall
(1986).
The descriptive statistics of the male and female samples
are shown in table 2.
Results and Discussion
The descriptive statistics of the samples (male and
female) revealed a highest average score of 5.62 in Q7
(i.e. In-branch banking involves too much queuing time) in
male samples and female samples has the highest average
score in Q18 (i.e. Electronic banking is convenient). These
two results tend to imply the same thing, while the male
customers denounced in-branch banking for its obvious
disadvantage of inconvenience, the female customer
upheld e-banking for being more convenient. Convenience
banking tend to be embraced by both male and female
customers (Q18 for male is 0.543). A further scrutiny of
the responses revealed that Q2, Q3, Q4, Q5, Q6, Q7, Q9,
Q10, Q11, Q12, Q13, Q14, Q15, Q16, Q17, Q18, Q20, Q21
had scores that ranged between 2.00 to 5.62 among males,
while Q1, Q2, Q5, Q6, Q7, Q8, Q9, Q15, Q16, Q17, Q18,
Q21 had scores that ranged between 5.00 to 5.62 among
females. The common factors where the male and female
customers had scores above 5.00 are Q2, Q5, Q6, Q7, Q9,
Q15, Q16, Q17, Q18 and Q21. These common factors are:
• Q2 : Transactions through electronic banking are
reliable
• Q5 : Electronic banking services are faster than in-
branch banking
• Q6 : Travelling to a bank involves too much queuing
time.
• Q7 : In-branch banking involves too much queuing
time.
Table 2: Descriptive Statistics of the Samples
Variables
Male Female
Mean
Std
Deviation
N Mean
Std
Deviation
N
Q1 4.92 1.742 580 5.36 1.953 480
Q2 5.29 1.643 580 5.18 1.652 480
Q3 5.12 1.618 580 4.77 1.828 480
Q4 5.18 1.600 580 4.69 1.067 480
Q5 5.34 1.430 580 5.23 1.597 480
Q6 5.17 1.478 580 5.33 1.644 480
Q7 5.62 1.434 580 5.05 1.716 480
Q8 4.95 1.501 580 5.05 1.654 480
Q9 5.16 1.664 580 5.10 1.465 480
2011 249
Variables
Male Female
Mean
Std
Deviation
N Mean
Std
Deviation
N
Q10 5.09 1.450 580 4.82 1.760 480
Q11 5.17 1.756 580 4.97 1.993 480
Q12 5.24 1.727 580 4.87 1.765 480
Q13 5.11 1.647 580 4.69 1.976 480
Q14 5.09 1.779 580 4.74 1.970 480
Q15 5.11 1.628 580 5.00 1.777 480
Q16 5.03 1.619 580 5.05 1.820 480
Q17 5.43 1.430 580 5.46 1.699 480
Q18 5.43 1.513 580 5.64 1.616 480
Q19 4.99 1.669 580 4.44 1.818 480
Q20 5.13 1.639 580 4.46 1.668 480
Q21 5.17 1.471 580 5.28 1.589 480
Q22 4.82 1.680 580 4.38 1.955 480
Q23 4.83 1.720 580 4.41 1.956 480
Q24 3.40 2.000 580 3.08 1.952 480
Q25 3.74 1.957 580 4.05 1.946 480
Q26 4.06 0.818 580 3.82 1.211 480
Q27 4.17 0.783 580 4.13 1.151 480
Q28 4.02 0.969 580 3.62 1.138 480
Q29 3.84 1.218 580 3.64 1.135 480
Source: Field work by Author
• Q9: Switching from electronic banking to in-branch
banking could be inconvenient to me.
• Q15: I like to use new methods to conduct transactions
e.g. ATM, telephone banking and internet banking.
• Q16: Electronic banking charges are expensive.
• Q17: Electronic banking is time saving.
Table 3: Results of Factor Analysis for Males and Females
Items Factor Leadings
All factors Male Female
Factor 1 – Service Quality (.568) (.551)
Q1 Transactions through electronic banking
are accurate.
.7056 .7868
Q2 Transactions through electronic banking
are reliable
.6799 .7758
Q3 I am familiar with e-banking .7913 .7593
Q4 I am comfortable with e-banking .7442 .7069
Q5 Electronic banking services are faster than
in-branch banking
.6068 .6855
Factor 2 – Perceived Risk Factors (.411) (.658)
Q6 Travelling to a bank involves too much
queuing time.
.6360 .7858
Items Factor Leadings
All factors Male Female
Q7 In-branch banking involves too much
queuing time.
.6924 .8545
Q8 Electronic banking services are faster than
in-branch banking.
.6847 .8378
Q9 Switching from electronic banking to
in-branch banking could be inconvenient
to me.
.6556 .6754
Q10 Going to a bank branch involves travel
costs.
.6335 .5622
Factor 3 – User Input Factors (.519) (.557)
Q11 I like to use electronic banking because it
ofers independence.
.8966 .8151
Q12 Electronic banking enables me to be fully
involved in my transactions.
.8032 .7408
Q13 E-banking is enjoyable to use. .9313 .6106
Q14 E-banking is user friendly. .8080 .7256
Q15 I like to use new methods to conduct
transactions e.g. ATM, telephone banking
and internet banking.
.7891 .6886
Factor 4 – Price Factor (.397) (.496)
Q16 Electronic banking charges are expensive. .5126 .4149
Factor 5 – Service Product Characteristics (.466) (.341)
Q17 Electronic banking is time saving. .7036 .7323
Q18 Electronic banking is convenient. .7267 .7261
Q19 Customer service in e-banking has
consistent standard.
.7836 .5852
Q20 Electronic banking has a wide variety of
services available.
.
Factor 6 – Individual Factors (.405) (.622)
Q21 I have a regular access to a computer. .6017 .5526
Q22 I have regular access to the internet. .6894 .6173
Q23 E-banking is easy to use. .6829 .7319
Q24 I use e-banking because my friends use it. .1042 .1622
Q25 The use of e-banking refects my social
status.
.3299 .2711
Factor 7 – Customers Satisfaction (.271) (.104)
Q26 How are you satisfed with time saving? .3560 .2978
Q27 How are you satisfed with convenience? .2439 .6529
Q28 How are you satisfes with the consistency
standards?
.3904 .4453
Q29 How are you satisfed with the wide
variety of services available?
.4463 .4216
Source: Field work by Author
• Q18: Electronic banking is convenient.
• Q21: I have a regular access to a computer.
The common reasons why both male and female
customers patronize e-banking in Nigeria is that it is
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
250 Business Intelligence Journal July
reliable, faster, does not involve travelling and queuing
time, it is time saving, and convenient. It is worthy to note
that an average customer sampled has access to computer.
And electronic banking is both seen to be expensive by
male and female customers.
The factors that are rated high by the male and not in
female, which can be called male factors, are: Q3, Q4, Q10,
Q11, Q12, Q13, Q14, and Q20. They are represented thus:
• Q3: I am familiar with e-banking.
• Q4: I am comfortable with e-banking.
• Q10: Going to a bank branch involves travel costs.
• Q11: I like to use electronic banking because it offers
independence.
• Q12: Electronic banking enables me to be fully involved
in my transactions.
• Q13: E-banking is enjoyable to use.
• Q14: E-banking is user friendly.
• Q20: Electronic banking has a wide variety of services
available.
The male customers patronizes electronic banking
because of its familiarity, comfort, cost effectiveness,
independence, high level of involvement, easy to use
feature, friendliness, and wide variety of services.
The factors that are peculiar to the female customers are
Q1 and Q8. They are represented thus:
• Q1: Transactions through electronic banking are
accurate.
• Q8: Electronic banking services are faster than in-
branch banking.
The female customer particularly patronizes the
electronic banking because of its accuracy and fastness.
Table 3 shows that the variables of service quality loaded
high for both male and female customers, they both patronize
electronic banking for the quality of service rendered, which
ranges from accuracy to speed. It is however important to
note that male customers loaded higher than the female
customers on all male factors (i.e. Q3, Q4, Q10, Q11, Q12,
Q13, Q14, and Q20) while female customers loaded higher
than male customers on factors peculiar to the female
customers as interpreted in table 2 (i.e. Q1 and Q8). Most
of the factors are important patronage of both male and
female customers except two of the individual factors for
both male and female customers; these are: Q24 – “ I use
e-banking because my friends use it”, Q25 – “The use of
e-banking refects my social status”.
This implies that both customers are not using
e-banking because of their friends and social status. These
factors should not be used as a unique selling proposition
for e-banking. In addition, strangely, male and female
customers loaded low on the group factor of customer
satisfaction, which means there is room for improvement
by banks to satisfy the customers better; presently the
expectations of the customers whether male or female are
higher than the service received particularly as it concerns
time saving (Q26 : .3560 for male and .2978 for female)
as well as consistency in standards (Q28 : .3904 for male).
The alignment of the male and female customer to some
of the factors of electronic banking is a signal to gender
implied impact on e-banking patronage, this result concur
with the fndings of Hanzaee and Aghasibieg (2008) who
established 10 – factors styles for males and 11 – factors
style for females, nine factors were common to both genders.
However both studies differ in the number of factors to
male and female, as well as common factors. This is so
because different product/service are being investigated
apart from using scale items peculiar to each study. For this
paper, 8 factors were common to male customers, 2 factors
to female customers and 10 factors were common to both.
Furthermore, the alignment of different factors by sex
confrms the fndings Kolodinsky et al (2000), Kolodinsky
and Hogarth (2001) who found that gender differences
existed in the adoption of new innovations and services
in banking in the United States. Special appeals could be
communicated to male and female customers based on the
identifed factors for further patronage.
The banks in Nigeria scored low in customer
satisfaction as it pertains to electronic banking, this calls
for improvement in service delivery in terms of timeliness
and consistency in standards.
Conclusion
It is evident that the male and female customers perceived
e-banking services differently and thus they patronize this
service for different reasons. Hence for the different banks
to sustain their interest there is dire need to emphasize the
common factors to the particular gender whose patronage
is essential depending on the product. The result also help
to identify the particular unique selling proposition to be
emphasized by the banks in the course of marketing any
gender inclined product, as their particular area of interest
is important for emphasis.
The issue of low satisfaction level of both genders is an
obvious area of improvement to be pursued by the banks.
2011 251
Some factors that are external to the operation of the banks
may be responsible for this, for example incessant power
failure. This means that banks may need to be proactive
to consistently keep their customers’ interest in a turbulent
business environment like Nigeria by ensuring that nothing
interrupts the fow of service delivery. Investments
should be made in this area to keep abreast of the level of
competition in the industry.
References
Akpan, I. (2009). Cross channel integration and
optimization in Nigerian banks. Telnet Press
Release, 20(1): 1-4.
Al-Ashban, A. & Burney, M. (2001) Customer
adoption of tele-banking technology: The case
of Saudi-Arabia. International Journal of Bank
Marketing 19(5): 191-200
Bakewell, C. & Mitchell, V.W. (2006). Male and
female decision making, Journal of Business
Research, 59: 1297- 1300
Bakewell, C. and Mitchell, V. W. (2003). Generation
Y female consumer decision-making styles.
International Journal of Retail & Distribution
Management, 31(2): 95-106
Chan, R.Y. (1997). Demographic and attitudinal
differences between active and inactive credit card
holders – The Case of Hong-Kong. International
Journal of Bank Marketing 15(4): 117 – 1200.
Cheung, M.I. and Liao, Z. (2002). Internet-based
e-banking and consumer attitudes: an empirical
study. Information and Management 39(4): 283 –
295.
Colgate, M; Nguyen, V. and Lee, C. (2003). Switching
barriers in New Zealand. University of Auckland
Business Review 5(1): 23 – 32.
Davidow, W. H. (1986). Marketing high technology:
An insider’s review. New York: Free Press.
De Vaus, D.(2002). Analyzing social science data.
London: Sage Publication
Delving, J. F (2002) Customer knowledge and choice
criteria in retail banking, Journal of Strategic
Marketing, 10:273-290.
Ezio, G. (2008) Authentication solution for e-banking.
New York: Prentice-Hall.
Fischer, E. & Arnold, S.J.(2004). Sex, gender identity,
gender role attitudes and consumer behavior.
Psychology & Marketing, 11(2):163-182
Gerrard, P. & Cunningham, J. (2003). The diffusion
of internet banking among Singapore consumers.
International Journal of Bank Marketing 21(1): 16-
28.
Hanzaee, K. H. & Aghasibieg, S. (2008) Generation
Y female and male decision-making styles in Iran:
Are they different? International Review of Retail,
Distribution and Consumer Research, 18(5):521-
537.
Horst, T. Andreas, P. & Arne, F. (2008) Success factors
of internet payment systems. International Journal
of Electronic Business 6(4): 365-385.
Jun,M. and Cai, S. (2001) The key determinants of
internet banking service quality: a content analysis.
International Journal of Banking Marketing
19(7):276-291.
Karjaluoto, H. (2002), Selection criteria for a mode
of bill payment: empirical investigation among
Finnish bank customers, International Journal of
Retail & Distribution Management, 30 (6):331-339.
Karjaluoto, H. Mattila, M. and Pento, T. (2002) Factors
underlying attitude formation towards online
banking in Finland. International Journal of Bank
Marketing, 20(6): 261-272.
Kolodinsky, Hogarth, J. M and Shue,J.(2001). Bricks
or Chicks? Consumers’ Adoption of Electronic
Banking technologies. Consumers Interest Annual
(46): 180-189.
Kolodinsky, J. and Hogarth, J. M (2001) The Adoption
of Electronic Banking Technologies by American
consumers. Consumer Interest Annuals, 47:1-9.
Kwan,C. W.,Yeung, K.W. & and Au, K.F.(2008)
Relationship between consumer decision making
styles and lifestyle characteristics: Young fashion
consumers in China. Journal of the Textiles Institute,
99(3): 193- 209.
Liao, Z., Cheung, M.T. (2002), Internet-based
e-banking and consumer attitudes: an empirical
study, Information and Management, 39:283-295.
Mitchell, V. & Walsh, G.(2004). Gender difference in
German consumer decision making styles. Journal
of Consumer Behaviour, 3(4):331-346.
Mitchell,V.W. & Bates, L. (1998). UK consumer
decision –making styles. Journal of Marketing
Management, 14: 199-225.
Mols, N.P. (1998). The internet and the banks strategic
distribution channel decisions. Journal of Internet
Research 8(4): 331 – 337.
Olalekan A. - E-banking Patronage in Nigeria: An Exploratory Study of Gender Difference
Asikhia, Olalekan
Business Intelligence Journal - July, 2011 Vol.4 No.2
252 Business Intelligence Journal July
Mukherjee, A. and Nath, P. (2003). A Model of Trust in
Online Relationship banking. International Journal
of Banking 21(1): 28-35.
Ng, T., & Palmer, E. (1999), Customer satisfaction
attributes for technology-interface services, MSIS
Department, School of Business and Economics,
University of Auckland, Auckland, unpublished
paper.
Nigeria Bureau of Statistics (2007). www.afdevinfo.
com/htm/reports/org/org_41583.html.
Polatoglu, V.N., & Ekin, S. (2001), An empirical
investigation of the Turkish consumers’ acceptance
of internet banking services, International Journal
of Bank Marketing,19(4):156-165.
Potchanakitummiai, S. & Speece, M. (2004). Barriers
to internet banking adoption. Journal of Electronic
Commerce Research 5(4): 270 – 286.
Report, W. (1997). The Financial System Inquiry Final
Report. New York: Free Press.
Sarin, S., Sego, T., Chanvarasuth, N. (2003). Strategic
use of bundling for reducing consumers’ perceived
risk associated with the purchase of new high-
tech products. Journal of Marketing Theory and
Practice, 11(3):71-83.
Sathye, M. (1999). Adoption of internet banking by
Australian consumers: an empirical investigation,
International Journal of Bank Marketing, 17(7):324-
334.
Slyke, C.V; Comunale, C.L; Belanger, F. (2002).
Gender differences in perceptions of web-based
shopping. Communications of the ACM 45(8): 82
– 86.
Sproles, G.B. & Kendall, E.L. (1986). A methodology
for profling consumers’ Decision-making
Styles, Journal of Consumer Affairs, 20(2): 267 – 279.
Sproles, G.B. (1985). From Perfectionism to Faddism.
Measuring Consumers’ Decision making Styles
proceedings. American Council on Consumer
Interest, 79 – 85.
Wan Omar, M; Ali, M.N; Hussin, Z; and Rahim, H.A.
(2009). Decision Orientations towards Shopping
and Buying among Young – Adult Malays in the
Universities, International Journal of Business and
Management 4(7): 26 – 30.
Zeithaml, V.A. (1981), How consumer evaluation
processes differ between goods and services, in
Donnelly, J.H., George, W.R. (Eds),Marketing
of Service, Proceedings Series of the American
Marketing Association.
2011 253
FOREIGN DIRECT INVESTMENT AND ITS EFFECT ON THE
NIGERIAN ECONOMY
Omankhanlen Alex Ehimare
Lecturer at the Department of Banking and Finance, Covenant University, Ota,
Ogun State, Nigeria.
Email: [email protected]
Abstract
This research study deals with the effect of Foreign Direct Investment on the Nigerian economy over the period 1980-2009.It helped
examined empirically if the following growth determining variables in the economy-Balance on current account (Balance of payment),
Infation and Exchange rate have any effect on Foreign Direct Investment. Also if Foreign Direct Investment have any effect on Gross
Domestic Product (GDP). Econometric models was developed to investigate the relationships between the aforementioned variables
and foreign direct investment. Based on the data analysis it was discovered that foreign direct investments have positive and signifcant
impact on current account balance in Balance of payment. While infation was seen not to have signifcant impact on foreign direct
investment infows. The exchange rate has positive effect on foreign direct investment. Therefore it is recommended that for Nigeria to
attract the desired level of FDI, it must introduce sound economic policies and make the country investor friendly. There must be political
stability, sound economic management and well developed infrastructure. Key words: Foreign direct investment, growth, human capital,
OLS, Nigeria, infrastructure, balance of payment, infation, exchange rate.
Governments have been trying to lift the country out
of the economic doldrums without achieving success as
desired. Each of these governments has not focused much
attention on investment especially foreign direct investment
which will not only guarantee employment but will also
impact positively on economic growth and development.
FDI is needed to reduce the difference between the desired
gross domestic investment and domestic savings.
Jenkin and Thomas (2002) assert that FDI is expected
to contribute to economic growth not only by providing
foreign capital but also by crowding in additional
domestic investment. By promoting both forward and
backward linkages with the domestic economy, additional
employment is indirectly created and further economic
activity stimulated.
According to Adegbite and Ayadi (2010) FDI helps fll the
domestic revenue-generation gap in a developing economy,
given that most developing countries’ governments do
not seem to be able to generate suffcient revenue to meet
their expenditure needs. Other benefts are in the form
of externalities and the adoption of foreign technology.
Externalities here can be in the form of licencing, imitation,
employee training and the introduction of new processes
by the foreign frms (Alfaro,Chanda,Kalemli- Ozean and
Sayek 2006).
Foreign direct investment consists of external resources
including technology, managerial and marketing expertise
and capital. All these generate a considerable impact
on host nation’s productive capabilities. The success of
government policies of stimulating the productive base
of the economy depend largely on her ability to control
adequate amount of FDI comprising of managerial,
capital and technological resources to boast the existing
production capacity. Although the Nigerian government
has being trying to provide conducive investment climate
for foreign investment, the infow of foreign investments
into the country have not been encouraging.
Given the Nigerian economy resource base, the country’s
foreign investment policy should move towards attracting
and encouraging more infow of foreign capital. The need
for foreign direct investment (FDI) is born out of the under
developed nature of the country’s economy that essentially
hindered the pace of her economic development. Generally,
policy strategies of the Nigerian government towards
foreign investments are shaped by two principal objectives
of the desire for economic independence and the demand
for economic development.
An analysis of foreign fow into the country so far have
revealed that only a limited number of multinationals or
their subsidiaries have made Foreign Direct Investment
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
254 Business Intelligence Journal July
in the country. Added to this problem of insuffcient
infow of FDI is the inability to retain the Foreign Direct
Investment which has already come into the country. Also
what effect have foreign direct investment have on such
variables as- Gross Domestic Product (GDP) and Balance
of Payment(BOP).Moreover, what effect does infation and
exchange rate have on Foreign Direct Investment? Carkovic
and Levine (2002) in their study concluded that exogenous
component of FDI does not exert a robust positive infuence
on economic growth.
According to Ayanwale (2007). The relationship
between FDI and economic growth in Nigeria is yet unclear,
and that recent evidence shows that the relationship may be
country and period specifc. Therefore there is the need to
carry out more study on their relationship.
Developing countries economic diffculties do not
originate in their isolation from advance countries. The
most powerful obstacle to their development comes from
the way they are joined to the international system. Added
to this problem is the poor external image Nigeria have
and the concept of European Economic Community that
include Eastern Europe.
This translates to the fact that investment fows that
would normally come from western countries now go to
poor European Economic Communities which include
Eastern Europe.
Foreign direct investment (FDI) is a major component
of capital fow for developing countries, its contribution
towards economic growth is widely argued, but most
researchers concur that the benefts outweigh its cost on the
economy. (Musila and Sigue, 2006).
Mc Aleese (2004) states that “FDI embodies a package
of potential growth enhancing attributes such as technology
and access to international market” but the host country
must satisfy certain preconditions in order to absorb and
retain these benefts and not all emerging markets possess
such qualities. (Boransztain De Gregorio and Lee 1998, and
Collier and Dollar, 2001). This paper is divided into fve
parts. Part one above is the introduction. Part two reviews
the relevant literature, part three discusses the methodology
employed in this study, and part four is data presentation
and analysis while part fve discusses the fndings and
recommendation.
This study will evaluate the fow of FDI in Nigeria and
its Effect on the Nigerian Economy. The period 1980-2009
will be investigated in the study. Only FDI, Government
Expenditure and Gross Capital formation will be used
as the explanatory variables. While GDP and balance on
current account of Balance of payment will be used as the
dependent variables.
Literature Review
The Changing international economic and political
environment has led to a renewed interest in the benefts
foreign direct investment (FDI) can offer to developing
countries in achieving economic growth. The growing
interest in foreign direct investment (FDI), stand from
the perceived opportunities derivable from utilizing this
form of foreign capital injection into the economy, to
augment domestic savings and further promote economic
development in most developing economies (Aremu 2005).
FDI is believed to be stable and easier to service than
bank credit. FDI are usually on long term economic
activities in which repatriation of proft only occur when
the project earn proft. As stated by Dunning and Rugman
(1985) Foreign Direct Investment (FDI) contributes to
the host country’s gross capital formation, higher growth,
industrial productivity and competitiveness and other spin-
off benefts such as transfer of technology, managerial
expertise, improvement in the quality of human resources
and increased investment.
According to Riedel (1987) as cited by Tsai (1994) while
the potential importance of FDI in less developed countries
(LDCs) development process is getting appreciated, two
fundamental issues concerning FDI remains unresolved. In
the frst place what are the determinants of FDI? Specifcally
from LDCs point of view are there factors in the control of
the host country that can be manipulated to attract FDI?
Or as some researchers claim that by and large LDCs
play a relatively passive role in determining the direction
and volume of FDI. This is the question about the demand
side determinants (or host country factors) of FDI which
are widely discussed in the literature.
There are also the supply side determinants or frm
specifc factors of FDI (Ragazzi 1973). The supply
side factors are beyond the control of LDCs. A body of
theoretical and empirical literature has investigated the
importance of FDI on economic growth and development
in less developed countries. For example see (Dauda 2007)
(Akinlo 2004) (Deepak, Mody and Murshid 2001) (Aremu
2005) e.t.c.
Modern growth theory rest on the view that economic
growth is the result of capital accumulation which leads to
investment. Given the overriding importance of an enabling
environment for investment to thrive, it is important to
2011 255
examine necessary conditions that facilitate FDI infow.
These are classifed into economic, political, social and
legal factors. The economic factors include infrastructural
facilities, favourable fscal, monetary, trade and exchange
rate policies. The degree of openness of the domestic
economy, tariff policy, credit provision by a country’s
banking system, indigenization policy, the economy’s
growth potentials, market size and macroeconomic stability.
Other factors like higher proft from investment, low
labour and production cost, political stability, enduring
investment climate, functional infrastructure facilities and
favourable regulatory environment also help to attract and
retain FDI in the host country. (Ekpo 1997).
According to the International Monetary Fund (1985)
foreign Direct Investment is an investment made to acquire
a lasting interest in a foreign enterprise with the purpose
of having effective voice in management. While Dunning
(1993) describe it as an investment made by an investor
based in a country to acquire assets in another country
with the intention to manage the assets. Mwillima (2003)
describe foreign direct investment as investment made so
as to acquire a lasting management interest (for instance
10% of voting stocks) and at least 10% of equity shares in
an enterprise operating in another country other than that of
the investor’s country.
Foreign Direct investment can also be describe as an
investment made by an investor or enterprises in another
enterprises or equivalent in voting power or other means
of control in another country with the aim to manage the
investment and maximize proft. This investment involves
not only the transfer of fund but also the transfer of physical
capital, technique of production, managerial and marketing
expertise, product advertising and business practice with
the aim to make proft.
In recent years due to the rapid growth and changes in
global investment patterns, the defnition of Foreign Direct
investment have been broadened to include the acquisition
of a lasting management interest in a company or enterprise
outside the investor’s home country.
Generally, the theory that explains the nexus between
FDI and growth in terms of output and productivity is
signifcantly positive. However, empirical literature
yields varying results. Some research studies fnd positive
outcome from outward FDI for the investing country (Van,
Poffelsberghe, De La Potterie & Lichtenberg, 2001), but
suggest a potential negative impact from inward FDI on
the host country. This results from a possible decrease in
indigenous innovative capacity or crowding out of domestic
frms. Other studies report more fndings that are positive.
For example, Nadiri (1993) fnds positive and signifcant
effects from US sourced FDI on productivity growth of
manufacturing industries in France, Germany, Japan and
United Kingdom. Borensztein, Gregorio and Lee (1998)
also fnd a positive infuence of FDI fows from industrial
countries on developing countries’ growth. However, they
also report a minimum threshold level of human capital for
the productivity enhancing impact of FDI, emphasizing the
role of absorptive capacity.
In the neo-classical production function approach, output
is generated by using capital and labour in the production
process. With this framework in mind, FDI can exert an
infuence on each argument on the production function.
FDI increases capital, and may qualitatively improve the
factor labour and by transferring new technologies, it also
has the potential to raise total factor productivity. Therefore,
in addition to the direct, capital augmenting effect, FDI
also have additional indirect and thus permanent effects
on output growth rate. Further, by raising the number of
varieties for intermediate goods or capital equipments, FDI
can also increase productivity (Borensztein, Gregorio &
Lee, 1998).
Therefore, though FDI could produce a signifcant effect
on output growth through speeding up capital formation
process, the effect tends to diminish in the long run because
of the principle of diminishing return.
As opposed to the limited contribution that the neo-
classical theory accredits to FDI, the endogenous growth
literature points out that FDI can not only contribute to
economic growth through capital formation and technology
transfers (Blomstrom, Lipsey & Zejan, 1996) but also do
so through the augmentation of the level of knowledge via
labour training and skill acquisition (De Mello, 1997).
Research Methodology Model
Specification
This study is based on the assumption that the infow
of FDI affects economic growth in Nigeria (GDP) and
Nigeria’s Balance of Payment (BOP). And again, that
infation and exchange rate in turn affect the infow of
Foreign Direct Investment (FDI). In other-words, GDP and
BOP are dependent on FDI, hence the model:
GDP = f (FDI)
(1)
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
256 Business Intelligence Journal July
Estimated Results
Model 1
t 0.842 1.568 10.237 3.063
GDP = ?
o
+ ?
I
FDI + ?
2
GOV + ?
3
GCF + e
GDP= 1.6709 + 4.0912FDI + 6.2835GOV. + 1.5457GFC
S.E.= (1.9847) (2.6086) (0.61381) (0.50454)
R
2
= 0.989607 F-Statistic= 825.24 D.W.= 2.74
Model 2
BCA = ?
o
+ ?
I
FDI - ?
2
GOV + ?
3
GCF + e
BCA= -1.3500 + 7.0662FDI + -0.49248GOV. + 0.42403GFC
S.E.= (1.1447) (21.5046) (0.35404) (0.29101)
t -1.179 4.696 -1.391 1.457
R
2
= 0.919443 F-Statistic = 98.917 D.W.= 1.72
BCA = f (FDI)
(2)
FDI = f (INFL., EXR.)
(3)
Where:
GDP = Gross Domestic Product
BCA = Current Account Balance
FDI = infow of Foreign Direct Investment
INFL. = Infation rate
EXR. = Exchange rate
Considering the fact that the GDP and BOP of an
economy are not determined by FDI alone, the inclusion
of two more growth determining variables is made so as to
get a more realistic model: Hence, equation (1) is extended
thus:
GDP = f (FDI, GOV, GCF)
(4)
BCA = f (FDI, GOV, GCF)
(5)
FDI = f (INFL, EXR.)
(6)
Where:
GOV = Government expenditure
GCF = Gross fxed capital formation
Equations (4) and (5) show that GDP and BCA are
dependent on FDI, GOV and GCF.
The statistical forms of the models are thus:
(7)
GDP = ?
o
+ ?
I
FDI + ?
2
GOV + ?
3
GCF + e
BCA = ?
o
+ ?
I
FDI - ?
2
GOV + ?
3
GCF + e
(8)
(9)
FDI = ?
o
- ?
I
INFL. - ?
2
EXR. + e
Where:
?0 = the intercept for equations (1) and (2)
?0 = the intercept for equation (3)
?I = the parameter estimate of FDI.
?2 = the parameter estimate of GOV.
?3 = the parameter estimate of GCF.
? I = the parameter estimate of INFL.
? 2 = the parameter estimate of EXR.
e = the random variable or error term.
Annual time-series data on the variables under study
covering thirty year period 1980-2009 are used in this study
for estimation of functions. Foreign Direct Investment
infow (FDI), Government Expenditure (GE) and Gross
fxed Capital Formation (GCF) are the relevant explanatory
variables. Equally, the Gross Domestic Product and Balance
on Current Account are the dependent variables. The Gross
Domestic Product is the quantitative variable that measures
economic performance of a country and the Balance on
Current Account measures BOP.
Presentation of Results
The regression analysis and tests of hypotheses are
conducted at 5% signifcance level. After running the
relevant regressions, the following results were obtained
and are presented below:
2011 257
Model 3
FDI = ?
o
- ?
I
INFL. - ?
2
EXR. + e
FDI= -14108. + -310.46 INFL. + 3731.5 EXR.
S.E.= (58549) (1678.9) (538.18)
t -0.241 -0.185 6.934
R
2
= 0.666903 F-Statistic= 27.029 D.W.= 0.453
Model 1
From the regressions result, the R-squared (R²) value of
0.989607 shows that at 98.96% the explanatory variables
explain changes in the dependent variable. This means that
at 98.96% the independent variables explain changes on
the Gross Domestic Product (GDP). This simply means
that the explanatory variables explain the behaviour of the
dependent variable at 98.96%. The calculated F-statistics
of 825.24 which is greater than the value in the F-table
(2.9751) implies that all the variables’ coeffcients in the
regression result are all statistically signifcant to GDP.
The Durbin-Watson (DW) as shown in the regression
analysis is 2.74 which shows that there is the presence of
autocorrelation.
The above model tested the effect of three different
variables namely – Foreign Direct Investment (FDI),
Government Expenditure (GOV) and Gross fxed Capital
Formation (GCF) on Gross Domestic Product (GDP). In
order to obtain the regression result, the OLS technique
with the help of the PC Give software was used.
The result obtained from the regression shows that there
is positive impact of Foreign Direct Investment (FDI) on
Gross Domestic Product (GDP) with a coeffcient of 4.0912.
However, this coeffcient is not statistically signifcant as
revealed by its corresponding standard error and t-values.
Hence, FDI is inelastic to GDP. This positivity in the
coeffcient of Foreign Direct Investment is in conformity
to the economic a priori expectation of a positive impact of
Foreign Direct Investment on the economic growth of the
economy (GDP).
Also, the regression result shows that the Government
Expenditure has a positive impact on GDP with a
coeffcient of 6.2835. The standard error and t-values
showed that this parameter is statistically signifcant. Thus,
the Government Expenditure is elastic to Gross Domestic
product. This positivity of the coeffcient of GOV conforms
to the economic a priori expectation of a positive impact of
Government Expenditure on GDP.
Furthermore, the result obtained from the regression
shows that Gross Fixed Capital Formation has a positive
impact on GDP. This is indicated in its positive coeffcient
of 1.5457. This coeffcient is revealed to be statistically
signifcant by the standard error and t-values. Thus, from
this it implies that Gross fxed Capital Formation is elastic
to GDP. The coeffcient of Gross fxed Capital Formation
being positive conforms to the economic a priori expectation
of a positive impact of GCF on the growth of the economy
vis-à-vis GDP.
Model 2
From the regressions result of model 2, the R-squared
(R²) value of 0.919443 shows that at 91.94% the explanatory
variables explain changes in the dependent variable. This
means that at 91.94% the independent variables explain
changes on Current Account Balance (BCA). This simply
means that the explanatory variables explain the behaviour
of the dependent variable at 91.94%. The calculated
F-statistics of 98.917 which is greater than the value in the
F-table (2.9751) implies that all the variables’ coeffcients
in the regression result are all statistically signifcant to
GDP.
The Durbin-Watson (DW) as shown in the regression
analysis is 1.72 which shows that there is the presence of
autocorrelation.
The above model tested the effect of three different
variables namely – Foreign Direct Investment (FDI),
Government Expenditure (GOV) and Gross fxed Capital
Formation (GCF) on Current account Balance (BCA). In
order to obtain the regression result, the OLS technique
with the help of the PC Give software was used.
The result obtained from the regression shows that
there is positive and signifcant impact of Foreign Direct
Investment (FDI) on Current Account Balance (BCA)
with a coeffcient of 7.0662. This coeffcient is statistically
signifcant as revealed by its corresponding standard
error and t-values. Hence, FDI is elastic to BCA. This
positivity in the coeffcient of Foreign Direct Investment
is in conformity to the economic a priori expectation of a
positive impact of Foreign Direct Investment on Current
Account Balance of the nation.
Also, the regression result shows that the Government
Expenditure has a negative impact on BCA with a
coeffcient of -0.49248. The standard error and t-values
showed that this parameter is not statistically signifcant.
Thus, the Government Expenditure is inelastic to Current
Account Balance. This negativity of the coeffcient of GOV
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
258 Business Intelligence Journal July
Discussion of Findings
The OLS regression analysis is carried out to determine
the impact of FDI, Government expenditure and Gross
fxed Capital Formation on GDP (proxy for economic
performance) and Balance of Payment through Balance
on Current Account (BCA), . Hence, GDP and BCA were
regressed on FDI, GOV and GCF. Though the impact of
FDI is of primary concern here, the other two economic
variables were included to serve as “control variables” to
check the overstating of the estimated coeffcient of FDI.
In model 3, the effects of two macroeconomic indicators,
infation and exchange rates were also examined. Hence,
FDI was regressed on infation and foreign exchange rates.
The results of the fndings show that FDI has positive
effect, though not statistically signifcant on GDP. In other
words, the infow of FDI into the Nigerian economy for the
stipulated period this research was carried out (1980-2009),
showed that FDI was not a major contributor to economic
growth of the nation. However, the fndings show that FDI
has positive and signifcant impact on BOP through current
account balance during the same period of analysis.
The effect of infation and foreign exchange rates on
FDI, brought under scrutiny, also showed that whereas
infation rate did not have major effect on the infow of
FDI into the Nigerian economy, foreign exchange rate had
great effect on the infow of FDI into the Nigerian economy
within the same period (1980-2009).
From the foregoing discussion, it should be pointed
out that although the government have made reasonable
efforts in attracting FDI, certain economic and political
circumstances prevalent in the country have hindered its
infow and its overall performance.
The primary objective of this study was to determine the
impacts and signifcance of FDI on the Nigerian economy
and the nation’s Balance of Payment (BOP). This was
achieved through the use of the OLS regression analysis of
data on the GDP, BCA, FDI, Government Expenditure and
Gross fxed Capital Formation sourced from the Central
Bank of Nigeria Statistical Bulletin.
The study also gave an opportunity for the examination
of the effects of infation rate and exchange rate on FDI. The
impact of Government expenditure on economic growth
was also examined. Therefore the following fndings were
revealed:
First, it was discovered that, FDI have not contributed
signifcantly to the economic growth of Nigeria in the
period under consideration.
conforms to the economic a priori expectation of a negative
impact of Government Expenditure on BCA.
Furthermore, the result obtained from the regression
shows that Gross Fixed Capital Formation has a positive
impact on BCA. This is indicated in its positive coeffcient
of 0.42403. However, this coeffcient is revealed not to be
statistically signifcant by the standard error and t-values.
Thus, from this it implies that Gross fxed Capital Formation
is inelastic to BCA. The coeffcient of Gross fxed Capital
Formation being positive conforms to the economic a priori
expectation of a positive impact of GCF on Balance of
Payment vis-à-vis BCA.
Model 3
From the regressions result of model 3, the R-squared
(R²) value of 0.666903 shows that at 66.69% the explanatory
variables explain changes in the dependent variable. This
means that at 66.69% the independent variables explain
changes on Foreign Direct Investment (FDI). This simply
means that the explanatory variables explain the behaviour
of the dependent variable at 66.69%. The calculated
F-statistics of 27.029 which is greater than the value in the
F-table (3.3541) implies that all the variables’ coeffcients
in the regression result are all statistically signifcant to FDI.
The Durbin-Watson (DW) as shown in the regression
analysis is 0.453 which shows that there is the presence of
autocorrelation.
The above model tested the effect of two different
variables namely –infation rate (INFL.) and Foreign
Exchange Rate (EXR.) on Foreign Direct Investment (FDI).
In order to obtain the regression result, the OLS technique
with the help of the PC Give software was used.
The result obtained from the regression shows that there
is negative and non-signifcant impact of infation on Foreign
Direct Investment (FDI) with a coeffcient of -310.46.
Hence, infation is inelastic to FDI. This negativity in the
coeffcient of infation is in conformity to the economic a
priori expectation of a negative impact of infation on FDI.
Again, the regression result shows that foreign exchange
has a positive effect on FDI with a coeffcient of 3731.5.
The standard error and t-values showed that this parameter
is statistically signifcant. Thus, the foreign exchange rate is
elastic to FDI. This negativity of the coeffcient of foreign
exchange rate does not conform to the economic a priori
expectation of a negative impact of foreign exchange rate
on FDI.
2011 259
Also, infation was found not to have any major effect on
the infow of FDI into the country. But exchange rate was
found to have major effect on FDI infow into the country.
In addition FDI contributed to Balance of payment
position through Current account balance. While Gross
fxed capital formation is inelastic to Balance on current
account.
In conclusion after the OLS regression analysis had been
carried out and with the study about the various factors
affecting FDI within the country, it is seen that:
1. There is no empirical strong evidence to support the
notion that Foreign Direct Investment has been pivotal
to economic growth in Nigeria; which could have
justify the effort of successive governments in the
country at using FDI as a tool for economic growth.
2. Governments direct involvement in the provision of
goods and services by establishing and controlling
corporations, for example, has contributed little
to economic growth in Nigeria. This justifes the
privatization policy of the various administrations in
our government to allow for the possible takeover by
investors (both foreign and domestic) of the government
corporations.
3. Though FDI has contributed signifcantly to Balance
of Payment (BOP) through the nation’s current account
balance. This is thus an effective measure of correcting
balance of payment disequilibrium in our economy.
Recommendations
The most signifcant factors that make Nigeria a good
host for FDI are her abundance in natural resources and
large population, indicating a large market.
The outcome of this study shows that though FDI was
not found to have signifcantly contributed to the nation’s
economic growth, if well harnessed it can contribute to
economic growth in Nigeria. To increase the infow of FDI
and its performance, the following recommendations from
this study are enunciated:
i. Balasubramanyam et al (1996) showed that most
economies beneft best from FDI when they are open to
foreign trade. Hence, the Nigerian government should
reduce the bureaucratic bottlenecks in foreign trade
especially the one constituted by the customs and port
authorities.
ii. Broensztein et al (1998) proved that there is a
high positive relationship between FDI and the level of
educational standard in the host economy. Based on this,
the country’s education should be in favour of science and
technology which would provide the economy with the
required skills that FDI require.
iii. Competitiveness should be encouraged, and as a
result, the existing and ‘yet-to-exist’ export processing and
free trade zones should be equipped with state-of-the-art
infrastructures and technologies.
iv. The infrastructures in the country need to be
enhanced to meet the needs/requirements of foreign
investors. For example, electricity should be provided at an
uninterrupted level to reduce the extra cost that investors
incur in the procurement of power generating sets coupled
with their maintenance. Also, good network roads and
adequate water supply should be provided so as to cut the
cost of investors doing business.
v. Appropriate measures should be placed to check
economic and fnancial crimes.
vi. The nation’s monetary authorities should develop
and implement measures that will ensure that both infation
and foreign exchange rates are sustained at levels that will
ensure increasing level of infow of FDI.
vii. The government and the private sector stakeholders
of our economy should consider harnessing infow of
FDI as a measure of improving the nation’s BOP through
current account balance, thereby ensuring the country’s
international competitiveness.
viii. Policy consistency should be emphasized.
References
Adegbite E.O and F.S. Ayadi (2010) “The Role of FDI
in Economic Development: A Study of Nigeria.”
World Journal of Entrepreneurship, Management
and Sustainable Development.Vol.6 No 1/2[Internet]
Available from www.worldsustainable.org
Akinlo, A.E. 2004. “Foreign direct investment and
growth in Nigeria: An empirical investigation”.
Journal of Policy Modeling, 26: 627–39
Alfaro, L., Chanda, A., Kalemli-Ozcan, S. & Sayek,
S. (2006). “How Does Foreign Direct Investment
Promote Economic Growth? Exploring the Effects
of Financial Markets on Linkages”. NBER Working
Paper no. 12522, National Bureau of Economic
Research, Cambridge, MA.
Aremu, J.A. 2005. “Foreign direct investment and
performance”. Paper delivered at a workshop on
Foreign Investment Policy and Practice organized
by the Nigerian Institute of Advanced Legal Studies,
Lagos on 24 March
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
260 Business Intelligence Journal July
IMF (1985) “Foreign Private Investment in
Developing Countries”, International Monetary
Fund, Washington .D.C.
Jenkins, C. and Thomas, L. (2002) “Foreign direct
investment in Southern Africa:determinants,
characteristics and implications for economic
growth andpoverty alleviation”, University of
Oxford [internet.] Available fromhttp://www.csae.
ox.ac.uk/reports/pdfs/rep2002-02.pdf
McAleese, D. (2004) Economics for business:
competition, macro-stability and globalisation. 3rd
ed. Edinburgh Gate: Pearson Education Limited
Musila, J.W. and Sigué, S.P. (2006) “Accelerating
foreign direct investment fow to Africa: from policy
statements to successful strategies”. Managerial
Finance, 32(7), 577-593.
Nadiri, M.I., 1993. “Innovations and Technological
Spillovers,” Working Papers 93-31, C.V. Starr
Center for Applied Economics, New York
University.
Nwillima N. (2008) “Characteristics, Extent and
Impact of Foreign Direct Investment on African
Local Economic Development”. Social Science
Research Network Electronic Paper Collection.http://ssrn.com Ragazzi, G. (1973) “Theories of the
Determinants of Direct Foreign Investment,” IMF
staff papers,20
Tsai, P.L. (1991) “Determinants of Foreign Direct
Investment in Taiwan: An Alternative Approach
with Time Series Data” World Development.
Van Pottelsberghe De La Potterie, B. and F. Lichtenberg
(2001), “Does Foreign Direct Investment Transfer
Technology Across Borders?” The Review of
Economics and Statistics Vol. 3 (3), 490-497.
Ayanwale A.B. (2007) “FDI and Economic Growth:
Evidence from Nigeria”. Africa Economic Research
Consortium Paper 165 Nairobi
Balasubramanyan, V., N. M.A. Salisu and D. Sapsford.
(1996). “Foreign Direct Investment and Growth
in EP and IS Countries”, Economic Journal, 106:
92–105
Borensztein, E., J. De Gregoria and J. Lee. 1998. “How
does foreign investment affect economic growth?”
Journal of International Economics, 45(1): 115–35.
Carkovic, M and Levine, R. (2002) “Does Foreign
Direct Investment Accelerate Economic Growth”
University of Minnesota Working Paper,
Minneapolis.
Collier, P. and Dollar, D. (2001) “Development
effectiveness: What have we learnt”? Development
Research Group, the World Bank[internet].
Availablefromhttp://www.oecd.org.
Dauda, R. O.S. (2007) “The Impact of FDI on Nigeria’s
Economic Growth: Trade Policy Matters”. Journal
of Business and Policy Research Vol. 3, No.2, Nov
– Dec
Deepak M., Mody A. and Murshid A.P. (2001)
“Private Capital Flows and Growth”, Finance and
Development. Vol. 38, June ,No.2.
De Mello L.R. (1997) “Foreign Direct Investment-
Led Growth: Evidence from Time Series and Panel
Data”, Oxford Economic Papers, 51: 133- 151.
Dunning J. (1993) Multinational Enterprises and the
Global Economy. England:Workingham Addison
Wesley Publishing.
Ekpo, A.H. 1995. “Foreign direct investment in
Nigeria: Evidence from time series data”. CBN
Economic and Financial Review, 35(1): 59–78
2011 261
Appendix I
REGRESSION RESULT
PcGive 8.00, copy for meuller session started at 13:39:56 on 24th October 2010
Data loaded from: alexpr~1.wks
EQ( 1) Modelling GDP by OLS - The present sample is: 1 to 30
Variable Coefcient Std.Error t-value t-prob PartR2
Constant 1.6709e+005 1.9847e+005 0.842 0.4075 0.0265
FDI 4.0912 2.6086 1.568 0.1289 0.0864
GOV._EXP. 6.2835 0.61381 10.237 0.0000 0.8012
GFC 1.5457 0.50454 3.063 0.0050 0.2652
R2 = 0.989607 F(3, 26) = 825.24 [0.0000] s = 820013 DW = 2.74
RSS = 1.748293983e+013 for 4 variables and 30 observations
EQ( 2) Modelling BCA by OLS - The present sample is: 1 to 30
Variable Coefcient Std.Error t-value t-prob PartR2
Constant -1.3500e+005 1.1447e+005 -1.179 0.2489 0.0508
FDI 7.0662 1.5046 4.696 0.0001 0.4590
GOV._EXP. -0.49248 0.35404 -1.391 0.1760 0.0693
GFC 0.42403 0.29101 1.457 0.1571 0.0755
R2 = 0.919443 F(3, 26) = 98.917 [0.0000] s = 472972 DW = 1.72
RSS = 5.816258697e+012 for 4 variables and 30 observations
EQ( 3) Modelling FDI by OLS - The present sample is: 1 to 30
Variable Coefcient Std.Error t-value t-prob PartR2
Constant -14108. 58549. -0.241 0.8114 0.0021
INFL. -310.46 1678.9 -0.185 0.8547 0.0013
EXR 3731.5 538.18 6.934 0.0000 0.6404
R2 = 0.666903 F(2, 27) = 27.029 [0.0000] s = 155120 DW = 0.453
RSS = 6.496770871e+011 for 3 variables and 30 observations
Ehimare O. A. - Foreign Direct Investment and its Effect on the Nigerian Economy
Omankhanlen Alex Ehimare
Business Intelligence Journal - July, 2011 Vol.4 No.2
262 Business Intelligence Journal July
ARTIFICIAL LEARNING AND SUPPORT VECTOR MACHINES:
DEFAULT RISK PREDICTION
1
FABIO FRANCH, Ph.D.
2
University of Trento (Italy)
471 4th Ave Apt F – Chula Vista, CA 91910
Email: [email protected]
Abstract
The fnancial crisis that started in 2008 has shown how much work has still to be done in order to precisely predict the bankruptcy
of those actors who ask for credit to their banks. In this paper I focus my attention on large companies, using a database kindly
provided by Unicredit, one of the most important European banking groups. The size and the complexity of the problem has required
the simplifcation of the database and the use of Principal Component Analysis (PCA) in order to reduce the problem to a dimension that
is manageable by the Support Vector Machine (SVM) software chosen for this study. The best confguration found allowed the correct
classifcation of 84% of all companies and such results are found to be higher than many other reported in the literature. Key words:
Bankruptcy prediction, Basle 2, Default Risk, Principal Component Analysis, Support Vector Machines.
1
Master thesis presented at the University of Trento, under the valuable supervi-
sion of Professor Alessandro Zorat of the University of Trento. Final grade:
110/110 cum laude.
2
Fabio Franch earned his Ph.D. from West Virginia University and studied at Uni-
versity of Trento, where he successfully completed his studies (BA and MA) in
2007. He was also an exchange student at the Abo Akademi in Turku, Finland.
His interests are machine learning and the dynamic analysis of economic phe-
nomena.
3
For applications of artifcial learning techniques to fnancial and economic
problems, see Zhang, G.Peter 2004, Deforche, Koen 2004, Fauci F. 2004,
Fernandez-Rodriguez, F.,Gonzalez-Martel, C., & Sosvilla-Rivero, S. 2000, Fu-
kushima, K. 1975, Gencay, R. 1996, Kuoa R.J., P. Wub, C.P. Wangc 2002, Ro-
driguez, Claudia P. 2004, Rosenblatt, F. 1962, Scarselli, Franco, Sweah, Liang
Yong, Markus, Hagenbuchner, Ah Chung Tsoi 2005, Tewoldeberhan, Tamrat
W., Alexander, Verbraeck, Simon S. Msanjila 2005, Tortorella, Francesco 2002,
Viswanathan, Murlikrishna, Young-kyu Yang, e Taeg Keun Whangbo 2005,
Wan, H. A.,Hunter, A., & Dunne, P. 2002. For applications of artifcial learning
for default risk prediction, see Yongqiao Wang, Shouyang Wang, and K. K. Lai
(…) it doesn’t matter whether the forecast of the future is
true or not. If a mathematician should predict a long and
happy reign for me, a time of peace and prosperity for the
Empire -- Eh, would that not be well? (Asimov, 1988)
Bankruptcy prediction has been studied by many
researchers from different felds
3
and it is now getting
more and more important due to the Basel 2 Agreement
4
and the ongoing fnancial and economic crisis. Despite
the relatively high precision achieved by more traditional
statistical techniques, much more is needed to reach
perfection at predicting bankruptcy.
I decided to apply Support Vector Machines (SVMs) on
the prediction of corporate defaults for several reasons
5
:
1. the corporate sector is the set of companies
6
that is
affected by a relatively higher default rate
7
.
2. the prediction of the risk of default of large companies
using data from the balance sheet is far from perfection.
In order to accomplish such objective, I decided to turn
to a big fnancial group which already has at its disposal a
large amount of information about its customers as required
by the Basel 2 Agreement. According to this fnancial
agreement, banks commit themselves to store an amount
of resources for each loan that refects as accurately as
possible the risk connected to that particular operation.
Unicredit collects a large amount of information in
order to decide whether they deserve a loan or not. These
data enable them not only to describe potential customers
2005, Martin-del-Brio, Udo, G. 1993, Odom, M. & Sharda, R. 1990, Barniv
R., A. Agarwal, and R. Leach 1997, Bel et al. 1990, Rudorfer, Gottfried 1995,
Ohlson, J. 1980, Kaski Samuel, Janne Sinkkonen, and Jaakko Peltonen 2001.
4
The Basel 2 Agreement has been in force since 2007 for most banks in Europe.
Countries members of the Basel 2 Agreement are Belgium, Canada, France,
Germany, Italy, Japan, Luxemburg, Netherlands, Spain, Sweden, Switzerland,
the United Kingdom and the United States of America. To know more on the
Basel 2 Agreement, see International Financial Risk Institute (IFRI) 2000,
Moody’s 2006, Panizzolo Davide 2006, Ronchino Beatrice 2005.
5
The Probability of Default (PD) is usually estimated for a homogeneous group
of borrowers (based on the type of loan and features of the borrowers) and rep-
resents an assessment of the percentage of borrowers that are going to bankrupt
within a year. It entails the assessment of an expected value and an unexpected
value, which is used as safety margin.
6
These companies are all Unicredit customers.
7
These information is related to the business activity earlier than year 2007.
2011 263
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
in quantitative terms, but also to track the change of the
fnancial status of the borrowers over time.
However, the protection of such information is vital
for the survival of the fnancial group in the competitive
arena of today’s markets. In other words, Unicredit has to
protect the identity of its customers from its competitors
(for obvious reasons of maintaining its customer base), but
it has also to do it in order to avoid that its competitors
discover its market strategies and its policies and nullifes
the advantage that comes with them.
Moreover, Unicredit protects the identity of the
variables it has selected as most relevant to predict a
correct value of the risk of default, in order to preserve its
competitive advantage that comes from a better estimate
of the risk, which leads to a lower amount of resources to
be stored and to an increase of available resources which
can be invested in a more proftably way. From this it
follows that the database I was presented does not contain
any label that provides a clear description of what those
variables represent. Thus, it is not possible to state whether
that variable is about the net income, or gross sales, or any
other variable that Unicredit deemed necessary to obtain a
better assessment of the probability that a particular record
(company) will fulfll his obligations.
Despite of the lack of information about the meaning of
each column, the database contains a wide large amount of
information for each client. In fact, it is possible to identify
registry, quantitative, and qualitative pieces of information,
all of which refer to year 2003
8
.
8
Unicredit kindly provided the same data also for year 2004. However, I decided
to focus only on year 2003 in order to see how much SVMs can be useful to
predict a bankruptcy event using the fnancial, economic data in a precise time
of the year (the end of the year), without considering their evolving over time.
Please note that a longer time span is required to predict the Probability of De-
fault (PD), as regulated by the Basel 2 Agreement.
It is possible to fnd data relative to:
1. the specifc industry of each borrower;
2. place and date of foundation;
3. current year;
4. bank account information of the registered offce and of
the local offces;
5. borrower’s county of residence;
6. county to which belong the Chamber of Commerce in
charge of taking care of each borrower;
7. amount of employees (as of the end of the year);
8. total amount of employees;
9. total amount of seasonal labor;
The database presents also 40 columns that monitored
the conformity to some unknown criteria).
Another section follows (labeled as EL_FOND) which
contains data about budget funds fows and has numerical
values
9
. The section labeled as “EL_VOCERIC” is 300
columns long and provides information about budget items
and reclassifed budget items. Another section follows,
which defnes the type of reclassifcation used in the balance
sheet and then 6 columns, labeled as “GPL_SEM” and an
other set of columns called “GPL_SEM”
10
.
At the end of the database it is possible to fnd the
columns labeled as “SEZ_12”, the ones that describe the
typology of the borrowers (even though it is unclear how
borrowers are classifed) and then a column that states
whether that particular borrower was able to comply with
his/her fnancial obligations
11
.
SVM: Implementation, Pre-Processing
And First Results
SVMs are artifcial learning techniques based on the
linearity assumption in a multidimensional space and on
the optimization theory. This theory implements a learning
method which fnds its origins in the statistical learning
theory. This learning method was frst introduced by Vapnik
and others (Burges 1998; Cristianini and Shawe-Taylor
2000) and has enabled the resolution of different kinds
of problems since its earlier adoption, offering a superior
performance to other statistical techniques (Abbasi and
Chen 2005; Boyacioglu, et al. 2009; Drucker et al. 1999).
Once all observations (or examples) have been collected
for a specifc problem (including the outputs or labels, the
values to be correctly predicted), a researcher can make a
decision concerning the hypothesis regarding the specifc
problem. Among these, linear hypotheses are the ones that
can be more easily understood and used in order to explain
how SVMs work.
9
In this case (zero means that there are no funds, thus it is not a missing value.
10
It is not possible to understand the meaning of the data reported in these col-
umns.
11
Default or no default. It is a dummy variable.
10. total amount of managers;
11. legal expiry date of the bank loan;
12. operational expiry date;
13. legal structure;
14. goods price/value (Euros/Italian Liras);
15. budget type/form;
Business Intelligence Journal - July, 2011 Vol.4 No.2
264 Business Intelligence Journal July
( ) ( ) ( )
y
x
y
x
l
l
S , ,..., ,
1
1
=
where
?
*
is the solution of the following quadratic
optimization problem
( )
l i
to subject
W
l
i i
i
j i j
l
j i
i
j i
l
i
i
y
x x
y y
,..., 1 , 0
0
2
1
max
0
1 , 1
= ?
=
? ? =
?
? ?
=
= =
?
?
?
? ? ?
the vector
?
=
=
l
i i i
i
x
y
w
1
*
?
is the maximum margin
hyperplane with a geometric margin equal to:
.
1
2
w
= ?
According to the Karush-Kuhn-Tucker condition, the
optimal solution needs to satisfy the condition
( ) | | . ,..., 1 , 0 1
* * *
l i
b x w
y
i i
i
i
= = ? + ?
?
\which implies that only to input
xi
for which the
functional margin is equal to 1 and which are in the
proximity of the hyperplane, correspond values of
?
*
i
which are different from zero.
All the other parameters
?
*
i
are equal to zero (Burges
1998; Cristianini and Shawe-Taylor 2000).
It follows that in the expression of the weights vector
only the points close to the hyperplane matter (see Figure
1), and it is for this reason that these points are termed
‘support vectors’.
Figure 1: Example of Support Vectors
WINSVM
12
was chosen as the tool to be used in this
study thanks to its Windows interface and its optimization
function, which enables the generation of a random set of
parameters and the testing of the SVM
13
with the generated
parameters.
The database presented several issues:
1. missing values: this issue was solved by deleting
records or columns that contained one or more of such
values;
2. redundancy of information: issue solved by deleting all
the columns that had values which were equal for all
50,000 records;
3. redundant information: Unicredit itself reported the
uselessness of using some pieces of information to
predict the risk of default;
4. mix of qualitative and quantitative values: issue
solved through the binarization of qualitative variables
(columns).
Given the size of the database and the insuffcient
computing power to delete/flter records or columns of the
entire database, I sorted all remaining records according to
the outcome of the bank loan operation (default/no default).
12
WINSVM is a Windows-based software which allows regression and classif-
cation. Seehttp://www.cs.ucl.ac.uk/staff/M.Sewell/winsvm/ for instructions on
how to use it. Other useful pieces of software are SVM and Kernel Methods
Matlab Toolbox (http://asi.insa-rouen.fr/~arakotom/toolbox/index.html), Near-
est Point Algorithm (http://guppy.mpe.nus.edu.sg/~mpessk/npa.shtml), SVM-
QP (https://projects.coin-or.org/SVM-QP), 2D Svm Interactive Demo (http://
www.learningwithkernels.org/software.html), and others (seehttp://www.ker-
nel-machines.org/software.html,http://www.support-vector.net/software.html,http://www.support-vector-machines.org/SVM_soft.html).
13
See Cristianini, Nello, John, Shawe-Taylor 2000 for a detailed description of
the theoretical assumptions on which is based the learning of a Support Vector
Machine.
In a linearly separable sample S
2011 265
Given this positive results, I then extended the dataset
to 1353 records, of which 60% for training and 40% for
testing. However, this time the optimization process was
more intensive than before; as a result, the system crashed.
Despite such an event, several confgurations have been
saved
18
: I decided then to run WINSVM on two computers
(similar to the computer described above), aiming at
obtaining as many confgurations as possible. This is
possible thanks to the random generation of the parameters
and their accuracy testing process. This operation required
The Unicredit database was further simplifed by:
1. deleting columns that the Unicredit Credit Risk
manager himself suggested as useless for our purpose;
2. keeping only columns whose values were expressed in
Euros;
3. keeping only the records of borrowers that reside on the
Italian territory;
4. deleting records which contain one or more missing
values;
5. deleting records that had more than 4 marks in the
section preceding the one labeled as “EL_FOND”
17
.
6. deleting of columns which presented constant values
for all records.
The Unicredit database, at the moment of its handing,
contained around 50,000 records, but only a small
percentage described companies which turned out to be in
fnancial troubles. After the pre-processing phase and the
simplifcation mentioned above the database included 2254
records and 607 columns. I then labeled “0” all records
which refer to bankrupted enterprises and “1” to all the
remaining ones.
I went on testing the effectiveness of the SVM by
removing all registry data, which required the binarization
and therefore an explosion of the amount of columns. The
goal at this stage was to test the computing power needed
to process the entire database. I randomly chose 79 records
for the training phase and 11 records for the testing. On
an ASUS laptop, 2.8Ghz processor, 512Mb RAM the
optimization phase lasted between 2 and 7 minutes, using
14
50% of the database constituted companies that paid back the loans after one
year, the remaining 50% were not able to do it.
15
A much larger amount of borrowers were able to fulfll their obligations due to
the fact that customers that received the loans were those that Unicredit deemed
deserving and enough wealthy to borrow money.
16
This provides a better assessment of which company will default and which one
is not going to.
17
Some records had more than 4, which was against the rules that were set by Uni-
credit before creating the database. This is probably due to wrong data entries
or misunderstandings within the organization itself.
both the regression and the classifcation. Less than a second
was the time required to train and test the SVM using the
selected parameters.
The SVM performed really well on this small dataset
(see Table I).
Average loss : 0.12007405
Avg. loss pos : 0.22013576 (6 occurences)
Avg. loss neg : 0 (5 occurences)
Mean absolute error : 0.5098412
Mean squared error : 0.39600989
Accuracy : 0.90909091
Precision : 0.83333333
Recall : 1
Predicted values:
| + | -
---+-----+-----
+ | 5 | 0 (true pos)
- | 1 | 5 (true neg)
Table I: output of the testing phase with 79 training samples and
11 testing samples.
I then selected all the records regarding those borrowers
that were not able to fulfll their obligations and an equal
amount of borrowers that were able to repay the loan
14
. This
was done to reduce the size of the database and to balance
the amount of records of the borrowers that did not pay
back their loans
15
. The equal size of the two classes (default
and non default) is supposed to enhance the capability of
the Support Vector Machine to understand the difference
between the two classes
16
.
18
Some of the confgurations lead to high accuracy predictions: accuracy is the
percentage of positive values and negatively values that were respectively pre-
dicted by the SVM as positive and negative out of the total of values.
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
266 Business Intelligence Journal July
19
200 confgurations were deemed suffcient to proceed with the analysis.
20
After this simplifcation, the database had 1799 records and 582 columns. We
assumed that the SVM would have probably performed badly with such a high
number of columns; thus, we deemed Principal Component Analysis necessary
to signifcantly reduce the amount of columns.
21
Unicredit provided data that was mostly about a particular feld it chose as the
most troublesome given the relatively higher percentages of bankruptcy.
I then
20
:
1. deleted every record (company) which operated in a
different industry
21
;
2. deleted the information about the industry to which
belongs each borrower;
3. deleted the data about the Chamber of Commerce and
the statistical ID of each borrower: this information
was deemed to be useless to predict the Probability of
Default;
4. deleted all the columns that presented data that was
used only to identify in some way each record;
5. deleted all the columns about employees: too many
missing values are in fact included in these columns;
6. added qualitative information that was removed in the
previous step;
Principal Component Analysis (PCA)
The original database was subject to a simplifcation
process that only marginally reduced the amount of columns
due to the assumption that a lower amount of them will
enhance the performance of the SVM when it processes
new samples. However, the steps described in the previous
section results to be insuffcient to signifcantly reduce the
complexity of the database.
I therefore decided to use Principal Component Analysis.
This statistical technique generates new factors that are
then chosen as axis of bi-dimensional maps which are then
interpreted in order to understand the degree to which these
factors can actually explain the underlying phenomenon.
Factors are generated by a correlation matrix; this
matrix is then processed in order to have as many factors
as the number of variables included in the process. Factors
have the property of being independent and are linearly
constructed using the original variables (Bowler 2002;
Bonifazzi 2006; Easton 2009).
Factors are then sorted according to their ability to
provide the highest amount of information, or, in other
words, to better explain the phenomenon. According to the
literature, a specifc amount of factors can be selected on
the basis of criteria like the following:
1. keep all factors which have an auto-value greater than
1;
2. keep all factors that can explain at least 66% of the
variability;
3. keep all factors that can explain at least 75% of the
variability;
4. keep those factors that are located on the left part of
the graph generated after choosing the variables to be
processed to generate the new factors.
5. keep all factors that are deemed to hold important
information (Mortara 2006).
It is important to remember that the software used for
this purpose, INSTAMAP
22
:
7. deleted all the columns that included the same value in
all records.
6 to 7 hours, because the execution of SVM had to be
repeated several times in order to obtain a large amount of
confgurations
19
.
I then tried several confgurations of parameters like C,
epsilon, nu, Kernel that led to a testing MSE that is less
than 0.01 and I located the one that returns the highest
accuracy value in the training phase. Such a confguration
has a radial Kernel, gamma equal to 0.3, C equal to 10 and
epsilon equal to 0.0001. Although this one was the best
confguration, it performed poorly in the testing phase:
accuracy is in fact rather low, 69%. I concluded from this
analysis that using fewer training samples and even less
testing samples made it easier for the SVM to recognize
the hyper-plane that correctly classifed 10 out of 11 testing
items. Raising the number of training samples apparently
led the SVM to defne a hyper-plane that did not separate
correctly the two classes of companies. It was plausible to
assume that such an issue was due to the higher complexity
of defning a multi-dimensional hyper-plane, using a much
larger amount of variables. In other words, the SVM seemed
to get confused by the excessive amount of data.
Thus, I proceeded further simplifying the database,
looking for a meaningful criterion which would allow
the removal of a considerable amount of columns while
keeping the number of training and testing samples as high
as possible.
22
See Mortara V. (2006) for more information.
1. cannot read column labels that are longer than 5
characters;
2011 267
Average loss : 0.55649621
Avg. loss pos : 0.18958699 (181 occurences)
Avg. loss neg : 0.92340542 (181 occurences)
Mean squared error : 0.99198794
Accuracy : 0.47513812
Precision : 0.40425532
Recall : 0.10497238
Predicted values:
| + | -
---+------+------
+ | 19 | 162 (true pos)
- | 28 | 153 (true neg)
Accuracy is at this point very low, less than 48%, with
mostly negative value only predicted. Thus, I decided to test
the SVM performance using all generated confgurations.
Only one confguration performed really well, with an
accuracy of 81.49%. I tried then to improve its performance
Table III: output of the classifcation procedure with the best set
of parameters
*** mySVMversion 2.1.4 ***
----------------------------------------
Predicting
Average loss : 0.40571761
Avg. loss pos : 0.75934415 (181 occurences)
Avg. loss neg: 0.052091077 (181 occurences)
Mean absolute error : 0.98534041
Mean squared error : 0.99136138
Accuracy : 0.62430939
Precision : 0.57525084
Recall : 0.95027624
Predicted values:
| + | -
---+------+------
+ | 172 | 9 (true pos)
- | 127 | 54 (true neg)
Table II: testing with the optimal confguration
Accuracy was at this point even lower than the one
previously estimated. Thus, I decided to test all the other
confgurations, but with no improvement in the testing
performance.
At this stage, the average loss was rather high in any
experiment conducted. Having an average loss equal to
0.5 means that predictions are greater than or less than the
correct value to be predicted by an average of 0.5. However,
the best confguration appeared to be the frst one, to which
corresponds an average loss equal to 0.35784271.
Given the poor performance of regression, I decided
to use classifcation. I chose to keep the confgurations
generated with regression, but I had to change the values
of the last column of the database
23
from “0” (no default
occurred) to “-1” (a default event occurred in the short),
while keeping all remaining labels the same.
I started again from the confguration of parameters that
performs the best in the training phase. The output of the
classifcation procedure with the set of parameters with the
lowest MSE is presented in Table 4.
23
Default “yes” or default “no”.
2. cannot process a large amount columns at a time:
I therefore chose to split the database in several
subsets containing 19 columns each and a last one
which included also the fnal column containing the
information about the occurring of the default event for
each record.
In this study I decided to keep as many factors as were
needed in order to explain 66% of the variability of the
data. Doing so reduced the size of the columns to 180 to
which I had to add the 75 columns of qualitative data that
were spared from the factor generation process given their
sparseness.
All factors were then copied into a new spreadsheet,
saved later in the TXT format which is needed in order
for WINSVM to read the data. I chose to use 80% of the
records for the training phase and the remaining 20% for
testing.
WINSVM crashed again due to the large amount of
information presented, but only after having generated
more than 300 possible optimal confgurations. I decided
that such an amount of options was enough to have an idea
of which Kernel can perform the best.
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
268 Business Intelligence Journal July
24
This confguration uses a very large Soft Margin, which is used to cope with
the outliers and the noise, all problems that probably affected the Unicredit
database.
by marginally decreasing or increasing one or more of its
parameters (see below). Values of epsilon equal to 0.01,
0.2, 0.15, 0.17, 0.18, 0.19 generated an accuracy that never
exceeded 81%.
I observed that with an epsilon value of 0.18 it was
possible to get the least average loss. I kept this value
of epsilon and changed C and gamma to optimize the
performance of the SVM. However, it was not possible to
refne the predictions using as starting confguration the
one whose MSE is equal to 0.2637. The best confguration
using the Radial Kernel had therefore a C value that is
equal to 1000000, an epsilon value of 0.1 and gamma equal
to 0.01
24
. In short, the performance of Radial Kernels was
usually very high in the training phase, but vey low in the
testing phase, which led us to think that Radial Kernels
defne a hyper-plane that is overly based on the specifc
training data it processed during the training. I needed to
fnd a confguration that is able to perform better in the
testing phase, which would support my claim of robustness
of the fndings of this work.
Thus, I went on testing other Kernels, choosing again
the confgurations to which corresponded a low error
(MSE). I chose to analyze the performance of Polynomial
Kernels. The best confguration seemed to be the one that
has a MSE equal to 0.85472, C equal to 1, epsilon equal to
0.0001, third degree. MSE was rather high, still it managed
to correctly classify all 1437 records in the training phase.
However, results in the testing phase were rather bad, with
an accuracy of almost 64%. Once again, I tried to optimize
the confguration of parameters, but only reducing the
degree from 3 to 2 led to an improvement of results. Not
even raising the C value positively affected the performance
which was stationary around an accuracy level equal to
71% (see below).
Table IV: accuracy in the testing phase with polynomial Kernel
I decided then to switch to an Anova Kernel, starting once
again from the best confguration and trying to optimize its
parameters in order to improve the accuracy in the testing
phase. The best confguration seemed to be the one that had
a MSE equal to 0.0691369, C equal to 100, epsilon equal to
0.0001, gamma equal to 0.9 and degree equal to 4. To such a
confguration of the SVM corresponded an accuracy equal
to 77%. Once again, I optimized the parameters, aiming at
fnding a better solution. It turned out that a high value of
gamma (gamma=30) improved the prediction’s accuracy of
6 percentage points
25
(see below).
25
Accuracy is thus equal to 83%.
Change in parameter Testing results
Degree=2
Average loss : 1.6596905
Mean squared error : 126.33603
Accuracy : 0.70718232
Precision : 0.69230769
Recall : 0.74585635
Change in parameter Testing results
Degree=2, C=0.0001
Average loss : 1.6596905
Mean squared error : 126.33603
Accuracy : 0.70718232
Precision : 0.69230769
Recall : 0.74585635
Degree=2
Average loss : 1.6596905
Mean squared error : 126.33603
Accuracy : 0.70718232
Precision : 0.69230769
Recall : 0.74585635
Degree=4
Average loss : 0.68685004
Mean squared error : 46.331544
Accuracy : 0.70718232
Precision : 0.68115942
Recall : 0.77900552
Change in parameter Testing results
Gamma=1
Average loss : 0.37226828
Avg. loss pos : 0.42555524 (238 occurences)
Avg. loss neg : 0.26999171 (124 occurences)
Mean squared error : 1.396154
Accuracy : 0.77071823
Precision : 0.71875
Recall : 0.88950276
Table V: testing with Kernel Anova
2011 269
I observed at this point the superior performance of
SVMs with a Neural Kernel. Almost all randomly generated
confgurations led to very accurate predictions. Moreover,
the last prediction has an accuracy that is almost equal to
84%, the highest observed so far. To such a high accuracy
corresponds a low number of misclassifed records. This
high-performance confguration was also characterized
by a very low average loss and especially by the highest
precision level
26
. Such a high value of precision is due
to the low amount of negative values that were predicted
as positive. Only the recall
27
value was not the highest
among those I observed until that moment. This was due
to the relatively higher amount of positive values that
were predicted as negative. Thus, I concluded that such
a confguration was the optimal one, given the highest
accuracy and precision achieved. The sharp decline of the
accuracy level when testing for robustness of the results
confrms the complexity of the problem under investigation
26
Precision is defned as the positive values that have been predicted as such out
of all values that have been predicted as positive even if they were negative.
27
Recall is defned as the amount of positive values predicted as such out of all
“real” positive values.
At this point I was not still satisfed with the results
obtained. I decided to see what the use of a Neural
Kernel entailed. I started from the confguration to which
corresponded a MSE equal to 0.314255, C equal to 0.01,
epsilon equal to 0.00001, a equal to 0.4 and b equal to 0.9
and a prediction accuracy equal to 81%.
Such accuracy was already interesting in itself; however,
I wanted to see if there was any way to further increase its
accuracy (see below). After having tried several options
by modifying only epsilon, I was unable to improve the
accuracy and only in one case the accuracy was as high
as in the former confguration. Thus I decided to increase/
decrease a and b and to keep all remaining parameters at
the same level (see below).
Change in parameter Testing results
Gamma=2
Average loss : 0.32014074
Avg. loss pos : 0.39044826 (234 occurences)
Avg. loss neg : 0.1916098 (128 occurences)
Mean squared error : 1.043599
Accuracy : 0.79281768
Precision : 0.73451327
Recall : 0.91712707
Gamma=30
Average loss : 0.2352293
Avg. loss pos : 0.34048536 (203 occurences)
Avg. loss neg : 0.10084578 (159 occurences)
Mean squared error : 0.60141935
Accuracy : 0.82596685
Precision : 0.76818182
Recall : 0.93370166
Variation in a e b
values
Risultati del testing
a=1
b 0.1
Average loss : 0.24368945
Avg. loss pos : 0.34219957 (198 occurences)
Avg. loss neg : 0.12475651 (164 occurences)
Mean absolute error : 0.65285786
Mean squared error : 0.60010934
Accuracy : 0.8121547
Precision : 0.7627907
Recall : 0.90607735
Table VI: results in the testing phase modifying the reported
parameters. C and epsilon are held constant.
Variation in a e b
values
Risultati del testing
a 0.1
b 0.1
Average loss : 0.20976328
Avg. loss pos : 0.31135803 (201 occurences)
Avg. loss neg : 0.08292759 (161 occurences)
Mean absolute error : 0.62960401
Mean squared error : 0.5516383
Accuracy : 0.83149171
Precision : 0.77272727
Recall : 0.93922652
a 0.1
b 1
Average loss : 0.19665233
Avg. loss pos : 0.20571275 (186 occurences)
Avg. loss neg : 0.18707711 (176 occurences)
Mean absolute error : 0.64513177
Mean squared error : 0.55654285
Accuracy : 0.83977901
Precision : 0.83606557
Recall : 0.84530387
Predicted values:
| + | -
---+------+------
+ | 153 | 28 (true pos)
- | 30 | 151 (true neg)
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
270 Business Intelligence Journal July
in this work. In fact, when using 30% of the selected data
for testing (up from 20%), the accuracy slightly drops to
79%. An additional ten percentage points in accuracy are
lost when using 40% of the data for testing purposes.
Discussion
It is important to understand the meaning of the
predicted values in order to measure the impact of using
such a model. I shall recall that, using classifcation,
predicted values range from “-1” to “+1”, where “+1” is
given to records to which corresponds an event of default
and “-1” in all other cases (no default). That means that
those 28 misclassifed records represented companies that
appeared to Unicredit as solid and deserving credit, but that
were at the end unable to fulfll their obligations, which
is exactly what Unicredit aimed at minimizing
28
. I should
therefore conclude that, after all, I should not choose such a
confguration, since I am only interested in having the least
number of companies possible that borrow money, even if
they cannot pay it back (a confguration that misclassifes
28 companies is therefore not the best one).
In order to draw even better conclusions, I have to
understand what the fact of having only 30 misclassifed
companies implies. The 30 companies predicted as positive,
but negative in reality, represent in fact companies that
the model classifes as weak and therefore not deserving
credit, but which are actually solid enough to fulfll their
obligations.
A large banking group like Unicredit should minimize
also such an occurrence, since it represents a loss of
customers. However, in a real world situation an important
banking group like Unicredit has a rather wide customer
base, which enables it choose to lend money only to those
companies that are clearly able to pay their loans back. In
such a situation, the confguration to which corresponded
an accuracy value of 84%
29
appears to be the optimal in
order to minimize losses and maximize revenues.
Conclusion
The purpose of this paper is to determine whether SVMs
can be helpful to predict companies’ default risk. This study
had to deal with several issues, among which the most
critical one was the lack of knowledge of the meaning of
28
This strategy is currently being pursued.
29
The confguration which uses a Neural Kernel.
the variables included in the database. Such an issue turned
out to be irrelevant, if I consider that PCA and the SVM
processed the data in such a way that the most important
variables were not rejected thanks to their informative
content.
Given an accuracy level equal to 84%, I can state that
SVMs can successfully be used to improve the ability to
predict their ability to fulfll their obligations.
The results achieved here appear to be even superior to
those found in the literature (see below).
Table VII: prediction accuracy in Yongquiao Wang, Shouyang
Wang, and K. K. Lai 2005
30
Source: Yongquiao Wang, Shouyang Wang, and K. K. Lai 2005
In conclusion, I believe that it is possible to further
enhance the performance of the classifcation process in
several ways, by:
1. performing again PCA, after having removed the most
infuential outliers. This would marginally reduce
the amount of records, but it would also increase the
variance of the data, making the task of fnd a better
hyperplane easier for the SVM. I chose not to go this
way because I believed that the qualitative information
would have, in some way, justifed the values
corresponding to the outliers.
2. performing again the PCA and reducing the informative
content to a lower level (less than 66%). I did not go
this way because the literature advises against such an
option
31
.
30
These results are the best ones obtained by Yongquiao Wang, Shouyang Wang,
and K. K. Lai 2005. They used a dataset containing 653 records and 15 features.
31
As reported in Mortara V. 2006.
The rating class is the class a borrower belongs to given its ability to fulfll its
obligations. The more solid a company is, the higher its ranking will be –AAA,
BB and so on –.
2011 271
3. processing as many records as possible, therefore
including a larger amount of records of companies that
were actually able to pay their loans back. I believe
that such an option could lead to some interesting
results, after having observed that most of the times the
amount of calculated Support Vectors was almost equal
to the amount of training samples. This means that the
SVM “considers” each training sample as “special”
and calculates a Support Vector for (almost) each of
them. In other words the hyper-plane is very close to
most of the training samples, which led us to think that
increasing the amount of training samples would help
the SVM defning a more accurate hyper-plane.
In conclusion, I have to point out that the predicted
values cannot directly be interpreted as the Probability of
Default (according to Basel 2 regulation, PD has also to
be calculation on a 5-year base at least: I only used data of
year 2003). These values have rather to be interpreted as a
measure of reliability of stating that a particular company
will bankrupt or not. In particular, the closer the predicted
value to “-1” or “+1” is (or even greater than 1), the more
credible/reliable the statement about the possible default
of that specifc company. The closer the predicted value to
zero, the more unreliable such a statement.
It would have been interesting to try to predict the rating
class of each borrower
32
, but to do so I would have needed
to:
1. have the information of the rating class of each
borrower, needed to train the SVM on them;
2. deal with multi-class classifcation problem, which
involves the use of as many SVMs as the number of the
rating classes.
Given the high accuracy I obtained by using a Neural
Kernel, I can state that SVM can proftably be used to
predict the fnancial solidity of potential borrowers. SVMs,
in other words, offer themselves as an additional tool with
which to fght and avoid dangerous fnancial crises like the
one that been plaguing the international community since
the second half of year 2008.
References
Abbasi, A. and H. Chen (2005) "Applying authorship
analysis to extremist-group web forum messages."
IEEE Intelligent Systems 20(5): 67-75.
Asimov, Isaac (1988) Prelude to Foundation, New
York: Bantam Books, p. 2.
Barniv R., A. Agarwal, and R. Leach (1997) Predicting
the outcome following bankruptcy fling: A
three-state classifcation using neural networks,
IntelligentSyst. in Accounting, Finance, and
Management, vol. 6.
Bell, T., G. Ribar, and J. Verchio (1990) Neural nets vs.
logistic regression: A comparison of each model’s
ability to predict commercial bank failures, in Proc.
(1990 Deloitte Touche/Univ. Kansas Symp. on
AuditingProblems, 29–53.
Belli Giorgio (2004) Valutazione delle imprese e
merito creditizio alla luce del nuovo accordo di
Basilea, thesis written at the University of Rome
“La Sapienza”, Academic Year 2003/2004.
Bonifazzi, C., M. Cinti, et al (2006) "Principal
Component Analysis." Physica Medica 21: 1.
Bowler, S. and T. Donovan (2002) "Democracy,
institutions and attitudes about citizen infuence on
government." British Journal of Political Science.
32(02): 371-390.
Boyacioglu, M., Y. Kara, et al (2009) "Predicting
bank fnancial failures using neural networks,
support vector machines and multivariate statistical
methods: A comparative analysis in the sample of
savings deposit insurance fund (SDIF) transferred
banks in Turkey." Expert Systems with Applications
36(2): 3355-3366.
Burges, C (1998) "A tutorial on support vector
machines for pattern recognition." Data mining and
knowledge discovery 2(2): 121-167.
Cristianini, Nello, John, Shawe-Taylor (2000) An
introduction to support vector machines and
other kernel-based learning methods, Cambridge:
Cambridge University Press.
Deforche, Koen (2004) Bayesian Network Learning
Università di Leuven per workshop The HIV Data
Management and Data Mining, South Africa.
Drucker, H., D. Wu, et al (1999) "Support vector
machines for spam categorization." IEEE
Transactions on Neural networks 10(5): 1048-1054.
Easton, D (2009) "A re-assessment of the concept
of political support." British Journal of Political
Science. 5(04): 435-457.
Fauci F (2004) Studio e sperimentazione di un
nuovo metodo per la classifcazione di immagini
mammografche University of Palermo.
Fernandez-Rodriguez, F.,Gonzalez-Martel, C., &
Sosvilla-Rivero, S (2000) On the proftability of
technical trading rules based on artifcial neural
Franch F. - Artifcial Learning and Support Vector Machines: Default Risk Prediction
Fabio Franch
Business Intelligence Journal - July, 2011 Vol.4 No.2
272 Business Intelligence Journal July
networks: Evidence from the Madrid stock market,
Economic Letters, 69, 89–94.
Fukushima, K. (1975) Cognitron: A Self-Organizing
Multilayered Neural Network, Biological
Cybernetics (20: 121–136.
Gencay, R. (1996) Nonlinear prediction of security
returns with moving average rules, Journal of
Forecasting, 15, 165–174.
Kaski Samuel, Janne Sinkkonen, and Jaakko Peltonen
(2001) Bankruptcy Analysis with Self-Organizing
Maps in Learning Metrics, IEEE Transactions on
neural networks, Vol. 12, n.4, July.
Kuoa R.J., P. Wub, C.P. Wangc (2002) An intelligent
sales forecasting system through integration of
artifcial neural networks and fuzzy neural networks
with fuzzy weight elimination.
Martin-del-Brio, B. & Serrano-Cinca, C. (1995) Self-
organizing neural networks: The fnancial state of
Spanish companies.
Mortara V (2006) INSTAMAP: un programma per
la produzione “istantanea” di mappe fattoriali.
University of Trento.
Odom, M. & Sharda, R. (1990) A neural network
model for bankruptcy prediction, Proceedings
of the IEEE International Conference on Neural
Networks, 163–168.
Ohlson, J. (1980) Financial ratios and the probabilistic
prediction of bankruptcy, Journal of Accounting
Research, 18(1), 109–131.
Rodriguez, Claudia P (2004) Energy Price Forecasting
in the Ontario Competitive Power System Market,
IEEE Transactions on power systems,. Vol. 19, n.1
February.
Rosenblatt, F (1962) Principles of Neurodynamics,
Spartan Books.
Rudorfer, Gottfried (1995) Early Bankruptcy Detection
Using Neural Networks.
Scarselli, Franco, Sweah, Liang Yong, Markus,
Hagenbuchner, Ah Chung Tsoi (2005) Adaptive
Page Ranking with Neural Networks, 936-937.
Tewoldeberhan, Tamrat W., Alexander, Verbraeck,
Simon S. Msanjila (2005) Simulating Process
Orchestrations in Business Networks: A Case Using
BPEL4WS.
Tortorella, Francesco (2002) Approccio statistico alla
classifcazione course slides of ‘Teoria e Tecniche
di Interpretazione delle Immagini’, University of
Cassino.
Viswanathan, Murlikrishna, Young-kyu Yang, e Taeg
Keun Whangbo (2005) Minimum Message Length
and Classical Methods for Model Selection in
Univariate Polynomial Regression ETRI Journal,
27, Number 6, December, 750.
Yongqiao Wang, Shouyang Wang, and K. K. Lai.
(2005) A New Fuzzy Support Vector Machine to
Evaluate Credit Risk IEEE Transactions on Fuzzy
Systems, 13, n.6.
Zhang, G.Peter (2004) Neural Networks in Business
Forecasting, Idea Group Publishing.
2011 273
REVAAM MODEL APPLIED TO MULTIPLE VALUATION
COMPARISON AMONG DIFFERENT WORLD REGIONS
Carlos Acosta-Calzado (MBA)
Business Intelligence Service
IBKAN Consultants, Investment Banking
Email: [email protected]
Abstract
This paper uses the REVAAM Model used to calculate adjusted multiples, through multiple linear regression, used commonly in
valuating companies such as the Price Earnings (PE) and/or the Enterprise Value to Earnings Before Interest, Taxes, Depreciation and
Amortization (EV/EBITDA). For this paper, we used REVAAM to calculate historical adjusted multiples for eight regions in the world:
Africa, Asia, Western Europe, Eastern Europe, Latin-America and the Caribbean, Middle East, Oceania and United States & Canada;
all the World; and the BRIC group (Brazil, Russia, India and China). We show the yearly obtained multiples, as well as the long term
multiples for each region and compared them. During our analysis, we found that although emerging markets do show higher multiples,
these are converging towards mature market multiples. When we compare long term multiples to those for each year we can distinguish
differences that could be consider bubbles due to speculation and to fundamentals (proftability factors). Key words: REVAAM, multiple
valuation, PE multiple, EBITDA multiple, EV/EBITDA
REVAAM
1
model was created as a way to make
commonly used multiple valuation a true "comparable"
valuation tool, rather than just a mathematical averaging
method that could be applied to the company we are
valuing. This model implies considering other fundamental
factors that truly affect valuation of a specifc company
and that are different from other companies even within
the same industry. These parameters refer to proftability
factors and are company specifc, but the methodology
of the model lets incorporate other parameters that could
include macroeconomic or industry specifc factors.
Although there are several multiples used in relative
valuation: Price Earnings
2
(PE), Enterprise Value to Earnings
Before Interest, Taxes, Depreciation and Amortization
3
(EV/EBITDA), Price to Sales
4
(PS), Price to Book Value
of Equity
5
(PBV), and variations to these, REVAAM model
concentrates in P/E and EV/EBITDA multiples as they are
the most used multiples and refer to the proftability of the
company.
For purposes of our work, we constructed a database
with historic information from thousands of companies
around the world and grouped them in 10 proposed regions,
from years ending 2000 to 2010. We wanted frstly to use
REVAAM model to calculate adjusted multiples for these
regions and compare them to see if there were differences
among regions, and secondly obtain and compare short term
multiples (calculated each year) and long term multiples
(without differentiating each year).
Hypotheses
As stated before, multiples vary among markets
according to region specifc risks, economic conditions
and risk-return parameters, as opposed to only fundamental
factors such as proftability at company level. Additionally,
we believe that markets are not rational in the short term
and tend to overreact to region specifc or world events,
thus making fair market value deviate from the theoretical
long-run value. Taking these arguments into consideration
we will try to test the following hypotheses:
Hypothesis 1: Adjusted long term and short term
multiples for emerging or underdeveloped regions
will be higher than those in mature markets.
1
Relative Valuation Adjusted Model
2
Price Earnings ratio is calculated as the market price per share divided by net
earnings per share
3
Enterprise Value to Earnings Before Interest, Taxes, Depreciation and Amortiza-
tion ratio is calculated as the value of operating assets divided by operating
earnings before interests and taxes plus depreciation and amortization
4
Price to Sales ratio is calculated as the market price per share divided by sales
per share
5
Price to Book Value of Equity is calculated as the market price per share divided
by the book value per share
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
274 Business Intelligence Journal July
Hypothesis 2: Adjusted yearly multiples will
signifcantly differ from the adjusted long term
multiples in each market.
Methodology
Model
REVAAM Model adjusts PE multiple by using the
Return on Equity
6
(ROE) and the Net Margin
7
as the
indicators of proftability for equity shareholders, and
adjusts EV/EBITDA multiple by using the Return on
Capital
8
(ROC) and the Pre-tax Operating Margin
9
. These
adjustments are made using multiple regression analysis in
order to minimize errors, where the independent variables
are the multiples and the other factors are the dependent
variables. These relationships are shown in the equations
below:
6
Return on Equity measures the rate of return on shareholders’ equity and is cal-
culated as the Net Income divided by the book value of equity
7
Net Margin measures the proftability and is calculated as the net income divided
by revenues
8
Return on Capital measures the rate of return on the company’s capital and is
calculated as operating earnings after taxes divided by the book value of equity
plus the book value of debt
9
Pre-tax Operating Margin or EBIT measures proftability before interests and
taxes and is calculated as operating revenues less operating expenses plus non-
operating income
(1)
Where,
P/E = the price earnings ratio calculated as the price per
share divided by net earnings per share
?
PE
= the value of the intercept calculated by the
regression analysis
?
ROE
= the value of the coeffcient for ROE
ROE = the return on equity, calculated as the net income
divided by the company’s beginning equity value
?
NM
= the value of the coeffcient for Net Margin
Net Margin = calculated as the net income of year t
divided by the sales of year
? = the sum of errors derived from the regression analysis
(2)
? Margin Operating tax - Pre * ?
ROC * ? ? EV/EBITDA
OM
ROC EVEBITDA
+
+ + =
Where,
EV/EBITDA = the EBITDA multiple calculated
as the Enterprise Value or Value of the Net Operating
Assets divided by the Earnings before Interests, Taxes,
Depreciation and Amortization
?
EVEBITDA
= is the value of the intercept calculated by the
regression analysis
?
ROC
= the value of the coeffcient for ROC
ROC = the return on capital, calculated as the after-tax
operating income (EBIT*(1-tax)) divided by the company’s
beginning equity value plus the debt balance
?
OM
= the value of the coeffcient for Net Margin
Pre-tax Operating Margin = calculated as the pre-tax
operating income of year t divided by the sales of year
? = the sum of errors derived from the regression analysis
? Margin Net * ROE * P/E
NM ROE PE
+ + + = ? ? ?
As stated in the REVAAM Model, the basic linear
regression model assumes that the contributions of the
different independent variables to the prediction of the
dependent variable are additive and they tend to follow
normal distributions. In our case, the relationships between
our variables may be multiplicative and also they have
highly skewed distributions (positive values). Hence it may
be possible to make their distributions more normal-looking
by applying the logarithm transformation, as shown in the
following equations:
From equations (1) and (2), we apply natural logarithms
to obtain:
? Margin) Ln(Net *
Ln(ROE) * Ln(P/E)
NM
ROE PE
+
+ + =
?
? ?
(3)
? Margin) Operating tax - Ln(Pre *
Ln(ROC) * DA) Ln(EV/EBIT
OM
ROC EVEBITDA
+
+ + =
?
? ?
(4)
Equations (3) and (4) will be used to obtain long-term
adjusted multiples for each region without considering
differences among years.
For short-term or yearly adjusted multiples we need to
differentiate samples for each year and, in order to simplify
the number of equations, we decided to consider dummy
variables for each year, so that the value will equal to 1 for
the corresponding year and zero for all other years.
2011 275
Hence, from equations (2) and (3), we introduce dummy
variables as follows:
(5)
(6)
Ln(EV/EBITDA) = ?
EVEBITDA
+ ?
ROC
*
Ln(ROC) + ?
OM
* Ln(Pre-tax Operating
Margin) + ?
2000
*2000 +?
2001
*2001 +
?
2002
*2002 + ?
2003
*2003 + ?
2004
*2004 +
?
2005
*2005 + ?
2006
*2006 + ?
2007
*2007 +
?
2008
*2008 + ?
2009
*2009 + ?
2010
*2010 + ?
Ln(P/E) = ?
PE
+ ?
ROE
*Ln(ROE) + ?
NM
*Ln(Net Margin) + ?
2000
*2000 +?
2001
*2001
+ ?
2002
*2002 + ?
2003
*2003 + ?
2004
*2004 +
?
2005
*2005 + ?
2006
*2006 + ?
2007
*2007 +
?
2008
*2008 + ?
2009
*2009 + ?
2010
*2010 + ?
Where,
20xx equals to 1 if it corresponds to the year and zero
for any other year
Sample
These regressions should be performed at company
level due to the incorporation of proftability factors and
results could be grouped by industry, country, region,
etcetera. REVAAM Model could be applied to any sample
of companies where proftability factors are available.
Since information of private companies is limited, we need
to use public information. For purpose of our analysis,
we obtained information of public companies from 113
countries with one or more stock exchanges, from years
2000 to 2010, where available. The following table
summarizes the number of companies we were able to
obtain for each region.
Table 1. Number of companies by sector per year
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total
Africa 0 0 307 357 380 440 512 622 378 813 857 4,666
Asia 0 0 3,075 6,087 7,353 9,081 9,675 10,621 11,949 20,245 22,083 100,169
Europe_east 0 0 3 613 737 886 1,269 1,719 569 1,213 1,462 8,471
Europe_west 0 0 3,992 3,343 3,695 3,821 4,297 4,384 4,301 4,840 4,871 37,544
Latam&Carib 0 0 628 501 578 532 589 694 776 882 923 6,103
Middle East 0 0 320 399 463 580 721 818 549 1,123 1,154 6,127
Oceania 0 0 0 495 531 654 681 722 881 1,571 1,428 6,963
US&Canada 5,752 7,154 7,440 7,575 6,072 6,162 8,579 8,246 8,059 9,303 7,863 82,205
World 5,752 7,154 15,765 19,370 19,809 22,156 26,323 27,826 27,462 39,990 40,641 252,248
BRIC 0 0 1,353 2,118 2,339 3,079 3,643 4,447 3,434 8,615 10,119 39,147
From the total 252,248 companies, some could be
considered in several years but with data corresponding to
each particular year.
When applying the samples to the model, we found that
the number of companies that can be used was reduced
because:
a. Negative fgures and values equal to zero cannot be
transformed to natural logarithms,
b. we took out "outlier" companies with ROE or ROC
higher or equal to 200% and those with Net margin or
Pre-tax Operating Margin higher or equal to 100%, and
c. some companies had missing information for some
proftability factors.
Thus, the following tables summarize the number of
companies we used for each multiple after the adjustments
on the samples.
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
276 Business Intelligence Journal July
Table 2. Final sample for P/E REVAAM calculation
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total
Africa 0 0 237 269 301 361 390 475 231 277 339 2,880
Asia 0 0 2,516 4,843 6,340 7,819 8,029 8,764 8,154 6,789 8,939 62,193
Europe_east 0 0 0 281 284 343 865 1,355 316 369 462 4,275
Europe_west 0 0 2,707 2,197 2,471 2,580 3,032 3,178 2,853 2,127 2,385 23,530
Latam&Carib 0 0 411 364 442 445 519 574 562 515 604 4,436
Middle East 0 0 212 261 342 437 549 604 309 335 530 3,579
Oceania 0 0 0 330 346 392 429 428 482 496 462 3,365
US&Canada 2,826 3,285 2,902 3,328 3,342 3,554 3,744 3,551 3,613 3,100 3,031 36,276
World 2,826 3,285 8,985 11,873 13,868 15,931 17,557 18,929 16,520 14,008 16,752 140,534
BRIC 0 0 1,086 1,577 1,806 2,368 2,809 3,553 2,069 1,968 2,493 19,729
Table 3. Final sample for EV/EBITDA REVAAM calculation
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total
Africa 0 0 255 268 288 338 385 411 184 226 263 2,618
Asia 0 0 2,561 4,958 6,297 7,772 8,007 8,570 7,708 9,425 11,607 66,905
Europe_east 0 0 0 389 462 577 880 1,242 266 388 460 4,664
Europe_west 0 0 2,648 2,514 2,802 2,866 3,076 3,223 2,713 2,281 2,481 24,604
Latam&Carib 0 0 500 430 456 456 519 558 515 544 586 4,564
Middle East 0 0 224 255 348 425 561 600 227 288 381 3,309
Oceania 0 0 0 338 380 428 401 395 438 458 454 3,292
US&Canada 3,382 3,752 3,615 3,951 3,929 3,969 4,179 3,892 3,914 3,811 3,654 42,048
World 3,382 3,752 9,803 13,103 14,962 16,831 18,008 18,891 15,965 17,421 19,886 152,004
BRIC 0 0 1,132 1,675 1,893 2,498 2,794 3,360 1,877 3,592 4,470 23,291
For both cases, the fnal sample used for the REVAAM
Model is substantially lower than the initial sample, but it
does contain suffcient data points to run linear regression
analysis.
For the case of Eastern Europe, both the original and the
fnal samples had few data points for year 2002; thus, we
decided no to include them for that particular year.
Results
From the samples obtained we ran the linear regression
model, frst using equations (5) and (6) to obtain the
adjusted multiples per year for each region. The following
are the resulting equations for each region. For additional
parameters, please refer to appendices 1 and 2.
Africa
Ln(EV/EBITDA) = 1.71E+12+ -0.3623*
Ln(ROC) + 0.1071* Ln(Pre-tax
Operating Margin) -1.71E+12*2002 -
1.71E+12*2003 -1.71E+12*2004 -
1.71E+12*2005 -1.71E+12*2006 -
1.71E+12*2007 -1.71E+12*2008 -
1.71E+12*2009 -1.71E+12*2010
(7)
2011 277
(8)
Ln(PE) = 7.60E+12+ -0.117* Ln(ROE) +
-0.0752* Ln(Net Margin) -
7.60E+12*2002 -7.60E+12*2003 -
7.60E+12*2004 -7.60E+12*2005 -
7.60E+12*2006 -7.60E+12*2007 -
7.60E+12*2008 -7.60E+12*2009 -
7.60E+12*2010
For the African region, we do not have information for
years 2000 and 2001, and the coeffcients for the intercept
and for each year are very high, and although using scientifc
notation for this fgures makes them appear equal among
them, the resulting numbers in the spreadsheet are different.
The resulting F-tests for both equations show that the linear
regression model can be used for the relationships we
proposed and the T-tests also show that our main variables
(ROC, ROE, Net margin, and Pre-tax Operating Margin) are
signifcantly different from zero. The R2 was 26.23% for
the EV/EBITDA multiple and 4.73% for the P/E multiple,
which means that the former, with the data available, has
an equation that explains better the corresponding multiple.
Asia
Ln(EV/EBITDA) = 8.37E+12 -0.3194*
Ln(ROC) + 0.1527* Ln(Pre-tax Operating
Margin) -8.37E+12*2002 -
8.37E+12*2003 -8.37E+12*2004 -
8.37E+12*2005 -8.37E+12*2006 -
8.37E+12*2007 -8.37E+12*2008 -
8.37E+12*2009 -8.37E+12*2010
(9)
Ln(PE) = -7.57E+11 -0.471* Ln(ROE) -
0.0444* Ln(Net Margin) +
7.57E+11*2002 + 7.57E+11*2003 +
7.57E+11*2004 + 7.57E+11*2005 +
7.57E+11*2006 + 7.57E+11*2007 +
7.57E+11*2008 + 7.57E+11*2009 +
7.57E+11*2010
(10)
For the Asian region, again we do not have information
for years 2000 and 2001. The resulting F-tests for both
equations show that the linear regression model can be
used for the relationships we proposed and the T-tests also
show that all main factors are signifcantly different from
zero. The R2 was 15.24% for the EV/EBITDA multiple
and 14.39% for the P/E multiple, so both show similar
proportion of variability explained by the available data.
We need to remember that the Asian region has the highest
number of companies in the whole sample.
Eastern Europe
(11)
Ln(EV/EBITDA) = 0.9011 -0.2276*
Ln(ROC) -0.0934* Ln(Pre-tax Operating
Margin) + 0.172*2003 + 0.1303*2004 +
0.4316*2005 + 0.3527*2006 +
0.3464*2007 -0.03506*2008 +
0.3574*2009 + 0.5927*2010
Ln(PE) = 1.534 -0.4556*
Ln(ROE) + 0.1702* Ln(Net
Margin) + 0.5978*2003 +
0.2182*2004 + 0.5321*2005 -
0.6714*2006 -1.0663*2007 -
0.2348*2008 + 0.4339*2009 +
0.7888*2010
(12)
For Eastern Europe we do not have information for
years 2000 and 2001, and as stated before we eliminated
year 2002 because data points for that year were few. The
resulting F-tests for both equations show that the linear
regression model can be used for the relationships we
proposed and the T-tests also show that all main factors are
signifcantly different from zero. The R2 was 7.14% for
the EV/EBITDA multiple and 4.98% for the P/E multiple.
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
278 Business Intelligence Journal July
Western Europe
Ln(EV/EBITDA) = 1.8465 -0.3479* Ln(ROC) +
0.1822* Ln(Pre-tax Operating Margin) -
0.0626*2002 + 0.0152*2003 + 0.1348*2004 +
0.2591*2005 + 0.4073*2006 + 0.1581*2007 -
0.4386*2008 + 0*2009 + 0.1693*2010
(13)
(14)
Ln(PE) = 1.4909 -0.3908* Ln(ROE) -0.1129*
Ln(Net Margin) + 0.0212*2002 + 0.1008*2003
+ 0.3085*2004 + 0.4554*2005 + 0.5943*2006
+ 0.2374*2007 -0.5214*2008 + 0*2009 +
0.2509*2010
In the case of Western Europe information starts in
2002 as well. In this case, the coeffcient or year 2009 was
calculated as zero in both equations even though there is
data for that particular year; this means that the model is
using 2009 as the base year. The resulting F-tests for both
equations show that the linear regression model can be
used for the relationships we proposed and the T-tests also
show that all main factors are signifcantly different from
zero, including the intercept value obtained. The R2 was
32% for the EV/EBITDA multiple and 22.28% for the P/E
multiple; a high correlation factor in fnancial regression
models.
Latin-America & the Caribbean
Ln(EV/EBITDA) = 1.5172 -0.3317* Ln(ROC)
+ 0.0895* Ln(Pre-tax Operating Margin) -
0.2196*2002 -0.1945*2003 -0.0642*2004 +
0*2005 + 0.216*2006 + 0.2106*2007 -
0.4391*2008 + 0.249*2009 + 0.3912*2010
(15)
Ln(PE) = 1.4069+ -0.5303* Ln(ROE) +
0.0559* Ln(Net Margin) -0.95*2003 -
0.9394*2004 -0.1834*2005 -0.0044*2006 -
0.5897*2007 -0.2693*2008 + 0.4695*2009 +
0.5748*2010
(16)
Regression equations for Latin-America and the
Caribbean are for years 2002 to 2010, where information
was available. F-tests show that both linear regression
models can be used for the relationships we proposed and
the T- tests also show that all main factors are signifcantly
different from zero, except for Net Margin in equation (16)
that suggests that this factor may not contribute to explaining
P/E multiple in this region. For purposes of this analysis and
in order to keep all models consistent among them, we keep
the variable in the model. Additionally, the use of dummy
variables for differentiating years could be misleading when
removing a critical factor such as Net Margin. The correct
procedure should be to run independent linear regression
models for each year and determine when is the factor not
important. R2 was 17.41% for the EV/EBITDA multiple
and 5.48% for the P/E multiple. This difference may be the
result of using Net Margin in the P/E model.
Middle East
Ln(EV/EBITDA) = 1.9015 -0.4348* Ln(ROC) +
0.1815* Ln(Pre-tax Operating Margin) -
0.0406*2003 + 0.2154*2004 + 0.3912*2005 +
0.0338*2006 + 0.1307*2007 -0.5468*2008 -
0.2345*2009 -0.1019*2010
(17)
Ln(PE) = -1.91E+11 -0.4064* Ln(ROE) -0.1208*
Ln(Net Margin) + 1.91E+11*2002 + 1.91E+11*2003 +
1.91E+11*2004 + 1.91E+11*2005 + 1.91E+11*2006 +
1.91E+11*2007 + 1.91E+11*2008 + 1.91E+11*2009 +
1.91E+11*2010
(18)
The Middle East region also has data available for years
2002 to 2010. F-tests also show that the linear regression
model is a good model for the proposed relationships. T-tests
also show that all main factors are signifcantly different
from zero, except for the intercept with a high probability
(95%) of being equal to zero. As in the equations for Latin-
America and the Caribbean we will keep this coeffcient.
The resulting R2 was 29.96% for the EV/EBITDA multiple
and 5.38% for the P/E multiple, which also shows that
maybe EV/EBITDA multiple for this region is better than
P/E when using the proposed relationships.
2011 279
Oceania
Ln(EV/EBITDA) = 1.9298 -0.3244* Ln(ROC)
+ 0.0872* Ln(Pre-tax Operating Margin) -
0.0629*2003 -0.0168*2005 + 0.1463*2006 -
0.007*2007 -0.7868*2008 + 0.0083*2009 -
0.1072*2010
(19)
Ln(PE) = 1.9966 -0.4642* Ln(ROE) -0.0399*
Ln(Net Margin) -0.0576*2003 -0.0141*2005 +
0.1294*2006 -0.1522*2007 -1.1194*2008 -
0.0343*2009 -0.2005*2010
(20)
Oceania region has data available for years 2003 to 2010.
As in the case of Western Europe, both resulting equation
calculate as zero the coeffcient for the base year, which
in this case is 2004. Both F-tests and T-tests show that the
linear regression model is a good model for the proposed
relationships and that all main factors are signifcantly
different from zero, including the intercept. The resulting
R2 was 30.17% for the EV/EBITDA multiple and 33.41%
for the P/E multiple, both high values that show similar
proportion of variability explained by the available data.
US&Canada
Ln(EV/EBITDA) = 1.6474 -0.2439* Ln(ROC) +
0.0492* Ln(Pre-tax Operating Margin) -
0.2262*2001 -0.3049*2002 + 0.0087*2003 +
0.1285*2004 + 0.1135*2005 + 0.1723*2006 +
0.0908*2007 -0.5673*2008 -0.1257*2009 -
0.0028*2010
(21)
Ln(PE) = 2.0608 -0.4683* Ln(ROE) + 0.1005*
Ln(Net Margin) -0.0359*2000 +0.0202*2001
+ 0.3556*2003 + 0.4529*2004 + 0.3781*2005
+ 0.4028*2006 + 0.2765*2007 -0.524*2008
+0.0466*2009+0.2228*2010
(22)
US and Canada is the second region with more data
available and it covers all years from 2000 to 2010. The
base year in both equations is 2002, where the coeffcient is
calculated at zero in both equations. All F-tests and T-tests
show that the linear regression model is a good model for
the proposed relationships and that all main factors are
signifcantly different from zero, including the intercept.
The resulting R2 was 12.81% for the EV/EBITDA multiple
and 23.64% for the P/E multiple, being the latter higher.
World
Ln(EV/EBITDA) = 1.6362 -0.3283* Ln(ROC) + 0.1219*
Ln(Pre-tax Operating Margin) -0.2074*2001 -0.0215*2002
+ 0.0957*2003 + 0.1519*2004 + 0.2482*2005 + 0.303*2006
+0.275*2007 -0.4338*2008+0.1507*2009+0.3397*2010
(23)
Ln(PE) = 1.6897 -0.4203* Ln(ROE) -0.0388*
Ln(Net Margin) + 0*2000 +0.0874*2001 +
0.0142*2002 + 0.0501*2003 + 0.1674*2004 +
0.2707*2005 + 0.2473*2006 + 0.0878*2007 -
0.4883*2008 + 0.1215*2009 + 0.2928*2010
(24)
World includes all available information from all
the previous regions, so data for years 2000 and 2001
correspond to the US and Canada region only. In both
equations the base year is 2000. F-tests and T-tests show
that the linear regression model fts the relationships for
both multiples, including intercepts. The overall R2 was
17.31% for the EV/EBITDA multiple and 10.87% for the
P/E multiple.
BRIC
Ln(EV/EBITDA) = 1.8929 -0.2861* Ln(ROC)
-0.0305* Ln(Pre-tax Operating Margin) -
0.0319*2003 -0.2788*2004 -0.207*2005 -
0.0374*2006 + 0.2486*2007 -0.4303*2008 +
0.2656*2009+0.4456*2010
(25)
Ln(PE) = -5.85E+11+ -0.5834* Ln(ROE) + 0.033*
Ln(Net Margin) +5.85E+11*2002 + 5.85E+11*2003 +
5.85E+11*2004 + 5.85E+11*2005 + 5.85E+11*2006 +
5.85E+11*2007 + 5.85E+11*2008 + 5.85E+11*2009 +
5.85E+11*2010
(26)
Finally, the BRIC region was constructed from a sample
including countries from several regions, so data was
available for years 2002 to 2010. Except for the intercept
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
280 Business Intelligence Journal July
Table 4. REVAAM EV/EBITDA multiples (short term)
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 7.33 7.84 8.33 10.42 13.96 13.29 4.93 6.93 7.92
Asia NA NA 12.98 11.14 10.45 11.39 11.86 12.72 5.73 10.45 12.07
Europe_east NA NA NA 6.05 6.18 7.77 7.44 7.28 5.03 7.44 9.24
Europe_west NA NA 8.35 10.06 11.25 12.38 14.16 10.87 5.49 8.04 9.31
Latam&Carib NA NA 7.10 7.03 7.72 8.16 9.94 9.99 5.05 9.31 10.17
Middle East NA NA 16.59 14.53 18.14 20.32 14.17 16.21 7.52 9.14 9.89
Oceania NA NA NA 11.55 11.86 11.41 13.62 11.65 4.98 9.94 9.25
US&Canada 7.57 5.91 5.53 7.75 8.68 8.41 8.87 8.17 4.25 6.53 7.53
World 7.60 5.90 7.98 9.39 9.82 10.55 11.16 10.82 5.20 8.89 10.42
BRIC NA NA 15.81 14.49 11.17 11.53 13.73 17.69 8.84 17.18 18.89
Table 5. REVAAM P/E multiples (short term)
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 7.33 7.70 6.85 13.06 16.15 18.83 7.16 11.15 15.03
Asia NA NA 16.85 16.68 18.22 18.89 19.51 19.55 10.08 19.70 20.99
Europe_east NA NA NA 16.44 9.97 13.31 4.00 2.65 6.02 13.87 17.46
Europe_west NA NA 15.14 15.18 18.51 20.76 23.81 15.79 7.76 14.60 18.13
Latam&Carib NA NA 11.25 4.45 4.23 8.40 10.18 5.56 7.81 15.74 16.04
Middle East NA NA 18.47 3.49 12.42 13.33 8.97 10.70 6.97 12.51 13.90
Oceania NA NA NA 19.14 19.41 18.72 21.54 15.80 6.42 19.34 17.65
US&Canada 14.31 16.10 16.56 23.51 25.84 23.50 23.75 20.82 9.42 16.32 20.97
World 13.86 15.98 15.62 16.45 18.02 19.13 18.62 15.55 9.09 17.23 19.77
BRIC NA NA 27.09 27.88 18.68 18.06 15.31 16.39 14.26 34.96 34.74
The following graphs show the multiples from tables 4
and 5 for each region and for each year. From Graph 1 it
can be seen that:
1. All regions reduced their multiples in 2008, in the
middle of the crisis
2. US and Canada region, show a decline in EV/EBITDA
multiples for 2001 and 2002 and then a recovery until
the fall of 2008.
3. From 2004 to 2007, the US and Canada region had
adjusted P/E multiples above other regions, even higher
than the BRIC region.
of equation (26) all other factors are statistically signifcant
different from zero. We will keep this coeffcient in our
formulas for purposes of our analysis. Also, coeffcients
for years and for intercept in equation (26) are very high.
R2 was 13.57% for the EV/EBITDA multiple and 9% for
the P/E multiple.
Now, for each year we apply the equations (7) to (25)
-those with odd number- to obtain the corresponding
multiples for each year. The factors (ROC, ROE, Pre-
tax Operating Margin and Net Margin) used in each case
correspond to the median values for each region in each
year. Table 4 shows the results for adjusted EV/EBITDA
multiples and Table 5 for adjusted P/E multiples.
2011 281
4. Middle East and BRIC regions have presented the
highest adjusted EV/EBITDA multiples. For the P/E
multiples, BRIC was the highest for years 2002-2003,
2008-2010.
5. The BRIC region had the highest EV/EBITDA and
P/E adjusted multiples during the crisis of 2008 and
recovered in one year and reached historical values in
2010.
6. Except for the BRIC region, variability of EV/EBITDA
multiple in 2002 has been reduced from 2002 to 2010.
Ranges in 2002, excluding BRIC, were between 5.5
and 16.5, whereas in 2010 multiples were between 7.5
and 12; that is a reduction in size of the range of 59%.
7. In the same token, variability of the P/E multiple was
also reduced in the same period in size of range of
64.5%, excluding the BRIC region.
0
5
10
15
20
25
30
35
40
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa Asia Europe_east Europe_west Latam&Carib
Middle East Oceania US-Canada World BRIC
Graph 2. Adjusted REVAAM P/E multiples per region per year
After we have obtained our adjusted multiples yearly
for each region, we will use equations (8) to (26) -those
with even number- to obtain the long term multiples of each
region. Then we calculate the multiples using the median
values of the factors (ROC, ROE, Pre-tax Operating Margin
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 10.24 9.22 9.22 9.42 9.40 9.46 9.19 8.42 8.29
Asia NA NA 12.24 11.67 11.29 10.93 11.00 10.91 10.73 10.06 9.56
Europe_east NA NA NA 7.04 7.51 6.98 7.23 7.12 7.23 7.23 7.10
Europe_west NA NA 9.61 10.77 10.68 10.37 10.22 10.06 9.18 8.64 8.44
Latam&Carib NA NA 8.99 8.72 8.46 8.39 8.27 8.34 8.13 7.65 7.30
Middle East NA NA 17.27 15.72 15.20 14.29 14.24 14.78 13.38 11.72 11.06
Oceania NA NA NA 11.05 10.63 10.39 10.53 10.49 9.77 8.80 9.20
US&Canada 7.12 6.95 7.04 7.23 7.18 7.07 7.03 7.02 7.07 6.97 7.11
World 8.67 8.28 9.29 9.73 9.61 9.38 9.40 9.37 9.14 8.71 8.45
BRIC NA NA 16.43 15.79 15.67 15.21 15.25 14.80 14.69 14.46 13.49
and Net Margin) in each year. Although the formulas for
the long term multiples do not differentiate each year, we
apply the value of the factors to obtain the multiple of that
year using the long term formulae. Tables 6 and 7 show
these results.
Table 6. REVAAM EV/EBITDA multiples (long term)
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
282 Business Intelligence Journal July
Region 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa NA NA 11.64 11.32 11.33 11.33 11.49 11.59 11.71 11.78 12.12
Asia NA NA 17.47 19.58 18.47 17.25 17.20 16.87 17.84 18.57 16.61
Europe_east NA NA NA 6.91 6.09 5.93 5.94 5.83 5.77 6.86 6.02
Europe_west NA NA 17.47 16.19 16.04 15.54 15.52 14.71 15.44 17.21 16.64
Latam&Carib NA NA 9.50 9.69 9.16 8.61 8.71 8.55 8.70 8.41 7.79
Middle East NA NA 12.05 11.16 10.71 9.98 9.38 10.13 9.62 11.44 11.42
Oceania NA NA NA 16.60 15.91 15.57 15.52 15.09 16.10 16.39 17.62
US&Canada 17.22 18.31 19.23 19.13 19.07 18.69 18.43 18.32 18.46 18.08 19.48
World 15.14 15.98 16.80 17.07 16.62 15.92 15.87 15.54 16.16 16.65 16.10
BRIC NA NA 25.50 25.48 23.81 20.35 20.51 19.33 20.35 21.02 18.15
Table 7. REVAAM P/E multiples (long term)
1. All regions long term multiples are smoothed and with
little variability, even though each year we calculated the
corresponding multiple using that year's median values
for ROE, ROC, Net Margin and Pre-tax Operating
Margin. This means that the long term equations are
not taking into consideration other external factors
rather than the fundamental and intrinsic aspects of
each company.
2. Average long term multiples for each region are:
Region EV/EBITDA P/E
Africa 9.21 11.59
Asia 10.93 17.76
Europe_east 7.18 6.17
Europe_west 9.78 16.08
Latam&Carib 8.25 8.79
Middle East 14.18 10.65
Oceania 10.11 16.10
US&Canada 7.07 18.58
World 9.09 16.17
BRIC 15.09 21.61
3. Mature markets, such as Western Europe, Oceania,
United States and Canada, and the World show a clear
gap between long term multiples EV/EBITDA and
P/E, being the former lower than the latter. In contrast,
emerging markets do show the same relationship, but
not as clear, and for the case of Middle East and Eastern
Europe, P/E multiple is lower than EV/EBITDA
Multiple.
4. Average long term multiples for mature markets are 16
to 18 in the case of P/E and 7 to 10 for EV/EBITDA,
whereas for emerging markets P/E ranges from 6 to 21
and from 7 to 15 in the case of EV/EBITDA.
5. All regions showed that for the crisis of 2007-2008
their short term or observed multiples were below the
long term multiples for that period, but all were above
the long term multiple in 2010. However, mature
markets' short term multiples for 2010 were slightly
above the long term multiples for that year; contrasting
with emerging markets whose observed multiples were
relatively higher than the long term multiples for 2010.
This may mean that investors cash was sent to emerging
markets seeking for better risk adjusted returns, but
also means that these regions are overvalued according
to long term multiples.
Table 8. Long term REVAAM average multiples
2011 283
Graph 3. Long and short term multiple comparison for African
region
Graph 4. Long and short term multiple comparison for Asian
region
Graph 5. Long and short term multiple comparison for Eastern
Europe region
-
5
10
15
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Asia
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
-
5
10
15
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Eastern Europe
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 6. Long and short term multiple comparison for Western
Europe region
-
5
10
15
20
25
30
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Western Europe
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 7. Long and short term multiple comparison for Latin-
America and the Caribbean region
-
5
10
15
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Latin-America and the Caribbean
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 8. Long and short term multiple comparison for Middle
East region
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Middle East
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
284 Business Intelligence Journal July
Graph 9. Long and short term multiple comparison for Oceania
region
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Oceania
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 10. Long and short term multiple comparison for Us and
Canada region
-
5
10
15
20
25
30
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
US and Canada
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 11. Long and short term multiple comparison for World
-
5
10
15
20
25
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
World
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Graph 12. Long and short term multiple comparison for BRIC
region
-
10
20
30
40
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
BRIC
EV/EBITDA ST
EV/EBITDA LT
P/E ST
P/E LT
Hypothesis 1 established that adjusted long term and
short term multiples for emerging or underdeveloped
regions should be higher than those in mature markets.
When we compared the adjusted short term EV/EBITDA
multiples we obtained a high variability year by year and
there is no suffcient evidence that emerging regions have
higher multiples than mature regions, except for the BRIC
and Middle East region; being the latter converging to
mature markets multiples after 2008. Then, looking at the
long term adjusted multiples, we found that for the EV/
EBITDA multiple, only Asia, Middle East and the BRIC
region were above the mature regions (United States and
Canada, Oceania and Western Europe). Regarding the
adjusted P/E multiple, fndings are not the same. For the
short term P/E adjusted multiples, only the BRIC region was
above mature markets multiples for all years except from
2004 to 2007. In 2010, BRIC region multiples sky rocketed
and also Asian multiples surpassed those of mature markets.
All other emerging regions' P/E multiple remained below
those of mature regions, but began increasing after 2008.
Now, comparing long term adjusted P/E multiples, the
only region above mature markets was the BRIC, which in
2010 converged to mature markets multiples. Additionally,
Asia's P/E adjusted multiple is within mature market range,
while the rest of the emerging regions are below mature
markets for all the period range.
Therefore, we could not fnd enough evidence to support
Hypothesis 1, thus we reject it. However, in individual
cases, we could on accept it for the BRIC region which
in all cases appears to have adjusted EV/EBITDA and P/E
multiples above those observed in mature markets.
2011 285
Hypothesis 2 suggests that yearly multiples - those
calculated from equations with odd number (7) to (25) - will
signifcantly differ from the adjusted long term multiples in
each market. By looking at graphs 3 to 12, we can see that
adjusted short term multiples do vary more than adjusted
long term multiples. The following table shows the standard
deviation for each multiple per region.
Table 9. Standard deviations for adjusted short and long term
multiples
Region
E
V
/
E
B
I
T
D
A
S
h
o
r
t
T
e
r
m
E
V
/
E
B
I
T
D
A
L
o
n
g
T
e
r
m
P
/
E
S
h
o
r
t
T
e
r
m
P
/
E
L
o
n
g
T
e
r
m
Africa 2.99 0.58 4.51 0.26
Asia 2.16 0.80 3.22 0.95
Europe_east 1.28 0.16 5.70 0.45
Europe_west 2.57 0.86 4.50 0.90
Latam&Carib 1.73 0.51 4.44 0.58
Middle East 4.35 1.93 4.34 0.92
Oceania 2.60 0.78 4.67 0.78
US&Canada 1.47 0.09 5.02 0.64
World 2.00 0.49 2.97 0.57
BRIC 3.37 0.86 8.24 2.67
So statistically speaking, deviations for adjusted short
term multiples are remarkably higher than those for adjusted
long term multiples, so we would accept Hypothesis 2 for
all regions. However, we need to notice that variability for
EV/EBITDA multiples is considerably lower than that for
P/E multiples, both in the short term. Another important
fnding is that, theoretically speaking, EV/EBITDA
multiples are lower than P/E multiples, because EBITDA
is higher than NI. In our adjusted multiples, we fnd that all
mature, BRIC and Asia regions present a clear gap between
both multiples; nevertheless, for Latin-America and the
Caribbean, and Africa the gap is small, and for Middle East
and Eastern Europe the relationship is inverse, adjusted P/E
multiples are higher than EV/EBITDA multiples.
It is important to mention that these analysis could have
been performed by using simple average for the multiples
used, however, using REVAAM model provides adjusted
multiples that should be more accurate than using simple
average multiples. Besides, what determines the fair market
value of a company is not just a multiple, but it should take
into consideration proftability parameters that in the end
determine the free cash fow for each specifc company,
and those companies with higher proftability should be
awarded a higher valuation, thus, higher multiples. In
appendix 5 we can see that using average multiples will
lead to higher values, even more for the P/E multiple.
Discussion and Conclusions
We used REVAAM model, with its limitations, to
calculate P/E and EV/EBITDA adjusted multiples for
regions of the world (some belong to emerging markets,
other to mature markets and few with countries in both
markets) and compare them both in the short run (year by
year) and in the long run. The use of REVAAM, through
multiple linear regression, let us differentiate proftability
factors at level company which in the end are what
determine the cash fow from each company, thus its value,
rather than applying just an arithmetic average of multiples.
General valuation literature suggests that multiples in
emerging markets are higher than mature markets, mainly
because risk-return analysis should award a better return in
economies that are not yet fully developed. However, as
we tested this hypothesis, we found that empirical evidence
does not support this proposal, except for the BRIC region.
One may argue that in order to get a higher return, investors
buy assets whose prices (multiples) are low and then try to
sell them at higher prices. Nevertheless, results showed
that emerging markets multiples were not higher than those
in mature markets, not in the short nor in the long term;
actually, in the long term multiples for emerging regions
were below those in mature regions.
We need to consider also that our time frame included
two crises, the one in 2001-2002 and the one in 20007-2008.
If we analyze the short term multiples, we could see that
in both periods multiples plunged but made a recovery in
the following years; two years for recovery after the 2001-
2002 crisis and just one year after that of 2007-2008, so
investors are already assuming that valuations in crises do
fall, but eventually need to recover along with government
intervention (i.e. US government bailout of banks and
European Central Bank funding to Greece fnancial
system). Additionally, variation in multiples was still high
among regions after the 2001-2002 crisis, whether after the
2007-2008 crisis this variability decreased signifcantly,
suggesting that valuation ranges for most regions are based
more on fundamentals or similar to mature markets. This
is not true for the BRIC region which during the 2007-
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
286 Business Intelligence Journal July
2008 crisis had the largest fall in multiples but presented
the highest multiples than any other region during recovery,
even higher than past levels. This effect could be caused
by investors preference or current trend to invest in these
countries due to their economic potential.
Besides comparing emerging to mature regions in
terms of valuation multiples, we wanted to compare long
term multiples which are not differentiated by year, but by
proftability factors. The reason behind this is that although
fnancial analysts believe that markets are rational there are
many external factors that affect valuations without really
affecting proftability factors of the companies. It is true
that economical shocks due affect interest rates, infation
and economic growth perspectives, and these variables
are also inputs in valuation; however, valuations should be
performed in the long run and must include these shocks in
the adequate short term frame. Actually, the most important
driver in free cash fow valuations is the terminal value
which takes into consideration long term proftability of the
company matched with long term conditions.
Results showed that adjusted short term multiples had
a considerably higher variability than adjusted long term
multiples. Since formulae for short and long term multiples
are differentiated by specifc years, we can conclude that
deviations from the adjusted long term multiple are more
speculation bubbles than true change in fundamental
value, and this is also supported by the fact that multiples
recovered in two years, and recently in only one year after
the crisis. Also, comparing short term to long term multiples
allows to somehow realize if a company, industry or region
is under or overvalued, so if we believe that in the long
run that company, industry or region should converge to the
long term multiple, we could decide to invest or short sell
to make a proft. In this token, I would say that the BRIC
region is overvalued. Of course, as we said later, other
macroeconomic and external factors do affect companies´
and industries´ valuations, but should not affect an entire
region. It is opinion of the author of this paper, that when
using multiple valuation analysts should consider the
following:
a. Using a method that incorporates factors that
differentiates one company from another, such as
REVAAM, rather than just using a simple average of
multiples, even within an industry. It is possible to add
additional factors such as macroeconomic or industry
specifc factors that affect valuation of our companies,
if information is available.
b. Regarding the last point, multiples should be used at
company level, as industry or region multiples are the
aggregated sum of all corresponding companies. For
our analysis, we ran regression models using raw data
from companies, not aggregated information.
c. It is better to use adjusted long term multiples than short
term multiples, because the former are less infuenced
by speculation and do adjust in time with a trend.
d. Prefer EV/EBITDA multiples over P/E multiples as the
latter present more variability than the former. The only
exception would be fnancial institutions and some
service companies in which EBITDA is not a natural
or easy measure.
e. If possible, use multiple valuation if free cash fow
valuation is not possible, as valuation is the present
value of actual and future cash fows.
It will be interesting to calculate multiples at the end of
2011 for all these regions in order to see how the Middle
East conficts affect valuations in nearby regions and in the
world, as it has affected oil prices and interest rates, but
even more investors perspectives on risk and speculations
about the future.
Finally, I would like to mention that using REVAAM
for adjusted multiples in the Middle East and Eastern
Europe resulted that P/E multiples were lower than EV/
EBITDA multiples, but when looking at average raw data
the opposite is true. This is because distributions for those
regions were skewed and regression analysis takes an
average of the distribution, which is not necessarily at the
50% of the sample.
References
Acosta-Calzado, Carlos et al. 2010. "REVAAM
Model to determine a company's value by multiple
valuation and linear regression anaylisis" Business
Intelligence Journal Volume 3, Number 2.
Schreiner, Andreas. 2007. “Equity Valuation Using
Multiples: An Empirical Investigation” Dissertation
der Universität St. Gallen (HSG)
Damodaran, Aswath. 2002. “Investment Valuation:
Tools and Techniques for Determining the Value of
Any Asset” Second Edition.
McKinsey & Company Inc., et al. 2005. “Valuation:
Measuring and Managing the Value of Companies”,
Edition, University Edition.
2011 287
Damodaran, Aswath. 2010. “The Dark Side of
Valuation: Valuing Young, Distressed, and Complex
Businesses”, 2nd Edition, Pearson Education, Inc.
Medlen, Craig. 2007. “The bubble machine: relative
capital valuation, distributive shares and capital
gains” An article from: Journal of Economic Issues.
Iwanowski, Raymond J. 1995. “An investor's guide
to foating-rate notes: Conventions, mathematics
and relative valuation (United States fxed-income
research. Portfolio strategues)”
Hoover, Scott. 2005. “Stock Valuation: An Essential
Guide to Wall Street's Most Popular Valuation
Models” McGraw-Hill Library of Investment and
Finance.
Frykman, David and Tolleryd, Jakob. 2003. “Corporate
Valuation: an easy guide to measuring value”.
Myers, Brealey. 2002. “Principles of Corporate
Finance”.
Pagano, Robert R. 1998. “Estadística para las Ciencia
del Comportamiento”, 5ª Edición.
Beaver, W. 1967. “Financial ratios as predictors of
failure”, Empirical Research in Accounting. (1966),
“Selected studies”, supplement to the Journal of
Accounting Research, January, 1966.
Greene W. 1993. “Econometric Analysis”, 2th Edition,
Mc. Millan Publishing Co.
Mandala G.S. 1983. “Limited Dependent and
Qualitative Variables in Econometrics”,
Econometric Society, Monographs number 3,
Cambridge, University.
Bloomberg Professional monitoring service.
Yahoo Finance (http://fnance.yahoo.com)
Capital IQ (http://www.capitaliq.com)
Value Line (http://www.valueline.com)
Ibkan Consultants Financial Data (http://www.ibkan.
com.mx)
Appendix 1
REVAAM Regression results for EV/EBITDA multiple for each region per year (short term)
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
1.71E+12 8.37E+11 0.9012 1.8465 1.5172 1.9015 1.9298 1.6474 1.6362 1.8929
?
ROC
(0.3623) (0.3194) (0.2276) (0.3479) (0.3317) (0.4348) (0.3244) (0.2439) (0.3283) (0.2861)
?
OM
0.1071 0.1527 (0.0934) 0.1822 0.0895 0.1815 0.0872 0.0492 0.1219 (0.0305)
?
2000
0.0000 0.0000 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
?
2001
0.0000 0.0000 0 0.0000 0.0000 0.0000 0.0000 (0.2262) (0.2074) 0.0000
?
2002
-1.7E+12 -8.4E+11 0 (0.0626) (0.2196) 0.0000 0.0000 (0.3049) (0.0215) 0.0000
?
2003
-1.7E+12 -8.4E+11 0.1720 0.0152 (0.1945) (0.0406) (0.0629) 0.0087 0.0957 (0.0319)
?
2004
-1.7E+12 -8.4E+11 0.1303 0.1348 (0.0642) 0.2154 0.0000 0.1285 0.1519 (0.2788)
?
2005
-1.7E+12 -8.4E+11 0.4316 0.2591 0.0000 0.3912 (0.0168) 0.1135 0.2482 (0.2070)
?
2006
-1.7E+12 -8.4E+11 0.3527 0.4073 0.2160 0.0338 0.1463 0.1723 0.3030 (0.0374)
?
2007
-1.7E+12 -8.4E+11 0.3464 0.1581 0.2106 0.1307 (0.0070) 0.0908 0.2750 0.2486
?
2008
-1.7E+12 -8.4E+11 (0.0351) (0.4386) (0.4391) (0.5468) (0.7868) (0.5673) (0.4338) (0.4303)
?
2009
-1.7E+12 -8.4E+11 0.3574 0.0000 0.2490 (0.2345) 0.0083 (0.1257) 0.1507 0.2656
?
2010
-1.7E+12 -8.4E+11 0.5927 0.1693 0.3912 (0.1019) (0.1072) (0.0028) 0.3397 0.4456
R
2
0.2623 0.1524 0.0714 0.3139 0.1741 0.2996 0.3017 0.1281 0.1731 0.1357
N 2,618 66,905 4,665 24,604 4,564 3,309 3,292 42,048 152,005 23,291
d.f. 2,606 66,893 4,654 24,593 4,553 3,298 3,282 42,035 151,992 23,280
F value 84.2193 1,093.219 35.79 1,124.991 95.9562 141.1046 157.5764 514.4430 2,650.741 365.0390
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.5558 1.6562 0.6839 100.9945 32.0016 29.3768 44.0296 101.9062 109.1892 54.3584
t
?ROC
(22.5541) (83.1291) (11.5578) (83.2199) (23.3648) (31.0492) (26.6464) (49.5357) (138.558) (36.248)
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
288 Business Intelligence Journal July
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
7.60E+12 -7.6E+11 1.5345 1.4909 1.4069 -1.9E+12 1.9966 2.0608 1.6897 -5.8E+11
?
ROC
(0.1170) (0.4710) (0.4556) (0.3908) (0.5303) (0.4064) (0.4642) (0.4683) (0.4203) (0.5834)
?
OM
(0.0752) (0.0444) 0.1702 (0.1129) 0.0559 (0.1208) (0.0399) 0.1005 (0.0388) 0.0330
?
2000
0 0 0.0000 0 0 0 0 (0.0359) 0 0
?
2001
0 0 0.0000 0 0 0 0 0.0202 0.0874 0
?
2002
-7.6E+12 7.57E+11 0.0000 0.0212 0 1.91E+12 0 0 0.0142 5.8E+11
?
2003
-7.6E+12 7.57E+11 0.5978 0.1008 (0.9500) 1.91E+12 (0.0576) 0.3556 0.0501 5.8E+11
?
2004
-7.6E+12 7.57E+11 0.2182 0.3085 (0.9394) 1.91E+12 0 0.4529 0.1674 5.8E+11
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
t
?NM
6.0593 38.7113 -4.2165 41.8399 6.0297 12.5162 7.0948 11.3174 51.1871 (3.5417)
t
2000
N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
t
2001
N/A N/A N/A N/A N/A N/A N/A (12.6226) (10.6053) N/A
t
2002
(0.5558) (1.6562) N/A (3.2943) (4.3463) N/A N/A (16.8566) (1.3046) N/A
t
2003
(0.5558) (1.6562) 0.1358 0.7854 (3.7166) (0.5643) (1.3230) 0.4941 6.0027 (0.8401)
t
2004
(0.5558) (1.6562) 0.1029 7.1434 (1.2448) 3.1926 N/A 7.2597 9.6603 (7.5224)
t
2005
(0.5558) (1.6562) 0.3409 13.8234 N/A 5.9976 (0.3754) 6.4263 15.9548 (5.8450)
t
2006
(0.5558) (1.6562) 0.2786 22.0669 4.3229 0.5403 3.2156 9.8756 19.5894 (1.0732)
t
2007
(0.5558) (1.6562) 0.2737 8.6446 4.2859 2.1150 (0.1525) 5.1209 17.8437 7.3126
t
2008
(0.5558) (1.6562) (0.0277) (23.2462) (8.7614) (7.3358) (17.6316) (32.0176) (27.7688) (11.5434)
t
2009
(0.5558) (1.6562) 0.2822 N/A 4.9946 (3.3132) 0.1858 (7.0525) 9.7150 7.7845
t
2010
(0.5558) (1.6562) 0.4680 8.8024 7.9484 (1.5135) (2.4138) (0.1566) 22.1371 13.2546
Prob(t
?EVEBITDA
)
0.5784 0.0977 0.4940 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?ROC
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004
Prob(t
2000
)
N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Prob(t
2001
)
N/A N/A N/A N/A N/A N/A N/A 0.0000 0.0000 N/A
Prob(t
2002
)
0.5784 0.0977 N/A 0.0010 0.0000 N/A N/A 0.0000 0.1920 N/A
Prob(t
2003
)
0.5784 0.0977 0.8920 0.4322 0.0002 0.5726 0.1859 0.6212 0.0000 0.4009
Prob(t
2004
)
0.5784 0.0977 0.9180 0.0000 0.2133 0.0014 N/A 0.0000 0.0000 0.0000
Prob(t
2005
)
0.5784 0.0977 0.7332 0.0000 N/A 0.0000 0.7074 0.0000 0.0000 0.0000
Prob(t
2006
)
0.5784 0.0977 0.7805 0.0000 0.0000 0.5890 0.0013 0.0000 0.0000 0.2832
Prob(t
2007
)
0.5784 0.0977 0.7843 0.0000 0.0000 0.0345 0.8788 0.0000 0.0000 0.0000
Prob(t
2008
)
0.5784 0.0977 0.9779 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
2009
)
0.5784 0.0977 0.7778 N/A 0.0000 0.0009 0.8526 0.0000 0.0000 0.0000
Prob(t
2010
)
0.5784 0.0977 0.6398 0.0000 0.0000 0.1302 0.0158 0.8756 0.0000 0.0000
Appendix 2
REVAAM Regression results for PE multiple for each region per year (short term)
2011 289
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
2005
-7.6E+12 7.57E+11 0.5321 0.4554 (0.1834) 1.91E+12 (0.0141) 0.3781 0.2707 5.8E+11
?
2006
-7.6E+12 7.57E+11 (0.6714) 0.5943 (0.0044) 1.91E+12 0.1294 0.4028 0.2473 5.8E+11
?
2007
-7.6E+12 7.57E+11 (1.0663) 0.2374 (0.5897) 1.91E+12 (0.1522) 0.2765 0.0878 5.8E+11
?
2008
-7.6E+12 7.57E+11 (0.2348) (0.5214) (0.2693) 1.91E+12 (1.1194) (0.5240) (0.4883) 5.8E+11
?
2009
-7.6E+12 7.57E+11 0.4339 0 0.4695 1.91E+12 (0.0343) 0.0466 0.1215 5.8E+11
?
2010
-7.6E+12 7.57E+11 0.7888 0.2509 0.5748 1.91E+12 (0.2005) 0.2228 0.2928 5.8E+11
R
2
0.0473 0.1439 0.0498 0.2228 0.0548 0.0538 0.3341 0.2364 0.1087 0.0900
N 2,880 62,193 4276 23,530 4,436 3,579 3,365 36,276 140,535 19,729
d.f. 2,868 62,181 4265 23,519 4,425 3,567 3,355 36,263 140,522 19,717
F value 12.94 949.91 22.3593 674.16 25.63 18.44 187.07 935.48 1,428.50 177.20
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.46 (0.99) 0.4197 49.79 7.95 (0.06) 33.15 95.32 55.14 (0.13)
t
?ROC
(2.90) (65.41) (6.8452) (46.55) (9.60) (8.05) (26.50) (73.84) (83.94) (29.81)
t
?NM
(1.98) (7.21) 2.8127 (16.02) 1.14 (2.99) (2.56) 17.58 (8.97) 1.77
t
2000
N/A N/A N/A N/A N/A N/A N/A -1.44 N/A N/A
t
2001
N/A N/A N/A N/A N/A N/A N/A 0.84 2.24 N/A
t
2002
(0.4566) 0.9918 N/A 0.6915 N/A 0.0565 N/A N/A 0.4330 0.1287
t
2003
(0.4566) 0.9918 0.1634 3.1325 (4.6847) 0.0565 N/A 14.8999 1.5742 0.1287
t
2004
(0.4566) 0.9918 0.0596 9.8559 (4.8645) 0.0565 0.0000 18.9897 5.3303 0.1287
t
2005
(0.4566) 0.9918 0.1455 14.6807 (0.9499) 0.0565 (0.2221) 16.0700 8.7124 0.1287
t
2006
(0.4566) 0.9918 (0.1837) 19.8478 (0.0238) 0.0565 2.0834 17.3060 8.0144 0.1287
t
2007
(0.4566) 0.9918 (0.2918) 7.9858 (3.2324) 0.0565 (2.4482) 11.7364 2.8571 0.1287
t
2008
(0.4566) 0.9918 (0.0642) (17.1806) (1.4704) 0.0565 (18.4664) (22.3193) (15.7557) 0.1287
t
2009
(0.4566) 0.9918 0.1187 N/A 2.5145 0.0565 (0.5692) 1.9188 3.8709 0.1287
t
2010
(0.4566) 0.9918 0.2157 7.9523 3.1689 0.0565 (3.2785) 9.1156 9.4601 0.1287
Prob(t
?EVEBITDA
)
0.6480 0.3213 0.6747 0.0000 0.0000 0.9549 0.0000 0.0000 0.0000 0.8976
Prob(t
?ROC
)
0.0038 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0483 0.0000 0.0049 0.0000 0.2537 0.0028 0.0107 0.0000 0.0000 0.0773
Prob(t
2000
)
N/A N/A N/A N/A N/A N/A N/A 0.1487 N/A N/A
Prob(t
2001
)
N/A N/A N/A N/A N/A N/A N/A 0.3989 0.0251 N/A
Prob(t
2002
)
0.6480 0.3213 N/A 0.4892 N/A 0.9549 N/A N/A 0.6650 0.8976
Prob(t
2003
)
0.6480 0.3213 0.8702 0.0017 0.0000 0.9549 N/A 0.0000 0.1154 0.8976
Prob(t
2004
)
0.6480 0.3213 0.9524 0.0000 0.0000 0.9549 1.0000 0.0000 0.0000 0.8976
Prob(t
2005
)
0.6480 0.3213 0.8844 0.0000 0.3422 0.9549 0.8242 0.0000 0.0000 0.8976
Prob(t
2006
)
0.6480 0.3213 0.8543 0.0000 0.9810 0.9549 0.0373 0.0000 0.0000 0.8976
Prob(t
2007
)
0.6480 0.3213 0.7704 0.0000 0.0012 0.9549 0.0144 0.0000 0.0043 0.8976
Prob(t
2008
)
0.6480 0.3213 0.9488 0.0000 0.1415 0.9549 0.0000 0.0000 0.0000 0.8976
Prob(t
2009
)
0.6480 0.3213 0.9056 N/A 0.0120 0.9549 0.5693 0.0550 0.0001 0.8976
Prob(t
2010
)
0.6480 0.3213 0.8292 0.0000 0.0015 0.9549 0.0011 0.0000 0.0000 0.8976
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
290 Business Intelligence Journal July
Appendix 3
REVAAM Regression results for EV/EBITDA multiple for each region for all years (long term)
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
1.71E+12 8.37E+11 0.9012 1.8465 1.5172 1.9015 1.9298 1.6474 1.6362 1.8929
?
ROC
(0.3623) (0.3194) (0.2276) (0.3479) (0.3317) (0.4348) (0.3244) (0.2439) (0.3283) (0.2861)
?
OM
0.1071 0.1527 (0.0934) 0.1822 0.0895 0.1815 0.0872 0.0492 0.1219 (0.0305)
R
2
0.2623 0.1524 0.0714 0.3139 0.1741 0.2996 0.3017 0.1281 0.1731 0.1357
N 2,618 66,905 4,665 24,604 4,564 3,309 3,292 42,048 152,005 23,291
d.f. 2,606 66,893 4,654 24,593 4,553 3,298 3,282 42,035 151,992 23,280
F value 84.2193 1,093.219 35.79 1,124.991 95.9562 141.1046 157.5764 514.4430 2,650.741 365.0390
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.5558 1.6562 0.6839 100.9945 32.0016 29.3768 44.0296 101.9062 109.1892 54.3584
t
?ROC
(22.5541) (83.1291) (11.5578) (83.2199) (23.3648) (31.0492) (26.6464) (49.5357) (138.558) (36.248)
t
?NM
6.0593 38.7113 -4.2165 41.8399 6.0297 12.5162 7.0948 11.3174 51.1871 (3.5417)
Prob(t
?EVEBITDA
)
0.5784 0.0977 0.4940 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?ROC
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004
Appendix 4
REVAAM Regression results for P/E multiple for each region for all years (long term)
Parameter Africa Asia
Eastern
Europe
West
Europe
Latam&
Carib
Middle
East
Oceania
US&
Canada
World BRIC
?
EVEBITDA
1.71E+12 8.37E+11 0.9012 1.8465 1.5172 1.9015 1.9298 1.6474 1.6362 1.8929
?
ROC
(0.3623) (0.3194) (0.2276) (0.3479) (0.3317) (0.4348) (0.3244) (0.2439) (0.3283) (0.2861)
?
OM
0.1071 0.1527 (0.0934) 0.1822 0.0895 0.1815 0.0872 0.0492 0.1219 (0.0305)
R
2
0.2623 0.1524 0.0714 0.3139 0.1741 0.2996 0.3017 0.1281 0.1731 0.1357
N 2,618 66,905 4,665 24,604 4,564 3,309 3,292 42,048 152,005 23,291
d.f. 2,606 66,893 4,654 24,593 4,553 3,298 3,282 42,035 151,992 23,280
F value 84.2193 1,093.219 35.79 1,124.991 95.9562 141.1046 157.5764 514.4430 2,650.741 365.0390
Prob F = 0 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
t
?EVEBITDA
0.5558 1.6562 0.6839 100.9945 32.0016 29.3768 44.0296 101.9062 109.1892 54.3584
t
?ROC
(22.5541) (83.1291) (11.5578) (83.2199) (23.3648) (31.0492) (26.6464) (49.5357) (138.558) (36.248)
t
?NM
6.0593 38.7113 -4.2165 41.8399 6.0297 12.5162 7.0948 11.3174 51.1871 (3.5417)
Prob(t
?EVEBITDA
)
0.5784 0.0977 0.4940 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?ROC
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Prob(t
?NM
)
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004
2011 291
Appendix 5
Comparison graphs among REVAAM multiples and Average multiples
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Africa
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
15
30
45
60
75
90
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Asia
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Eastern Europe
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Western Europe
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
10
20
30
40
50
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Latin-America & Caribbean
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
10
20
30
40
50
60
70
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Middle East
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
Acosta C. C. - Revaam Model Applied to Multiple Valuation Comparison Among Different World Regions
Carlos Acosta-Calzado
Business Intelligence Journal - July, 2011 Vol.4 No.2
292 Business Intelligence Journal July
-
10
20
30
40
50
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Oceania
EV/EBITDA ST
EV/EBITDA
Average
P/E ST
PE Average
-
10
20
30
40
50
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
US and Canada
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
10
20
30
40
50
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
World
EV/EBITDA ST EV/EBITDA Average
P/E ST PE Average
-
20
40
60
80
100
120
140
160
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
BRIC
EV/EBITDA ST
EV/EBITDA Average
P/E ST
PE Average
2011 293
MARKETING PRINCIPLES OF A NEGLECTED BANKING
SERVICE: SAFE DEPOSIT BOX RENTAL
Monireh Panahi
PhD candidate in Strategic Management in Shahid Beheshti University. Marketing consultant; author of
articles on marketing and strategic management in Iranian journals.
School of Management & Accounting, Shahid Beheshti University, Daneshjoo Blvd., Evin, Tehran.
Email: [email protected]
Mohammad Pakeniat
Researcher in SRRC, Sharif University of Technology. MBA from School of Management & Economics,
Sharif University of Technology. Business consultant and researcher in the fields of strategic
management in developing countries and service marketing.Shahid Rezaii Research Center,
Sharif University of Technology, Azadi Avenue, Tehran, Iran
Email: [email protected]
Abstract
Purpose of this paper is identifying factors which hinder demand for Safe Deposit Box service and factors that are important for
customers when they want to get Safe Deposit Box Service. Methodology consisted of the use of literature review, interviews and focus
groups to form our questionnaires; then two surveys were conducted to extract our results. Findings show (a) customer perception
of lack of safety and privacy are the main causes of little demand for Safe Deposit Box Service; (b) making the customers feel safe,
by giving the information or insuring the boxes can be very useful in marketing Safe Deposit Box service; (c) Bank CRM, personnel
behavior and skills and proximity to home/offce are also important factors for customers who demand SDBR service. There was no
prior formulation of SDBR service marketing practices openly published, so the use of general literature on bank marketing and focus
groups was performed. Focus groups take the ideas of some groups of people for granted; as few people can take apart in them. This
work provides a base for marketing Safe Deposit Box Rental Service. Although the direct income from this service may not seem to
be high, the effects of this service on customer satisfaction have been evaluated to be dramatic. Former research had focused on retail
bank marketing in general; we failed to fnd a research aimed at marketing Safe Deposit Box Services. Our work has directly focused on
marketing this service. Key words: Safe Deposit Box Rental, Marketing, Customer needs, Safety, Banking Services
Financial institutions have prepared a safe place to keep
the valuables of their customers, called “Safe Deposit Box”.
This service is ought to respond to the need for safety. People
use this service to protect their valuables from theft, fre,
etc. Today almost every important bank in the world and
also some non-bank companies are delivering this service.
This means that a huge sum of money has been invested in
building boxes, delivering SDBR services, managing and
planning to develop them.
Most banks have built limited numbers of safe boxes
in their selected, mainly prestigious, branches. Iranian
banks, among others have pursued the same policy. In Iran
state-owned as well as private banks provide this service in
selected branches. These banks have been able to market
few numbers of their boxes (in some cases only about
10%). Heavy investment and little sale have raised the
question about the reasons of low demand in the market and
the factors which are important for customers. Answering
these questions can help banks market their SDBR service
easier.
In bank marketing literature former researchers have
studied marketing related topics in retail banking in general
(e.g., Rosenberg & Czepiel, 1983; Lewis, 1989; Smith and;
Oliver, 1993; Avkiran, 1994; LeBalnc and Nguyen, 1998;
Ndubisi, 2003; wang et al., 2003; Manrai and Manrai,
2007); however, there has been little research about banking
services separately; and we have found no document to
discuss the marketing side of Safe Deposit Box Rental
(SDBR) service. To help fnancial institutions to understand
the behavior of customers of this service better, we have
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
294 Business Intelligence Journal July
conducted a research to answer the following questions:
Why few people use Safe Deposit Boxes? And what is
important for customers when they think about having a
Safe Deposit Box?
To answer these questions, we reviewed pertinent
literature, interviewed with experts and formed an expert
focus group; then we formed a SDBR customer focus group.
We used the outcomes of interviews and focus groups to
form a questionnaire. We conducted a survey to answer
each of the two questions in our fnal step. This research
has been done in different cities in Iran. Although our
respondents, especially in focus groups and surveys, were
limited to Iranians, we expect our fndings to bring about
ideas for other researchers who may try to fnd answers for
similar questions in other social and economic settings.
Our paper is organized as follows: First, we discuss the
history and importance of this service to understand if this
service is still valuable for banks and fnancial institutions.
Second, we discuss our research method, and next we
propose our fndings and conclusion comes at the end.
Importance of the service
In Today’s banking sector SDBR service is delivered
as a necessary banking service. Former research reveals
that in 1970’s about 97% of American banks provide this
service for their customers (Heggestad and Mingo, 1976).
Although most banks spend a lot to offer this service to
their special customers, we failed to fnd a research that
analyses the importance of this service.
To study the importance of this service we reviewed the
patents registered in this feld, and interviewed with experts
from different banks.
a. Patent Analysis
Patent analysis is a method used for several reasons. One
of the applications of this method is identifying the rate of
technological development of a particular product. High
number of patents that have been registered in a period of
time shows that there has been a high investment done in
building this product, which has yielded in many patents.
Low number of patents shows that there has been little
investment done in that product. Higher investment implies
that investors have assumed future market favorable and
expect market growth. Low investment means that frms
are hopeless to have a lucrative business in that product in
future.
In this section, we have reviewed patents and applications
in United States Patent and Trademark Offce (USPTO)
since 1850 that included Safe Deposit Box in their title.
Figure 1 shows the number of patents registered in USPTO
from 1850 till 2010.
Figure 1. Number of patents registered in USPTO about Safe Deposit Boxes
2011 295
As seen in Figure 1, the patents registered show two
upward cones at around 1920’s, a gradual growth since
1950, and a downward cone in 1990. Sharp growth at
1920’s could be attributed to the growth in the number of
customers. It could be argued that the increasing number of
patents related to Safe Deposit Boxes implies that investors
have estimated this product to be proftable.
b. Interviews
To identify the importance of this service for providers,
we carried out interviews with experts from different banks.
Interviewees were selected from globally recognized
fnancial institutions as well as Iranian banks. These banks
included HSBC, Lloyds and Barclays in UK, Standarder
and Banco Pastor from Spain and Melli (largest bank in
the Muslim world), Mellat, Maskan, Saman and ENBank
in Iran.
Results show that most banks do not identify this service
important. All of the interviewees except one admitted that
this service does not bring about direct benefts to banks.
More than half of them expressed that their banks identifes
this service as a means to increase customer satisfaction
rather than a separate service that is separately proftable. So
these banks offer this service to their premium or superior
customers, even sometimes for free. Others, although said
that they treat this service separately, identifed it not very
important to their banks.
Some of the answers that have leaded us to this
conclusion are:
“This service is costly. I think the costs of providing this
service is more than its income; we need to allocate a large
place for boxes, an operator for them, and run a 24 hour
safeguarding, but the price of this service is very low”.
“We offer this service to our selected customers. We
try to increase their satisfaction and we think this would
be proftable for us. Can you imagine the income that
this service will cause by increasing the satisfaction of a
millionaire, even only a little?”
“Today banks’ competition makes you to increase
customer loyalty. We don’t like to lose our customers,
especially premiums, because of lack of SDBR service.”
This evidence questioned the proftability of delivering
this service, but there were other evidence that ensured us
this service is lucrative, even if treated as a separate service,
not just as a means to increase customer satisfaction. This
evidence was non-bank institutions that deliver SDBR
services. Some of these companies sell this service with
other security services (e.g. UK Security Services
1
) and
some provid solely one service: Safe Deposit Box rental
(e.g. metropolitan Safe Deposit Boxes
2
, safe deposit
centers
3
). This evidence showed that this service makes
enough income itself that one runs a business for.
Literature Review
Intense competition has persuaded frms to put more
effective marketing methods into action. Studies performed
based on this want reveal that the cost of serving a loyal
customer is fve to six times less than the cost of fnding
and serving a new customer (Rosenberg & Czepiel, 1983;
Ndubisi, 2003). Therefore further research was carried out
to identify the reasons of customers’ satisfaction, loyalty and
switching behavior. (e.g., Smith and Lewis, 1989; Oliver,
1993; Avkiran, 1994; LeBalnc and Nguyen, 1998; wang et
al., 2003; Manrai and Manrai, 2007). The need for customer
satisfaction urged frms to increase their service quality.
As frms in service sectors decided to increase customer
loyalty and satisfaction, they had to increase their service
quality. Bolton and Drew (1991), Cronin and Taylor (1992)
and Spreng and Mckoy (1996) have argued that service
quality is the main foundation of customer satisfaction for
service companies. Therefore the literature on customer
satisfaction in service sectors has emphasized service
quality as the basis for customer satisfaction, especially
in banking sector. (Smith and Lewis, 1989; Oliver, 1993;
Avkiran, 1994; Levesque and McDougall, 1996; LeBalnc
and Nguyen, 1998; Newman, 2001; Caruana, 2002;
Kheng et al., 2010). Banks today identify service quality
as the main objective that helps them survive in today’s
competitive market (Wang et al., 2003).
Studying customer satisfaction in banking sector added
other factors to service quality. We propose a brief review
of the literature on customer satisfaction in banking sector
here:
There are related topics in banking sector being studied:
customer satisfaction (e.g., Spreng et al., 1995; Levesque
and Mcdougall, 1996; Fienberg, 1996; Mittal and Lassar,
1998; Mittal et al., 1998; Athanassopoulos, 2000; Yavas et
al., 2004), customer loyalty and switching behavior (e.g.,
Jain et al., 1987; Moutinho and Brownlie, 1989; Manrai
and Manrai, 2007). Colgate and Hedge (2001) have argued
1
www.uksecurityservices.com
2
www.metropolitansafedeposits.co.uk
3
www.Safedepositcenters.co.uk
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
296 Business Intelligence Journal July
that there are three main reasons for customer switching:
service failure, pricing and denied services. Chacravarty
et al. (2004) argue that the higher the relationship of
a customer with a bank (economic or other relations)
the lower the propensity of the customer to switch his
bank. Yavas et al. (2004) admit that service quality is the
foundation of customer satisfaction. Manrai and Manrai
(2007) conclude that dissatisfaction with service quality
is the main reason for switching banks. Financial offering
(interest rate) beside convenience and service are the main
parts of service quality identifed (Gwin and Lindgren,
1986; Manrai and Manrai, 1995, Chacravarty, 2004).
Manrai and Manrai (2007) conclude that there are
four overall dimensions that cause customer satisfaction
with bank services: Financial offerings, environment and
atmosphere, convenience and personnel related issues
(Manrai and Manrai, 2007). Although there has been much
debate about their priority (See for example Thwaites and
Vere, 1995).
SDBR service in Iran
Before discussing our research, we propose some
information about SDBR service in Iran, as most of our
feld data has been gathered in Iran.
First openly known Safe Deposit Box in Iran was
delivered in early 1940’s in Melli Bank. This service was
not openly promoted or used till 1980’s, during Iran-Iraq
war; when most of the rich used this place to keep their
valuables safe. The war increased the demand for SDBR
service, and this encouraged the banks to deliver this
service. Today this service is provided in several banks in
Iran; they provide this service at their selected branches.
Most of them offer similar prices, due to Central Bank
regulation. Customers leave some money as a deposit and
pay annual rent to get this service. In Iranian banks access is
limited to working hours but banks follow different criteria
to evaluate suitable customers to deliver SDBR service to
them. Neither of Iranian banks provides insurance for the
contents of safe boxes.
Research Design
In this research we have tried to answer to two questions:
First, Why few people are interested in SDBR service? And
what is important for customers when they think about
having a Safe Deposit Box, in other words, what are key
success factors in marketing Safe Deposit Box services?
To answer the frst question we needed to understand
how people who don’t use this service think about it and
why they don’t use it. To gather their ideas we needed to
conduct a survey. As there was no studies to show what
may be the factors that affect people’s behavior in this
regard, we extracted the factors from:
• A focus group session with people who don’t use this
service.
• A focus group session with bank experts.
In the frst focus group we tried to involve people from
different demographic/occupational groups. Seven people
were invited and the session lasted about 120 minutes.
In the second focus group we invited experts from
Melli (largest bank in the Muslim world), Mellat, Maskan,
Saman and EN banks. These included private as well as
state-owned banks, some of which had a small branch
network, and some a large one. All of these banks deliver
SDBR service. We invited these people as we thought they
are exposed to retail banking customer behavior, and may
have ideas why few banks’ customers demand this service.
The session lasted about 120 minutes.
Results of focus group discussions were analyzed using
content analysis techniques to identify the items of our
questionnaire. This open-ended structured questionnaire
asked about factors that cause few people to use Safe
Deposit Boxes. Randomly selected people who said they
don’t use this service flled out this questionnaire. We tried
to have different demographic groups in our respondents.
784 people answered this questionnaire in 11 selected cities,
to reach the confdence level of 95% at confdence interval
of 3.5. This questionnaire had options based on Lickert
scale, ranking from “frmly agree” to “frmly disagree”.
To answer the second question we needed to know
the idea of SDBR service customers to identify which
factors are important for them when they want to select a
SDBR service provider. To gather their ideas we needed
to conduct a survey. However, as we failed to fnd any
former research to study this phenomenon, we relied on
researches studying retail banking customer satisfaction
and marketing in general. We used the results of these
studies and checked the validity of these factors with bank
experts. As a result we extracted a list of factors that affect
customer satisfaction and successful marketing in retail
banking which were applicable to SDBR service. However,
to complete the factors we added results from:
2011 297
• Interviews with bank experts
• Bank experts focus group
• Customer focus group
Experts from HSBC, Lloyds and Barclays from UK,
Banco Pastor and Standarder from Spain and Melli, Mellat,
Maskan, Saman and EN banks from Iran were interviewed.
These banks deliver SDBR service. We interviewed these
people as we thought they are exposed to customer needs
and expectations from SDBR service providers, and may
have ideas about the factors which affect customer decision
when they want to select a bank to get this service from.
Each interview lasted about 90 minutes.
In bank expert focus group experts from Melli, Mellat,
Maskan, Saman and EN banks were invited. In this session
experts we asked to discuss about factors that are important
for customers and banks can success in marketing SDBR
service by paying attention to them. The session lasted
about 120 minutes.
Seven customers of SDBR service from different banks
were invited to participate in customer focus group. In
this focus group we tried to involve people from different
demographic/occupational groups. The session lasted about
120 minutes.
Results of interviews and focus groups’ discussions
were analyzed using content analysis techniques to identify
the items of our questionnaire. This open-ended structured
questionnaire was aimed at understanding the key factors
of success in marketing SDBR service. 784 current SDBR
service customers flled this questionnaire in 11 selected
cities to reach the confdence level of 95% at confdence
interval of 3.5. This questionnaire had options based on
Lickert scale, ranking the importance from “very low” to
“very high”.
Findings/Discussion
In this section we propose the fndings of our research.
Findings would be proposed in two sections: why people
may not use this service, and which attributes of this service
are important when they look for one.
Why few people use SDBR service?
In this section we present our fndings to answer the frst
question: The reasons of people’s little demand for SDBR
service. To fnd the reasons, as stated earlier we have used
interviews and focus groups with people who don’t use this
service and bank experts. Here we present some quotes
from our interviews and focus groups that we have found
important to derive our results from:
a. Experts’ Focus Group/interview:
“Few people know about the real price of this service;
they think this is a luxury service and should be costly”.
“Even when we introduce this service to our selected
customers, most of them think that this is a special service
for very rich people and they can’t get it.”
“I heard that a customer had told she desired to use this
service, but she thought this service is only for our special
customers”
“People may accuse banks of revealing customer
information to government agencies. People don’t like this,
especially because of tax. Although this does not happen,
but people are concerned about it”
“Compared to other banking services, banks rarely
introduce this service to their customers. This is because
they don’t identify it lucrative”.
b) Non-customers’ focus group
“I don’t use this service; because I may need my
belongings on holidays, when banks are closed”.
“Safe Deposit Boxes are safe, but not my way to the
bank.”
“One of my concerns about this service is my privacy.
When I want to use my box, other customers may know
what I have there. As far as I know, one should use his
SDB in front of others; there should be a private place for
customers”
“Insurance is utmost important for me. No bank provides
insurance on boxes [in Iran].”
“This service is for the rich. I prefer to keep my valuable
belongings in my house.”
“We use this service to increase the safety in keeping
our things, but banks have not assured us they can really
increase safety.”
“No bank in our neighborhood provides this service; I
will have to travel a long distance to get and use this service”
“Safe Deposit Boxes? I have nothing valuable to get
one”
These were quotes we derived from our interview and
results of our focus groups. We then implemented a content
analysis method to analyze these fndings. The factors that
were identifed could be classifed into: little awareness
about the price, little trust to banks, little awareness, lack of
access on holidays, unsafe way to/from bank, little privacy,
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
298 Business Intelligence Journal July
To analyze the results of table 1 we summed “frmly
disagree” and “disagree” percentages, and “frmly agree”
and “agree” percentages. Analysis showed that lack of safety
(85%) and lack of insurance (60%) are the outstanding
reasons. Unsafe way (54%) is ranked third. 47% think that
distance between their place and SDBR provider is a reason
they don’t use this service and 44% have claimed that they
don’t think they have valuables. Other factors don’t seem
to be so important.
The frst three important factors (bank safety, insurance
and safety of people’s way) have all targeted safety;
in other words people’s image of service safety is the
most important factor that affects their decision towards
demanding this service. Our fnding is in harmony with one
of our customers’ claim:
“The main feature of this product is safety. We decide to
use this because we need more safety”.
Geographic proximity is also important for people; as it
takes their time, and affects the safety of their ways to/from
bank. Assuming safety the value that SDBR service brings
about for customers, one would identify the results of the
survey acceptable.
Less than half the respondents have said that other
factors affect their decision. These factors include: people’s
image of privacy of SDBR service, their trust to banks,
their propensity to keep valuables at home, and their need
to access boxes on holidays, and lack of valuables. This
shows that these factors are not as important as formerly
mentioned factors (safety and proximity of providers to
home/offce).
Less than 20% have said that price is important for
them. This fact, beside the fact that most of the people
overestimate the price of this service shows that price does
not hinder people’s demand for SDBR service.
Which attributes of SDBR service are
important for customers?
In this section we discuss the factors that affect that
are important for customers when they want to select a
SDBR service provider. Identifying these factors would
help service providers to understand customer needs and
expectations to develop their services in harmony with
them.
To fnd these factors, we frst reviewed the literature, in
which we failed to fnd a fne-tune literature about SDBR
service. We therefore decided to review the literature on
retail banking as SDBR service is a retail banking service
Factor
F
i
r
m
l
y
d
i
s
a
g
r
e
e
D
i
s
a
g
r
e
e
N
o
i
d
e
a
A
g
r
e
e
F
i
r
m
l
y
a
g
r
e
e
I don’t have access on holidays 9% 20% 36% 25% 10%
My way to bank is not safe 7% 16% 23% 38% 16%
Others may know what I have 21% 34% 26% 15% 4%
I don’t trust banks 27% 32% 18% 15% 8%
There is no insurance 7% 11% 21% 28% 32%
I prefer to keep my valuables at home 14% 36% 18% 24% 8%
Banks are not safe 3% 3% 9% 46% 39%
I have nothing valuable to use SDBR 8% 29% 19% 27% 17%
I cannot aford it 7% 28% 46% 15% 4%
There is no SDBR in our neighborhood 4% 18% 31% 33% 14%
Beside the factors reported in Table 1 we tested
customers’ awareness about this service. We asked them if
they know how to get this service. 15% of respondents
said that they are not aware about this issue at all; and 57%
admitted they don’t know the conditions of getting this
service “exactly”.
In our focus groups we found out that people think
SDBR service is a luxury service and should be expensive.
However, real prices of this service didn’t seem so high.
So we also asked customers’ image about the price of
this service. Survey results recommend that only 27% of
respondents had answered the question about the price
correctly; 19% estimated the price fve times more, and
11% estimated the price ten times more. This implies that
people may imagine this service an expensive service.
Table 1 . Results of our survey about the reasons of little demand
for SDBR service
not insured boxes, propensity to keep belongings in house,
customer imagination of lack of safety and having little
valuables.
These were our most outstanding fndings among many.
To test our fndings from interviews and focus groups
and generalize them to the Iranian society, we designed a
survey based on our former fndings. We ran our survey in
11 selected cities in different parts of Iran. At confdence
level of 95% and confdence interval of 3.5 we collected
784 questionnaires. People that don’t use this service were
asked to fll out the questionnaire. Questions were ranked
in their answers from one (frmly disagree) to fve (frmly
support). Results are shown at table 2.
Asked “Why don’t you use Safe Deposit Box rental
service?” respondents have answered as follows:
2011 299
(provided in the literature review section 3). From the
literature review we derived that: Financial offerings,
environment and atmosphere, convenience and personnel
related issues are important factors in retail banking.
As these factors had not been developed specifcally for
SDBR service, we checked the validity of these four main
groups with bank experts. Based on experts’ ideas we then
concluded that fnancial offering in SDBR service might
not be the interest paid (similar to deposit services) but it
could be the price, which is composed of a deposit and an
annual rent. Environment and atmosphere in case of SDBR
service could be translated to branch safety. We assumed
convenience to be geographic proximity of SDBR service
provider to customers’ place. On personnel related issues,
we derived personnel skill and behavior (including friendly
behavior, caring customers’ needs, etc).
We then formed focus groups with customers and
experts to understand the expectations and important items
for SDBR service customers. As participants in focus
groups are limited, they should be selected in a way to
represent the ideas of different customer groups (Patton,
2001); therefore participants in our focus group were from
different demographic and occupational groups. Here we
present some quotes we have found important to derive our
results from:
“To select a SDBR service, I looked for the one with
insurance, but as I didn’t fnd one [in Iran] I got an uninsured
one, but if it was insured I would feel my belongings are
much safer.”
“I selected this bank, because I felt it is the safest place.”
“It is important to fnd one close to your house or offce.”
“I want to share my box with my father; we keep our
valuables together. Our bank provides us only with one
card. This means that every time the one who wants to use
the box should get the card from the other one. This is not
easy always and takes a lot of time”.
“The least expensive was my option.”
“When I selected this bank, I thought this is the only
bank that provides this service.”
“Every time I go to the bank, I meet a new guy. Bank’s
policy to rotate jobs causes me to feel unsafe.”
“We need the bank to inform us about things like the
need to extend our contract, etc.”
We analyzed our fndings using content analysis
techniques and added our fndings from the literature to it.
We identifed the following possible factors: cost, proximity
to home/offce, insurance, bank branch safety, possibility
of shared access, bank CRM activities, personnel skill,
personnel behavior, fxed SDB staff.
Factor
V
e
r
y
L
o
w
L
o
w
M
o
d
e
r
a
t
e
H
i
g
h
V
e
r
y
H
i
g
h
Cost 4% 7% 25% 26% 38%
Proximity to home/ofce 4% 5% 20% 32% 39%
Insurance 5% 3% 10% 21% 61%
Branch safety 2% 3% 8% 22% 65%
Possibility of shared access 7% 11% 22% 31% 29%
Bank CRM activities 2% 2% 12% 37% 47%
Personnel skills 5% 8% 15% 33% 39%
Personnel behavior 5% 6% 11% 30% 48%
Fixed SDB staf 7% 11% 25% 32% 25%
Table 2. Customer ideas about important factors in SDBR
service delivery
We designed a questionnaire based on the mentioned
factors. We conducted a survey in 11 cities in Iran. At
confdence level of 95% and confdence interval of 3.5 we
collected 784 questionnaires. Cities were selected from
different parts of the country. Respondents were customers
of SDBR service of different banks. We asked them to
evaluate the importance of the identifed factors. The
spectrum composed of “very low” to “very high” options.
It could be observed that bank branch safety and
insurance are the most important factors for customers.
This implies that safety is very important for customers.
Therefore to succeed banks should assure their customers
that their valuables are kept safe.
Also another important factor for customers is banks’
relation with them. As SDBR customers rarely go to bank,
it is important to keep them in touch with their banks to
succeed in increasing customer satisfaction and marketing
this service.
Table 2 shows that all the factors that have been
evaluated are ranked above average. This shows that all of
the factors we have introduced are important. To identify
which factors are more important than others, we used
Friedman test and ranked them. Results are shown below:
Table 3 . Friedman Test for factors identifed to affect SDBR
service marketing
Test Statistics
N 660
Chi-Square 747.819
Df 12
Asymp. Sig. .000
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
300 Business Intelligence Journal July
The results of Friedman test are reported in Table 4.
Table 4. Results of Friedman Test for important factors in
Marketing of SDBR Service
Rank Factor Mean Rank
1 Branch safety 8.79
2 Insurance 8.40
3 Bank CRM activities 7.67
4 Personnel behavior 7.33
5 Proximity to home/ofce 6.87
6 Personnel skills 6.63
7 Cost 6.59
8 Possibility of shared access 5.75
9 Fixed SDB staf 5.47
As the table shows the most important factors are:
Branch safety, insurance and bank CRM activities. This
ensures us that this service should assure customers that
their belongings are kept safe. Other important factors
include personnel behavior and proximity to home/offce.
Factors that ranked lower are: personnel skills, cost,
possibility of shared access and fxed SDB staff.
Conclusion
In this paper we tried to answer two questions that
can help bankers to market their Safe Deposit Box Rental
Service. The frst question was about the reasons few people
use this service. Second question was to identify the factors
that are important for customers to select a service provider.
Results of our research showed that there are several
factors that hinder people’s demand for SDBR service. One
main reason was that people had doubts about the safety of
this service. That the majority doubted the safety of boxes
and identifed insurance as an important factor show that
safety is an outstanding factor. Their concerns about the safe
way to/from bank also strengthen this assumption. Besides
people’s dissatisfaction with the long distance between
their place and bank implies that selecting branches to
locate Safe Boxes is an important issue that helps the bank
to market this service; this would also relieve customer
concerns about unsafe way, due to geographic proximity.
Some people think that the information about contents of
their boxes may be revealed to others (family, government,
etc.), something they resent. Some also think that other
customers may know what they have when they want to use
their boxes and question the private environment in banks.
Our research also showed that many people identifed
SDBR a luxury service. This could be derived from their
imagination about the price of this service (as about notable
proportion estimated the price fve or ten times more), and
that some think this service is dedicated to the very rich
class solely.
Other factors include propensity to keep valuables at
home, lack of access on holidays and lack of valuables to
use SDBR service.
To know which factors are important for customers in
marketing SDBR service (second question), our research
showed that branch safety is ranked frst. This means that
banks should try to transfer a safe image of their boxes. This
could be done by informing them about the construction
of boxes and their location and insuring contents of boxes.
As SDBR customers rarely go to bank (because they
rarely need what they have put in their safe boxes) bank
performance in customer relationship management (CRM)
is important for customers and they have ranked this issue
third. This could be achieved by informing customers about
new pertinent services, termination of their contracts, etc.
by calling, sending mails and emails, etc.
Other important factors which can help banks attract
customers are personnel related issues. This includes
personnel behavior (friendliness, courtesy, etc) and skill.
Banks can strengthen these factors by educating their
personnel in areas like how to do the checking process
faster and following special customer needs whenever
possible. One other factor that was identifed to affect
customer satisfaction in SDBR service delivery was fxed
SDB staff. However, results of our survey showed that this
factor is not very important.
Customers’ request to have this service close to their
home/offce could be answered by distributing the boxes
in a number of branches instead of locating a high number
of them in one branch. Another solution can be providing a
delivery service of boxes to customers’ place. This service
is being delivered in some banks, especially for small boxes.
Cost does not seem to be very important for SDBR
customers; this implies that they are mostly above average
income people. Also the cost of this service (at least in the
Iranian case) is not very high; the rent of the small size
is about 0.3% of the minimum labor income. Therefore
offering lower prices does not seem to be very effective in
marketing this service.
Some customers may need to use Safe Boxes together.
In these cases banks provide one access card for them; and
the one who needs to use the box will need to get the card
from the other; although both have been introduced to the
2011 301
bank as users and their information is registered. This could
be solved by providing one card for each of them. Although
possibility of shared access (with family, colleagues, and
friends) is of very low importance factors identifed in our
research.
It is worth noting that surveys were done in Iran and
the results of our research are based on Iran’s social,
cultural and economic setting. This research has been done
in a developing country; due to similarities in banking
system, social, cultural and economic situations across the
developing world, the results of our work can be generalized
in developing countries. Also, our results can bring about
ideas for further research in marketing SDBR service in
other countries.
Acknowledgement
We are grateful to Mr. Hashemi, Dr. Baseri, Mrs.
Mohseni, and Ms. Rezaeian from Maskan Bank Research
Center for their support of this research. We would also
like to thank Dr. Samadi for his comments on our work.
Mr. Khaksar, Mr. Mohammadzaman and Mr. Mesgari are
acknowledged for their cooperation in analyzing the results
of interviews, focus groups and surveys.
References
Athanassopoulos A. D., 2000, Consumer satisfaction
cue to support market segmentation and explain
switching behavior, Journal of Business Research,
Vol. 47(3), pp.191.
Avkiran N. K., 1994, Developing an instrument to
measure customer service in branch banking,
International Journal of Bank Marketing, Vol.
12(6), pp.10-18.
Bolton, R. N. and J. H. Drew, 1991, A Multistage
Model of Customers’ Assessments of Service
Quality and Value, Journal of Consumer Research,
17 (March), 375-384.
Caruana, A. (2002). Service loyalty: The effects of
service quality and the mediating role of customer
satisfaction, European Journal of Marketing, Vol.
36(7/8), pp. 811-828.
Chacravarty S., R. Feinberg and E. Rhee, 2004,
Relationships and individuals’ bank switching
behavior, Journal of Economic Psychology, Vol.
25(4), pp.507.
Colgate M., and R. Hedge, 2001, An investigation into
the switching process in retail banking services,
International Journal of Bank Marketing, Vol. 19
(4/5), pp.201-212.
Cronin, J. J. and Taylor, S. A. (1992), measuring
service quality: a reexamination and extension,
Journal of Marketing, Vol. 56, July, pp.55-68.
Fienberg R., 1996, Customer dissatisfaction and
delays: the robust negative effects of service delays,
Journal of consumer satisfaction, dissatisfaction
and complaining behavior, Vol. 9, pp.81-85.
Gwin J. M. and Lindgren Jr. J.H., 1986, reaching the
service-sensitive retail consumer, Journal of retail
banking, Vol. 8(3), pp.36-42.
Heggestad A. A. and J. T. Mingo, 1976, Prices, non-
prices and concentration in commercial banking,
Journal of Money, Credit and Banking, Vol. 8 (1),
pp. 107-117.
Jain A.R., Pinson C., Malhotra N. R. , 1987, customer
loyalty as a construct in the marketing of banking
services, International Journal of Bank Marketing,
Vol. 6, pp. 49-72.
Kheng L. L., O. Mahamad, T. Ramayah, 2010, The
Impact of Service Quality on Customer Loyalty: A
Study of Banks in Penang, Malaysia, International
Journal of Marketing Studies, Vol. 2(2), pp. 57-66.
LeBalnc G., and Nguyen N., 1998, Customers’
perception of service quality in fnancial institutions,
International Journal of Bank Marketing, Vol. 6(4),
pp.7-18.
Levesque T. and Mcdougall G.H.G, 1996, determinants
of customer satisfaction in retail banking,
International Journal of Bank Marketing, No. 7,
pp.12-20.
Manrai L.A. and Manrai, A.K.,1995, why do
consumers switch bank services: some managerial
implications (Abstract). In: H. Timmermans (Ed)
European Institute of Retailing and the Canadian
Institute of Retailing and Services Studies.
Manrai L.A. and A. K. Manrai, 2007, A feld study of
customers’ switching behavior for bank services,
Journal of Retailing and Consumer Services,
Vol.14, pp.208-215.
Mittal B., Ross W., Baldasare P.M., 1998, The
asymmetric impact of negative and positive
attribute-level performance on overall satisfaction
and repurchase intentions, Journal of Marketing,
Vol. 62 (1), pp.33-47.
Mittal B., Lassar W. M., 1998, why customers switch?
The dynamics of satisfaction versus loyalty, Journal
of Services Marketing, Vol. 12(3), pp.174-194.
Panahi M., Pakeniat M. - Marketing Principles of a Neglected Banking Service: Safe Deposit Box Rental
Monireh Panahi, Mohammad Pakeniat
Business Intelligence Journal - July, 2011 Vol.4 No.2
302 Business Intelligence Journal July
Moutinho, L., Brownlie D., 1989, Customer
satisfaction with bank services-a multidimensional
space analysis, International Journal of Bank
Marketing, Vol. 7 (5), pp.23-27.
Newman, K. (2001). Interrogating SERVQUAL: a
critical assessment of service quality measurement
in a high street retail bank, International Journal of
Bank Marketing, Vol. 19(3), pp. 126-139.
Ndubisi N. O., 2003, Service quality: understanding
customer perception and reaction, and its impact
on business, International Journal of Business, Vol.
5(2), pp.207-219.
Oliver, R.L., 1993, A conceptual model of service
quality and service satisfaction: compatible goals,
different concepts, in Swartz T.A., Bowen D.A. and
Brown S.W.(Eds), Advances in Service Marketing
and Management, Vol.2, JAI Press, New York, NY,
pp.65-85.
Rosenberg, L. J., & J. Z. Czepiel, 1983, A marketing
approach for customer retention, The Journal of
Consumer Marketing, Vol. 1, pp.45-51.
Smith A. M. , Lewis B. R., 1989, Customer care in
fnancial service organizations, International
Journal of Bank Marketing, Vol. 7 No.5, pp.13-
22 MacKenzie S.B. and Olshavsky, R.W., 1995,
A reexamination of the determinants of consumer
satisfaction, Journal of Marketing, Vol.60 (3), p.15
M. Q. Patton (2001) Qualitative Research and
Evalutaion Methods, SAGE Publications.
Spreng R. and R. Mckoy, 1996, An empirical
examination of a model of perceived service quality
and satisfaction, Journal of Retailing, vol.72, pp.
201-214.
Thwaites D., Vere L., 1995, Bank selection criteria-a
student perspective, Journal of Bank Marketing,
Vol. 11 (1-3), pp. 133-149.
Wang Y., H. P. Li and Y. V. Hui, 2003, The antecedents
of service quality and product quality and their
infuences on bank reputation: evidence from
banking industry in China, Managing Service
Quality, Vol. 13(1), pp.72-83.
Yavas U., Benkenstein M., Stuhldreier., 2004,
Relationships between service quality and
behavioral outcomes: a study of private bank
customers in Germany, International Journal of
Bank Marketing, Vol. 22 (2/3), pp.144.
2011 303
STRATEGIC MARKETING AND FIRMS PERFORMANCE: A
STUDY OF NIGERIAN OIL AND GAS INDUSTRY
Akinyele, S.T. (PhD)
School of Business, Covenant University,Ota-Nigeria.
Email: [email protected]
Abstract
The purpose of this paper is to investigate the impact of strategic marketing and frms performance of the Nigerian oil and
gas industry. This study adopted a survey research methodology to examine strategic marketing and frms performance of Nigerian
oil and gas marketing companies in an attempt to attain their desired level of performance. One hypothesis was formulated from the
statement of research problem. Analysis of Variance, Pearson Moment Correlation Analysis, Factor Analysis among other statistical
tools were used in testing the hypotheses. The overall results suggest that strategic marketing is a driver of organizational positioning
in a dynamic environment, and that it helps to enhance the development of new product/service for existing markets. These fndings,
along with other interesting fndings of the study, are discussed. From the empirical and anecdotal managerial evidence as well
as from the literature implications are drawn for the effcient and effective strategic marketing in the Nigerian oil and gas industry.
Key words: Strategic Marketing, Firms, Deployment, Resources, Performance
The sensitivity of petroleum resource is clearly refected
in the fact that it has remained or continued to be the goose
that lay a golden eggs for the Nigerian economy as well
as the supreme foreign exchange earner contributing over
80% of government revenues and helps the development
of Nigeria’s infrastructures and other industries (Chukwu
2002). However, due largely to the highly technical nature of
exploration and production, the sector depends substantially
on imported technologies and facilities for most of its
operations. In view of the critical importance of the sector
to the nation’s economy and its capacity to generate far-
reaching multiplier effect, the grooming of highly skilled
indigenous manpower to participate keenly in the activities
of the sector to redress the foreign dominance becomes
imperative (Baker 2006). With a current production capacity
of about 30 million barrels per day (bpd), Nigeria plans to
increase her production capacity to about 40 million bpd
by 2010 (Mathiason 2006). Already, Nigeria is the leading
oil and gas producer in Africa, currently ranked the seventh
highest in the world (The Guardian 2006).
In addition to the above, Nigeria which is widely
referred to as a gas province, has natural gas reserves that
triple crude oil reserves, being estimated in excess of 187.5
trillion standard cubit feet(scf) (Africa Oil and Gas 2004).
Baker (2006) observes that Nigeria began exporting oil
in 1958 with crude oil production of 5000 barrels per day
(bpd) rising by 1979 to a peak of 2.3 million bpd.. However,
gas resources are largely untapped and Nigeria’s gas
reserves place it among the top ten countries in the world
in that category ( Ekpu 2004). Ekpu (2004) also observe
that other energy resources such as hydro power, wind
energy, and coal, which is produced in Enugu and Benue
States abound in the country. Energy consumption is in the
area of petroleum products, which according to (Dixton et
al 2005), accounted for between 70% and 80% of total
energy consumed in Nigeria between 1970 and 1980, the
major consumers being the transportation, household and
industrial sectors.
Marketing has been defned and conceptualized in
various ways, depending on the author’s background,
interest, and education (Osuagwu 1999). For example,
marketing can be seen as a matrix of business activities
organized to plan, produce, price, promote, distribute, and
megamarket goods, service, and ideas for the satisfaction
of relevant customers and clients. Achumba and Osuagwu
(1994) also posit that marketing is important for the success
of any organization, whether service- or product-oriented.
Mathiason (2006) argues that the oil and gas service
sector constitutes a service industry that has currently
been changed by aggressive strategic marketing behaviour.
According to (Li et al 2000), indigenous Nigerian oil
and gas marketing companies were not profoundly
entrepreneurial at the beginning for the following reasons:
lack of trained manpower, poor infrastructural development,
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
304 Business Intelligence Journal July
lack of adequate or suffcient capital base on the part of the
indigenous oil and gas marketing companies and intense
competition from superior foreign companies.
This study is intended to expand the body of knowledge
in respect of the application of strategic marketing to the
oil and gas sector especially in a developing economy like
Nigeria that earns over 80% of her foreign exchange from
oil and particularly, as the Federal Government is putting
in place policies and strategies to improve the oil sector’s
contributions to the Nigeria economy (Garuba 2006).
Statement of Research Problem
The problem statement, describes the content for the
study and it also identifes the general analysis of issues
in the research necessitating the need for the study
(Osuagwu 1999). The research is expected to answer
questions and provide reasons responsible for undertaking
the particular research (Pajares 2007). Many research
efforts in the area of marketing practices in developing
economies have dealt with macro issues and emphasized
the management of company’s structure and strategies,
conduct and performance of marketing activities as they
relate to performance indices such as market share, growth,
effciency and well being of consumers and clients (Samli
and Kaynak 1994) lament that the key defect with this static
and macroanalysis of marketing practices in developing
economies is that it minimizes the impact of marketing
environment on the achievement of performance measures.
Also, although some research efforts have been undertaken
to explain marketing practices in developing economies at
the organizational level (Chukwu 2002), many of these
research efforts do not provide answers to issues pertaining
to the impact of company’s structure and strategies on the
performance of mineral prospecting industries particularly
the oil and gas marketing companies. The deregulation of
the Nigerian economy through the Structural Adjustment
Programme (SAP) affected the oil and gas sector in Nigeria
in many ways . Consequently, the Nigerian oil and gas
companies incorporated the usage of various market mix
elements to improve their market outreach/ coverage, new
product ratio, price positioning, competitive orientation,
etc to survive and grow (Johne and Davies 2002). The
poor condition of some oil and gas marketing companies
in Nigeria is a function of some interrelated problems.
According to (Li et al 2000 ), the causes of the oil and gas
marketing companies failure or poor performance, are due
to microeconomic or macroeconomic factors (performance
industry environmental factors indice coupled with the
management of marketing content and product marketing).
Garuba (2006) have, however, posited that oil and gas
companies poor performance in Nigeria is a function of
industry environmental factor indices and marketing of oil
and gas services. It evolves as a result of the interplay of
the marketing mix elements and the environmental factors,
which impact on these elements (Li et al 2000 ; Mavondo
2000). Therefore, the function of marketing strategy
deals with determining the nature, strength, direction,
and interaction between marketing mix elements and the
environmental factors in a particular situation (Osuagwu
2001; 2001; 2004). However, achieving effcient and
effective product marketing strategy by an organization
is diffcult, as a result of the ambiguity and instability
of environmental factors . Sound and robust marketing
commitment on the part of organization and sales-people
are important to the survival and growth of the oil and gas
industry, considering the subtle, unstable and seemingly
hostile business environments in which contemporary
business organizations operate (Osuagwu 1999). In order
to formulate and implement effective and effcient goal
actualization and inter-industry marketing commitment in
product distribution, oil and gas companies should have
a thorough and continuous understanding of the relevant
environment that impacts on their marketing strategies.
Objectives of the Study
The main focus of marketing activities of oil and gas
marketing companies is the identifcation and satisfaction
of clients’ needs and wants. These objectives can be
attained by identifying the likely marketing mix variables
and strategies, including relevant environmental impacts
on them. There is, therefore, the need to carry out this
research given the enormity of the problem facing the oil
and gas industry. Specifcally the study sought to explore
the possible differences in the organizational structure and
strategies adopted and how they affect the performance of
Nigerian oil and gas marketing companies.
Research Hypothesis
The organizational structure and strategies adopted
do not affect the market share of Nigerian oil and gas
marketing companies.
2011 305
Literature Review
A service may therefore be seen as an activity or beneft
which can be offered to an organization or individual by
another organization or individual and which is essentially
intangible. It is a separately identifable but intangible
offer which produces want-satisfaction to the client, and
which may or may not be necessarily tied to the sale of
a physical product or another service (Osuagwu 1999).
Services include a wide range of activities and form some
of the growing sectors of the economies of developed and
developing countries. Services include professional services
(legal, accounting, medical, management consulting,
etc), general services (insurance, postal, telephone,
transportation, internet ,tourism, etc), maintenance and
repair services, and services from marketing researchers
and product manufacturers, among others. Oil and gas
service is not a tangible thing like food, clothing and car.
Sound and robust marketing strategies are important to
the survival and growth of any business, including oil and
gas business, considering the increasingly subtle, unstable
and seemingly hostile business environments in which
contemporary business organizations operate . Therefore,
in order to formulate and implement effcient and effective
marketing strategies, business organizations should have
a thorough and continuous understanding of the relevant
environment that impacts on their marketing strategies.
According to (Mavondo 2000), marketing strategy has
been a salient focus of academic inquiry since the 1980s.
There are numerous defnitions of marketing strategy
in the literature and such defnitions refect different
perspectives ( Li et al 2000). However, the consensus is
that marketing strategy provides the avenue for utilizing the
resources of an organization in order to achieve its set goals
and objectives. Generally, marketing strategy deals with
the adapting of markeing mix-elements to environmental
forces. It evolves from the interplay of the marketing mix
elements and the environmental factors (Li et al 2000).
Therefore, the function of marketing strategy is to determine
the nature, strength, direction, and interaction between the
marketing mix- elements and the environmental factors in
a particular situation . According to (Levie 2006), the aim
of the development of an organization’s marketing strategy
development is to establish, build, defend and maintain
its competitive advantage. Against this background, the
present research attempts to assess the marketing strategies
of Nigerian oil and gas marketing companies, the impacts
of environmental factors on such strategies and the
effectiveness of the marketing strategies. The marketing
strategies of Nigerian oil and gas marketing companies are
expected to be adaptable to these environmental factors
in order to achieve set performance goals. The oil and gas
industry seems to have witnessed some form of corporate
performance over the years which can be attributed to their
distinct level of market share (Akinyele, 2010 ).
Marketing strategies and tactics are concerned
with taking decisions on a number of variables to
infuence mutually-satisfying exchange transactions and
relationships. Typically, marketers have a number of
tools they can use, these include megamarketing and
the so-called 4Ps of marketing , among others. Marketing
seems easy to describe, but extremely diffcult to practice
(Kotler and Connor 1997). Marketing is one of the salient
and important organic functions which help to service
organizations to meet their business challenges and
achieve set goals and objectives . The word “ service” is
used to describe an organization or industry that “does
something” for someone, and does not “ make something”
for someone (Silvestro and Johnston 1990). “Service” is
used of companies or frms that meet the needs and wants
of society such organizations are essentially bureaucratic .
“ Service” may also be described as intangible its outcome
being perceived as an activity rather than as a tangible
offering. The question of the distinction between services
and tangible products is based on the proportion of service
components that a particular offering contains.
Definition of Strategic Marketing
The early strategic marketing - performance studies
date from the time of rapid expansion of formal strategic
marketing in the 1960s (Levie 2006). Although same studies
employed diverse methodologies and measures, they shared
a common interest in exploring the fnancial performance
consequences of the basic tools, techniques, and activities
of formal strategic marketing i.e. systematic intelligence-
gathering, market research, SWOT analysis, portfolio
analysis, mathematical and computer model of formal
planning meetings and written long- range plans. Osuagwu
(2004) observes that strategic marketing is a disciplined
effort to produce fundamental decisions and actions that
shape and guide what an organization is, what it does, and
why it does it, with a focus on the future. Akinyele (2010),
reports that newly invented strategic marketing displaced
long range planning because of the growing discontinuity of
the environment. Strategic marketing on the other hand does
not necessarily expect an improved future or extrapolatable
past. Grewal and Tansuhaj (2001), argued that a manager
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
306 Business Intelligence Journal July
Table 2 presents the descriptive statistics of the
effectiveness of strategic marketing practices of the studied
oil and gas industry. The fnding shows that strategic
marketing practices have been reasonably effective in oil and
gas industry, with strategic marketing effectiveness being
very high from the analysis above. The essence of strategic
marketing is to achieve set objectives, and these objectives
can be measured in terms of proft, market share, marketing
cost, gross earnings, capital employed, asset quality, quality
of marketing management, liquidity, turnover of marketing
staff, and management of departmental crisis. The
Table 2: Descriptive Statistics of Effectiveness of Strategic
Marketing Using Qualitative Measures of Performance(n=
286).
Variables Mean
Std.
Dev.
Variance Skewness
Company makes proft by
selling large quantities of
goods/services
5.17 .96 .89 -.1.27
Experience to cut costs is an
important goal
4.67 .99 .99 -.79
Sales executive move faster
than competitors in responding
to customers needs
4.93 .98 .87 -.99
Develops an exhaustive set
of alternatives before making
improvement management
decision
4.79 .89 .79 -.69
Demands better services
provided by customers
4.17 1.23 1.29 -.39
Emphasize opening up new
branches
4.69 1.23 1.27 -.32
Ability to gain market share is
high
3.57 1.39 1.79 -.34
is not necessarily a strategist and that a manager’s vision is
also not an entrepreneurial vision. He explains that while
the manager would rather have an orientation point of
guiding a company in a specifc direction, an entrepreneur
having strategic competence should state his vision clearly,
aggressively and in an optimistic manner.
Methodology
A cross- sectional survey was selected for this study
because it was easy to undertake compared to longitudinal
survey and the result from the sample can be inferred
to the larger population. In addition, some extraneous
factors could have manifested in the observed change
other than the independent variable concerned. The target
population in the study was the employees of product
pipeline marketing companies in Lagos, Nigeria. The
questionnaire was pre-tested with respondents in other
product marketing company, to authenticate reliability.
The pre-testing was done to avoid any possible infuence
on trial respondents before the actual survey. The analyzed
data was presented using descriptive statistics, frequency
tables, Analysis of Variance, and Correlation coeffcients.
Descriptive statistics allow the generalization of the data to
give an account of the structure or the characteristics of the
population as represented by the sample.
Data Analysis, Finding and Discussions
Table 1: Reliability Coeffcients of Research Measures
(Cronbach’s Alpha)
S/N Variable Measure
Cronbach’s Alpha
Coefcients
1 Management of Marketing strategy 0.76
2 Oil and Gas Performance Measurement 0.73
3 Efect of Environmental factors on
marketing strategies
0.65
4 Organizational structure and strategi 0.84
Table 1 above shows Cronbach’s alpha coeffcients of
the major research measures. “Management of marketing
strategy contructs” and “Oil and gas performance
measurements” met Nunnally’s (1978) internal consistency
(reliability) standard for newly- developed research
measures, while “Effect of environmental factors on
marketing strategies” failed to meet Nunnally’s (1978)
standard for reliability. Specifcally, Nunnally (1978)
recommended 0.70 Cronbach alpha value (internal
consistency) for newly developed research instruments.
Therefore, subject to the specifc and usual limitations
associated with this type of research, the research instrument
appears reliable and valid.
This study has provided empirical evidence pertaining
to the perception placed on oil and gas marketing strategies,
and oil and gas performance measures and impact of
environmental factors on such strategies. The research
fndings show that product and mega marketing strategies
received relatively low perception. These fndings have
managerial and research implications.
2011 307
effectiveness of strategic marketing practices in the studied
oil and gas industry is encouraging. These are the CAMEL
measures of performance. According to (Osuagwu 1999),
the effectiveness of oil and gas strategies determines the
survival and growth of downstream sector in Nigeria,
especially in an ever- changing environment. Effective oil
and gas management through strategic marketing assists
in the employment of capital raised, and manages the oil
and gas asset portfolio in viable business options so that
the assets are seen to be performing and yielding returns.
The marketing strategies of oil and gas, in order to show
reasonable levels of effectiveness along the CAMEL
measures, have to emphasize a marketing management
team with foresight, experience, and commitment towards
the survival and growth of the oil industry, among others.
Osuagwu(2001), posit that the most widely accepted
measure of performance of oil and gas industry is current
proftability, which is measured in terms of return on assets
and return on equity. Nigerian oil and gas industry that
creates comparatively large amounts of value (in relation
to its equity) through it strategic marketing practices can
be said to show high level of effectiveness. And as Table
3 shows, the studied oil and gas industry’s have shown
appreciable levels of effectiveness using the identifed
measures of performance.
Table 3: Comparison of Environmental Characteristic
a
Dimension of the Environment
C
o
m
p
a
n
y
A
T
o
t
a
l
M
e
a
n
C
o
m
p
a
n
y
B
O
a
n
d
o
M
e
a
n
C
o
m
p
a
n
y
C
T
e
x
a
c
o
m
e
a
n
C
o
m
p
a
n
y
D
A
g
i
p
M
e
a
n
Markets
Product diversity 4.64 4.29*** 3.22 3.12
Geographical diversity 5.40 4.22*** 2.34 3.21
Level of product information 4.89 4.80 4.33 4.02
Diversity of promotional media 4.92 4.46* 4.06 3.42
Competition
Intensity of rivalry 5.69 5.36** 4.34 4.54
Inability to infuence market
conditions
4.21 3.60*** 3.34 3.18
Average proftability of the principal
market
4.32 4.26 4.21 4.65
Entry barriers to the principal market 4.69 5.42*** 4.32 4.24
Constraints imposed by inter-industry
relationships with major stockholders
2.84 3.09 3.33 3.24
With major distributors and customers 2.76 3.78*** 3.43 3.11
With major suppliers-subcontractors 2.49 3.71*** 3.34 3.54
With government 4.38 3.27*** 3.12 3.43
With competitors 2.14 2.64*** 2.42 2.56
Notes: a. The higher the mean score, the more typical is the characteristics
* Signifcant at 0.5 level by t-test of means
** Signifcant at 0.1 level by t-test of means
*** Signifcant at.001 level by t-test of means
Dimension of the Environment
C
o
m
p
a
n
y
A
T
o
t
a
l
M
e
a
n
C
o
m
p
a
n
y
B
O
a
n
d
o
M
e
a
n
C
o
m
p
a
n
y
C
T
e
x
a
c
o
m
e
a
n
C
o
m
p
a
n
y
D
A
g
i
p
M
e
a
n
Ability of labour market
For managers 3.64 1.79*** 3.11 3.23
For technological experts 3.47 1.99*** 2.23 2.65
From the above table, there is also a signifcant difference
in labour market-ability between the four companies. Total
frms face a less mobile labour market than oando, Texaco
and Agip frms. Not new, the fndings is consistent with the
prevalent view that the Total labour market is less mobile
because of its many tangible incentives incorporated
into their employment system. The strengths and range
of constraints imposed by interrelationships with other
organizations are also different in Total and other oil and
gas companies under study. Oando and other oil and gas
frms face stronger constraints imposed by government and
regulatory bodies, while Total frms feel the constraints
imposed by their relationships with distributors, customers,
suppliers and competitors to a greater degree than Oando,
Texaco and Agip frms. This result suggests that Total
companies create closer inter-organizational networks with
various kinds of organizations. The networks, although
constraining decisions within organizations, may have a
number of benefts including risk-sharing and long term
stabilization of business. The strong constraints imposed by
the government upon oil and gas companies probably stem
from the relatively adverse historical relationship between
business and government in Nigeria as well as from motives
to protect the public and promote competition.
To sum up, Total frms face a less diverse, less competitive,
more volatile and high opportunity environment, and less
mobility of market. They are, moreover, constrained by
interrelationships with other organizations to a greater
extent than the other oil and gas marketing frms under
study. A frm must set operational objectives, the priorities
of which are contingent upon the opportunities provided
and constraints imposed by its environment matched
against the internal capabilities of the organization.
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
308 Business Intelligence Journal July
a
R Squared= .382 (Adjusted R Squared=.361)*
signifcant at 0.005 two tailed test.
The fndings in Table 5 indicate that there was a
signifcant difference in performance at f (4,285)= 5.457, at
0.05 signifcant level. This implied that the frst hypothesis
was rejected and the alternative hypothesis retained, which
meant that the company’s structure and the strategies
adopted by Total yielded a better market share than that
of their counterparts in the industry.
Discussion of the Findings
As stated earlier, the discussion of this study followed
the hypotheses raised and tested and they are presented
below:
The organizational structure and strategies adopted
do not affect the market share of Nigerian oil and gas
marketing companies.
The null hypothesis one which stated that” the
organizational structure and strategies adopted do
not affect the market share of Nigerian oil and gas
marketing companies” was rejected. This implied that
the organizational structure and strategies adopted by
oil and gas marketing companies affect market share
positively. Several empirical studies have concluded that
an organizational structure and strategies adopted is indeed
an important element in companies’ success. (Levie 2006),
gave evidence that the level of organizational structure and
strategies is positively related to company effectiveness,
while (Grewal and Tansuhaj 2001 ) reported that more
successful companies have well defned organizational
structures, in sharp contrast to less successful companies.
Focusing on large frms, (Ekpu 2004); found a positive
relationship between the unstructured organizational
patterns and large frm fnancial performance. The fndings
by (Osuagwu, 2004; Akinyele 2010) underscore this
fnding, as they established that changes in the market
environment, business strategy and organizational structure
have impact on companies performance.
Table 4: Mean and Standard Deviation on Company’s Structure
and Strategies Adopted and Market share Attained by Nigerian
Oil and Gas Marketing Companies
Questions Freq. Mean
Standard
Deviation
Company’s structure and strategies adopted by
Total frm
286 70.40 2.18
Company’s structure and strategies adopted by
Oando frm
286 68.40 1.18
Company’s structure and strategies adopted by
Texaco frm
286 42.64 1.22
Company’s structure and strategies adopted by
Agip frm
286 68.20 0.22
Oil and gas companies performance 286 64.40 1.16
Table 4 above shows the difference between means of
company’s structure and strategies and resultant market
share of Nigerian oil and gas marketing companies. Total
frm with a mean of 70.40 and standard deviation of 2.18
had a better structure compared with other counterparts in
the industry with mean of 68.40 and standard deviation of
1.18 respectively. The Table also shows that with the mean
of 64.40 and standard deviation of 1.16 when divided
between the companies, the market share of Total was better
than that of the other counterparts. The question then is,
was the difference signifcant enough or was it the result
of a sampling error? The answer is presented in Table 5.
Table 5 Summary of Analysis of Variance on Company’s
Structure and Strategies Adopted and Market share Attained by
Nigerian Oil and Gas Marketing Companies Respectively
Source
TypeIII
Sum of
squares
Df
Mean
square
F Sig.
Corrected Model 294.808 9 32.644 35.415 .000
Intercept 1887.508 1 1887.508 1749.596 .000
Company Types 20.014 1 5.654 6.240* .000
Company’s structure
&strategies
22.757 4 5.684 6.231 .000
Company Types*
Company’s structure &
strategies
24.745 4 5.086 5.457 .001
Source: Field Survey, 2007.
Hypothesis Testing
The frst hypothesis tested in the study states that ‘
The organizational structure and strategies adopted do not
affect the market share of Nigerian oil and gas marketing
companies.
Source
TypeIII
Sum of
squares
Df
Mean
square
F Sig.
Error 581.174 283 .816
Total 5758.000 285
Corrected Total 667.984 286
2011 309
Conclusions
This section elaborates on the conclusion of the research.
Based on the fndings of this research, the following
conclusions are warranted:
1. The evidence from fndings suggested that oil and gas
marketing companies have comparative advantages in
adopting various marketing strategies using different
technologies. Oil and gas marketing companies
appeared to specialize in the use of traditional methods
of marketing, which is based on “soft” information
culled from close contacts by marketing and sales
department rather than the use of the specialized
strategic marketing methods that are based on “hard”
quantitative information.
2. Most of the fndings of the research are consistent with
previous normative and empirical works. For instance,
the companies face a less diverse, less competitive,
more volatile and high opportunity environment, and a
less mobility of market. They are however, constrained
by interrelationships with other organizations to a
greater extent.
3. The research instrument shows high validity and
reliability.
4. This study has provided empirical evidence pertaining
to the perception of oil and gas marketing strategies, and
the industry environmental factors on such strategies..
Managerial and Research Implications
The fndings of this study have several managerial
implications for Nigerian downstream oil and gas. First,
Nigerian oil and gas managers are advised to place more
emphasis on the adoption of various marketing strategies
using different technologies. Second, all organizations
face an external business environment that constantly
changes. Changes in the business environment create both
opportunities and threats to an organization’s strategic
development, and the organization cannot risk remaining
static. It must monitor its environment continually in
order to: build the business, develop strategic capabilities
that move the organization forward, improve the ways in
which it creates products/services and develops new and
existing markets with a view to offering its customers
better services. Third, anticipating competitors’ actions
and reactions to the organizational’ moves may be the key
determinant of success for any marketing strategy. Fourth,
with the competitive downstream oil and gas industry of
today, participants can put more emphasis on relationship
marketing to ensure effectiveness. This essentially entails
personalizing the oil and gas services offered to clients,
attending to clients’ cultural and social activities, in relation
to other non- business activities, which are of interest to
clients. Fifth, Nigerian oil and gas marketers should be
sensitized to the importance of their offerings to their
clients, including the impressions their clients have of
those offerings. As oil and gas clients are demanding more
quality from their petroleum product marketing companies
(PPMC), it may be strategic to inject the idea of total
quality management (TQM) and its variants among product
marketers.
Recommendations, Limitations and
Suggestions for Further Studies
This study indicates that strategic marketing have
a signifcant impact on performance variables and that
they interact with the different components to facilitate
performance. It also indicates that different performance
factors moderate strategic marketing practice. Therefore,
organizations hoping to enhance corporate performance
in a dynamic business environment should consider the
following:
Recommendations
a. The concepts and principles of total quality
management (TQM) are recommended for holistic study,
in addition to contemporary marketing management issues
such as relationship marketing, value analysis, business
process re-engineering, megamarketing, re-marketing, co-
marketing, bench marketing, and permission marketing,
among others.
b. Efforts should be made by organizational marketers
to understand the relevant factors that affect both clients’
behaviours, and the strategic options to be adopted to cope
with such behaviours.
c. Oil and gas marketing scholars or researchers
should endeavour to study holistically the relevant business
functions and activities which may enhance or hinder the
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
310 Business Intelligence Journal July
Journal of Management and Business Research,
vol.10(6):60-77.
African oil and gas(2004). Energy policy agenda
.International energy services Ltd, Lagos.
Baker, M.(2006). Petroleum technology development
fund: Intervention to discover and develop
technocrats for the energy sector of tomorrow.
Petroleum Journal, Vol. 5(3),112-115.
Chukwu, I.(2002). Crude oil development, The Post
Express, Lagos, Thursday, October 1, p. 24-25.
Dixton, D. , Donald, F. , & Roger, A. (2005). Basic
Marketing: Concepts, Decisions and Strategies,
Englewood Cliffs, N.J. : Prentice- Hall Inc.
Ekpu, R.(2004). Associated gas utilization, Nigeria’s
oil and gas, 3(8) August,pp.17-18.
Garuba, D.S.(2006). Survival at the margins:
Economic crisis and coping mechanisms in rural
Nigeria, Local Environment, Vol. 11, (1), p.17-36.
Grewal, R. and Tansuhaj, P.(2001). Building
organizational capabilities for managing economic
crisis: The role of market orientation and strategic
management. Journal of Marketing, 65: 67-80.
Johne, A. and Davies, R.(2002). Innovation in medium-
sized oil companies: How marketing adds value,
The International Journal of Bank Marketing, 18(1).
Lavie, D.(2006). Capability reconfguration: An
analysis of incumbent responses to technology
change. Academy of Management Executive,
31(1):153-174.
Li, S. , Kinman, R., Duan, Y. and Edwards, J.S.(2000).
Computer-based support for marketing strategy
development, European Journal of Marketing,
34(5/6).
Mathiason, N.(2006). Oil delta burns with hate, The
observer(UK), Sunday, January, 29(http://observer.
Guardian.co.uk/business.
Mavondo, F.T.(2000).Marketing as a form of
adaptation: Empirical evidence from a developing
economy, Marketing Intelligence and Planning,
18(5),256-272.
Nunnally, J.(1978).Psychometric Theory, 6th Edition,
New York:McGraw –Hill.
0suagwu, L.(1999). Marketing: Principles and
Management, Lagos: Grey Resources Limited.
understanding and subsequently applicability of relevant
modern management concepts and principles to oil services
marketing.
d. Firms that are not operating in a dynamic business
environment need not adopt a strategic marketing practice
as this may cause the frm to look inconsistent in the eyes of
its customers and eventually reduce effective performance.
e. The need for the identifcation of options and
resources and of capabilities of deployment constitutes an
impetus to effective strategic marketing implementation,
since the practice derives from capabilities in assembling
and maintaining an appropriate resource portfolio and
coupling the resource portfolio with the identifcation and
recognition of options.
Suggestions for further studies
This research leads to some observations that might be
of interest to future researchers, as they represent the seeds
from which future research can be developed.
a. This same research can be carried out in other
nations so that a broad comparison of the concepts of
strategic marketing as it affects frm performance can be
made.
b. Research into the combined effects of strategic
marketing practice and performance factors as mediators
of frm performance relationship.
c. Research into the effects of key characteristics of
industries environmental indices and marketing strategy
could be carried out to further explain the differences in the
frm’s adoption of strategic marketing.
d. Finally, future research works are to be undertaken
in order to refne the cobwebs found in the present research,
and orient it to more specifc contexts (business, time,
location, etc) in Nigeria’s oil and gas industry.
References
Achumba, I.C. and Osuagwu, L.(1994). Marketing
Fundamentals and Practice, Rock Hill: USA: Al-
Marks Educational Research, Inc.1-400.
Akinyele, S.T. (2010). Signifcance of strategic
marketing to enterprise performance: An empirical
analysis of Nigerian oil and gas industry. Global
2011 311
Osuagwu, L.(2001). An evaluation of the marketing
strategies of Nigerian insurance companies,
Academy of Marketing Studies Journal, Vol.
5(2),pp.17-30.
Osuagwu, L.(2001). Adoption of marketing strategies
in Nigerian institutions, Academy of Marketing
Studies Journal, Vol. 5(1): 1-16.
Osuagwu,( 2004). Relationship marketing strategies in
Nigerian companies, The Marketing Management
Journal, Vol. 14(2): 114-128.
Pajares, F.(2007). Elements of a proposal, available
from the author at www.aes.emory.edu/mfp/
proposal.html, download on 10th May, 2008.
Samli, A.C and Kaynak, E.(1994). Marketing practices
in less developed countries, Journal of Business
Research, 12:5-18..
Silvestro, R. and Johnson, R.(1990). The determinants
of service quality-enhancing and hygiene factors,
Proceedings of the QUISS 11 Symposium, New
York: St. John’s University.
The Guardian(2006). What is wrong with the Nigerian
oil industry. Wednesday, January 25,p.1-2, Lagos.
Akinyele S. T.. - Strategic Marketing and Firms Performance: A Study of Nigerian Oil and Gas Industry
Akinyele, S.T.
Business Intelligence Journal - July, 2011 Vol.4 No.2
312 Business Intelligence Journal July
INTUITIVE MANAGERIAL DECISION MAKING IN MALAYSIA
AND THE UNITED STATES
Dr. Isola Oluwabusuyi (MBA, MA, DBA)
School of Business and Accounting
Brown Mackie College, Atlanta
Email: [email protected]
Abstract
The purpose of this paper is to examine and compare the factors that infuence intuition as a decision-making tool for managers
in Malaysia and in the United States. The underlying study examined the relationships among gender, management level, country of
operation, and the reported use of intuition in decision making. Agor’s Intuitive Measurement Survey (AIM) survey was adapted (with
permission from copyright owner) from Weston Agor’s study to measure the relationship between a manager’s reported use of intuition
in decision making and the manager’s management level, the manager’s gender, and the manager’s country of operation. The research
shows signifcant relationship between research variables. Male managers in Malaysia’s reported use of intuition in decision making
was signifcantly lower than US managers reported use. Limitations of the study include the following: Samples were a combination of
criterion, maximum variation and convenience based selection of companies in Malaysia and the United States. United States samples
were selected from the East and West coast of the United States. The limitations could impact study’s external validity. Study’s fndings
are quite signifcant to global business managers intending to shift more of their activities to Asia in the near future. Multinational
corporations would have to provide more data for their Malaysian managers. This Study had fve signifcant fndings. Business schools
in the West may need to redesign their curriculum as more business managers feel more comfortable with intuitive decision making
techniques. Key words: Decision Making, Intuition, Global Management, Culture
Managerial decision making has always been a subject of
passionate academic discussion. What is the most effective
way of choosing between alternatives? Should managerial
decision making be intuitive or rational? Are managers
of companies facing relatively stable environments more
intuitive in their decision making than their counterparts
managing companies facing more unstable environments?
Are female managers more intuitive in their managerial
decision making than their male counterpart? Is managerial
decision making infuenced by a manager’s country of
operation?
Khatri &Ng (2000) discovered that top executives
of computer companies surveyed reported using more
intuition than their counterparts in banks and utilities.
The study also showed that computer industry was more
unstable than banking industry which was moderately
unstable and utilities’ industry which was relatively stable.
Financial performance in the computer industry was also
positively related to the use of intuitive decision making by
top executives in the industry.
Cappon (1993) tested over 3000 individuals and found
out that women did not have more intuition than men. He
believed that everyone had intuition and that it could be
developed in individuals (Fields, 2001). Cappon’s request
to administer his research instrument was turned down by
many intuition-sensitive companies but companies in the
manufacturing industry were quite receptive to his request.
This might have skewed his fndings. Cappon was a medical
doctor and psychotherapist.
Dane &Pratt (2007) in examining factors contributing to
effectiveness of intuitive decision making expatiated on the
role of domain knowledge, implicit and explicit learning,
and task characteristics on the effectiveness of intuition.
They concluded that as tasks become more judgmental, the
relationship between complex, domain relevant schemas
and effective decision making becomes stronger.
Intuition and gender
One widely-held view is that successful managers
are aggressive, forceful, competitive, self confdent,
independent and have a high need for control (Hayes et al,
2004). Loden (1985) argued that women have a lower need
for control and are more cooperative than men. Green and
2011 313
Cassell (1996) suggest that women are often characterized
as relatively submissive, nurturing, warm, kind and selfess.
In a study of sex stereotypes and leader behavior; Brenner
and Bromer (1981) reported that men are described as being
more analytical and logical and women as more intuitive.
Sex differences have been cited as the reason why
women are under-represented in management; they lack
the qualities for success and cannot perform as effectively
as male managers (Hayes et al, 2004). In agreeing with
this more compassionate and intuitive gendered¬ view of
women, Clare (1999), referred to intuition as one of the
valuable contributions that women bring to management.
Studies conducted to investigate the validity of this
stereotyping did not produce consistent results.
In a study by Wajcman (1996), successful women
managers were found to be in most respects, indistinguishable
from men in equivalent positions. Alban-Metcalfe and West
(1991) found a remarkable similarity in the way female and
male managers perceived themselves at work. Donnell and
Hall (1980) found no signifcant difference between male
and female managers in their study of 1,000 matched pairs
of female and male managers. However, Eagly and Johnson
(1990) found support for the absence and presence of
differences in their Meta analysis of studies of gender and
leadership style (Hayes et. al., 2004). Pacini and Epstein’s
(1999) study showed that women perceived themselves
as intuitive. They report that women are more likely than
men to identify themselves as engaging in experiential
processing and to judge themselves as being good at it
(Aarnio & Lindeman, 2005).
In summary, the search for gender differences in
information processing style has so far produced mixed
results. Self report studies produced results showing that
men and women support the existence of gender differences
in information processing; however some of the results are
actually contradictory. Some studies showed that women
see themselves as more intuitive than men while a few
self-report and in-depth studies showed men as being more
intuitive than women. In making sense of the results, Hayes
et al. (2004), suggested that observed pattern “appears to
lend support to the utility of the structural (Kanter, 1977)
and gendered culture (Green and Cassell, 1996) approaches
to understanding behavior in organizations”. The fact that
female managers showed more intuition than their non-
managing counterparts was construed to be their way of
adapting to a male dominated environment in which success
was determined by conformity to certain modes of conduct.
Country of Operation
Merriam-Webster online dictionary (2007) had four
different defnitions for a country: an indefnite usually
extended expanse of land, the land of a person's birth,
residence, or citizenship, a political state or nation or
its territory, the people of a state or district, and rural as
distinguished from urban areas. One of the variables
studied in this research is the country of operation. Research
question four examined the impact of country of operation
on reported use of intuition by executives. The literature
review on country of operation is hence focused on culture
as countries’ distinguishing factor.
Culture can be defned as the way of life of a group of
people. Damen (1987) also defned culture as “learned and
shared human patterns or models for living”. These patterns
and models pervade all aspects of human social interaction.
The use of proverbs to study cultures is a well-known
method in Anthropology. Lovell (2001) said “proverbs
can be the eyes that provide a window to a culture’s soul.”
Prahlad (2001) did a study of Jamaican culture through
Jamaican Proverbs gathered from Reggae music. In
studying Malaysia and the United States as countries of
operation, a review of literatures containing Western and
Malaysia proverbs was conducted.
Malaysia is a country of three major ethnic groups;
Malays, Chinese and Indians. The book of Analects is one of
the most revered sources of information on Chinese culture.
The book contains most of the sayings of Confucius and
other highly respected Chinese teachers. Relevant contents
of the book were contrasted with Hofstede’s (2003)
cultural universals to develop a comprehensive outlook
on the Chinese culture. The same procedure was carried
out for Western, Indian and Malay cultures to develop a
comprehensive view of those cultures.
Hofstede (1980) defned culture as a kind of collective
programming of the mind which distinguished members
of one category of people from another. In the 1970's, he
measured elements of national cultural systems that impact
behavior in work situations. His studies produced a total
of 116,000 questionnaires in two surveys held in 1968
and 1972 (Hofstede, 2003). The studies revealed four
main dimensions on which country cultures differ. They
were labeled power distance (PDI), uncertainty avoidance
(UAI), individualism (IDV), and masculinity (MAI). Later
research, which dealt with Asians as the subject, added
the dimension called "long-term orientation”. These fve
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
314 Business Intelligence Journal July
dimensions were used to compare the two cultures through
the eyes of their proverbs.
Power distance index (PDI) focuses on the degree of
equality, or inequality, between people in the country's
society. A high power distance ranking indicates that
inequalities of power and wealth have been allowed to grow
within the society. “Malaysia’s power distance score is 104
compared to the other Far East Asian countries average of
60. This is indicative of inequality of power and wealth
within the society (Hofstede, 2003)”. A review of the book
of Analects revealed the following Chinese proverbs.
“Yu Tzu said: There are few who have developed
themselves flially and fraternally who enjoy offending
their superiors. Those who do not enjoy offending superiors
are never troublemakers. The Superior Man concerns
himself with the fundamentals. Once the fundamentals are
established, the proper way (Tao) appears. Are not flial
piety and obedience to elders fundamental to the enactment
of humaneness?" (Lau, 1992, p. 1) The concept of superiors
practiced by the Chinese is a refection of the people’s belief
in power distance. Yu Tzu in this proverb demonstrates the
need to avoid offending those who occupy higher positions
in the power structure of the society. Parents are the most
prominent superiors in Chinese worldview.
Malay proverbs dealing with power distance are listed
below:
A lost wife can be replaced, but the loss of character
spells ruin. A deep look at this saying reveals a stratifed
society that values female fairly less than their male
counterparts. It is not of the characteristics of societies in
which power distance is widely accepted.
Individualism (IDV) focuses on the extent to which
the society reinforces individual or collective achievement
and interpersonal relationships (Hofstede, 2003). A
high individualism ranking indicates that individuality
and individual rights are paramount within the society.
Individuals in these societies may tend to form a larger
number of looser relationships. A low individualism
ranking typifes societies of a more collectivist nature with
close ties between individuals. Malaysia ranked low on
individualism with a score of 26, third highest for Far East
Asian countries, behind Japan's 46 ranking, and compared
to an average of 24 for Asian countries. Corresponding
quotes from the book of Analects are stated as follows:
“Tzu Kung asked, ‘Does the Superior Man also have
things that he hates?’ Confucius said; He does. He hates
those who advertise the faults of others. He hates those
who abide in lowliness and slander the great. He hates
those who are bold without propriety. He hates those who
are convinced of their own perfection, and closed off to
anything else. How about you, what do you hate? Tzu
Kung said I hate those who take a little bit of clarity as
wisdom; I hate those who take disobedience as courage;
I hate those who take disclosing people's weak points to
be straightforwardness." (Lau, 1992, p.17) Confucius here
demonstrates the need to avoid exposing others weaknesses.
The need to outdo others is almost non-existing in Chinese
worldview. This is in sharp contrast to Western worldview
of self before others.
Malay proverbs dealing with the topic of individualism
are listed below:
If you have, give; if you lack, seek.
The body pays for a slip of the foot, and gold pays for a
slip of the tongue.
A heavy load should be borne together as well as a light
load
As a bamboo conduit makes a round jet of water, so
taking counsel together rounds men to one mind. The need
for collectivism is emphasized by these Malay proverbs.
Masculinity (MAS) focuses on the extent to which a
society reinforces, or does not reinforce, the traditional
masculine work role model of male achievement, control,
and power (Hofstede, 2003). A high masculinity ranking
indicates the country experiences a high degree of
gender differentiation. In these cultures, males dominate
a signifcant portion of the society and power structure,
with females being controlled by male domination. A low
masculinity ranking indicates the country has a low level
of differentiation and discrimination between genders. In
these cultures, females are treated equally to males in all
aspects of the society. Quotes relevant to masculinity in the
book of Analects are stated as follows:
“Being robbed, Chi K'ang Tzu was upset, and
questioned Confucius about what to do. Confucius said, If
you were desire less, they wouldn't steal from you, even
if you were to offer them a reward to do so." (Lau, 1992,
p. 12) The authentic original Chinese culture does not
favor materialism. In fact, the society esteems peace and
harmony above competitiveness. Confucius here seems to
convey the need to shun materialism.
“Do not wait for the rice to be served at the knee” is
a Malay proverb that seems to extol hard work and shun
laziness.
“He who works as a slave, eats as a king” is an Indian
proverb extolling the virtues of hard work.
Uncertainty avoidance index (UAI) focuses on the level
of tolerance for uncertainty and ambiguity within the society
- i.e. unstructured situations (Hofstede, 2003). A high
2011 315
uncertainty avoidance ranking indicates the country has a
low tolerance for uncertainty and ambiguity. This creates a
rule-oriented society that institutes laws, rules, regulations,
and controls in order to reduce the amount of uncertainty.
A low uncertainty avoidance ranking indicates the country
has less concern about ambiguity and uncertainty and has
more tolerance for a variety of opinions. This is refected
in a society that is less rule-oriented, more readily accepts
change, and takes more and greater risks.
Malaysia is relatively low in uncertainty avoidance.
The country’s (UAI) is only 36, compared to an average of
63 for the Far East Asian countries. A search through the
book of Analects revealed the following sayings about the
dimension:
Chi Lu asked about serving the spirits. Confucius said,
"If you can't yet serve men, how can you serve the spirits?"
Lu said, "May I ask about death?" Confucius said, "If you
don't understand what life is, how will you understand
death?" (Lau, 1992, p. 11) Confucius in this teaching shows
how Chinese people are generally intolerant of ambiguity.
The emphasis in this teaching is to focus on what you know
and leave the ambiguous for other people.
Malay proverbs dealing with this dimension are listed
below:
Do not empty the water jars just because you hear the
thunder in the sky
Do not leave the tortoise at your feet and hunt for the
turtle on the sea shore. These proverbs also demonstrate the
society’s disdain for uncertainty.
Indian proverbs relevant to uncertainty avoidance index
were also gathered and listed below:
Don't bargain for fsh which are still in the water.
He who is a guest in two houses, starves.
Long-term orientation (LTO) focuses on the extent to
which the society embraces, or does not embrace long-term
devotion to traditional, forward thinking values. High long-
term orientation ranking indicates the country subscribe
to the values of long-term commitments and respect for
tradition. This is thought to support a strong work ethic
where long-term rewards are expected as a result of today's
hard work. A low long-term orientation ranking indicates
the country does not reinforce the concept of long-term,
traditional orientation. In this culture, change can occur
more rapidly as long-term traditions and commitments do
not become impediments to change. LTO related quotes
from the book of Analects are stated below:
“Confucius said: The superior man stands in awe of
three things: He is in awe of the decree of Heaven; He is in
awe of great men; He is in awe of the words of the sages.
The inferior man does not know the decree of Heaven;
he takes great men lightly, and laughs at the words of the
sages” (Lau, 1992, p. 16) Respect for tradition is implied by
the reference to great men in this teaching. Malaysia’s score
on this index is not available.
Only seven (7) countries in the Geert Hofstede (2003)
research have individualism (IDV) as their highest
dimension: USA (91), Australia (90), United Kingdom
(89), Netherlands and Canada (80), and Italy (76). The high
individualism (IDV) ranking for the United States indicates
a society with a more individualistic attitude and relatively
loose bonds with others. The populace is more self-reliant
and individuals look out for themselves and their close
family members. Review of Western proverb literatures
produced two groups of proverbs relating to individualism.
The frst group promotes individualism while the second
group promotes cooperation. The two groups are presented
below.
Every man must carry his own cross If you want a thing
done right, do it yourself If you want breakfast in bed,
sleep in the kitchen Paddle your own canoe Good fences
make good neighbors You are responsible for you The
need for individuals to fetch for themselves is expressed
in these proverbs. The central theme of the six proverbs is
that individuals should be prepared to solve their problems
without relying on others for help. This is in line with the
USA’s high score of 91 on this cultural dimension. However,
the literature review also revealed some proverbs that
promote cooperation more than individualism. Examples
of proverbs in this category are presented below.
A bicycle can't stand on its own because it's two-tired.
Honey catches more fies than vinegar.
No man is an island
The nail that sticks out gets pounded
A big tree attracts the woodsman's axe
The central theme of these fve proverbs seems to be at
odds with the earlier six. They encourage cooperation rather
than individualism. The possibility of some of the proverbs
being foreign is also real since the American society is not
entirely white. The most unusual of the proverbs is fourth
one that talks about nails getting pounded. The researcher
felt this particular proverb might have been imported from
Asia.
The next highest Hofstede (2003) dimension for the
United States is masculinity (MAS) with a ranking of 62,
compared with an average of 50 for all countries. This
indicates the country experiences a higher degree of gender
differentiation of roles. The male dominates a signifcant
portion of the society and power structure. This situation
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
316 Business Intelligence Journal July
generates a female population that becomes more assertive
and competitive, with women shifting toward the male role
model and away from their female role.
The literature review for this section turned up proverbs
that mostly support USA’s score on the dimension. The
supporting proverbs are presented before those that are not
supportive of the position.
Half a loaf is better than none.
If at frst you don't succeed, try, try again.
It's the early bird that gets the worm.
Make hay while the sun shines.
The need to keep trying until success is achieved is
expressed in the second proverb, this is indicative of the
male achievement model valued by the culture. The essence
of timeliness in trying to achieve the desired societal status
is emphasized by the third proverb. Two proverbs that are
not supportive of the male dominant, achievement oriented
tendency are listed below. The two proverbs look more like
what one would fnd in Asian cultural literatures.
Winning isn't everything.
Health is better than wealth.
The United States was included in the group of countries
that had the long term orientation (LTO) dimension
added. The LTO is the lowest dimension for the US at 29,
compared with an average of 45 for all countries. This low
LTO ranking is indicative of the societies' belief in meeting
its obligations. The frst group of proverbs gathered from
the literature review supports the low score of the country
on the dimension, while the second group seems to suggest
that the score on the dimension should have been higher.
The supporting proverbs are presented hereby presented.
Never put off till (until) tomorrow what you can do
today.
No time like the present.
A stitch in time saves nine
The second proverb in this second group is actually
more favorable towards long term orientation as people
are encouraged to make attempts to understand the past in
dealing with the future. The third is also similar in its theme
as people are encouraged to focus more on the long run.
All things come to him who waits.
He who fails to study the past is doomed to repeat it.
History repeats itself.
Nature, time, and patience are three great physicians.
The next lowest ranking Dimension for the United
States is power distance (PDI) at 40, compared to an
average of 55 for all nations. This is indicative of a greater
equality between societal levels, including government,
organizations, and even within families. This orientation
reinforces a cooperative interaction across power levels and
creates a more stable cultural environment.
A cat may look at a king.
Green leaves and brown leaves fall from the same tree.
If you want to judge a man's character, give him power.
Power corrupts; absolute power corrupts absolutely.
The society’s disdain for power permeates through the
four proverbs. This is also refected in the country’s very
low PDI score. Egalitarianism is one of the central tenets
of the American society. The frst proverb relates human
freedom to look to that of a cat. “If a cat may look at the king
- then I have a right to look where I please” Egalitarianism
is also further stressed by the fact that we all emanate from
the same source as expressed in the green leaf proverb. The
proverb implies that we are all the same inside regardless of
what we look like outside.
The last Geert Hofstede (2003) Dimension for the US is
uncertainty avoidance (UAI), with a ranking of 46, compared
to an average of 64 for all countries. A low ranking in the
uncertainty avoidance dimension is indicative of a society
that has fewer rules and does not attempt to control all
outcomes and results. It also has a greater level of tolerance
for a variety of ideas, thoughts, and beliefs. Researcher’s
review of Western proverb literatures produced two groups
of proverbs relating to uncertainty. The frst group promotes
risk taking while the second group promotes the need to
exercise caution. The two groups are presented below.
It's easier to ask forgiveness than permission.
It is better to die on one's feet than live on one's knees.
He who hesitates is lost.
He who dares wins
Don't cross a bridge before you come to it.
A watched pot never boils.
A coward dies a thousand times before his death.
The valiant never taste of death but once.
This frst group of proverbs encourage risk taking.
The ffth proverb claims fretting about future problems is
superfuous and the seventh proverb decries cowardice. It
teaches that worrying about a forthcoming disaster may
cause as much (or even more) pain as the disaster when it
occurs (but does neither change it nor make it easier). The
central theme of this group of proverbs is that fretting is
more destructive than risk taking. This position is affrmed
by the country’s low UAI score.
The second group promotes cautious approach to risk
taking and the proverbs in this group are presented below.
A bird in the hand is worth two in the bush.
A picture is worth a thousand words.
Cobbler, stick to thy last.
2011 317
Don't burn your bridges before they're crossed.
Don't count your chickens before they're hatched.
It's better to be safe than sorry.
The frst proverb in this second group asserts that what
you already have is worth more than what you dream about
and the third proverb tries to encourage people to stick to
what they know. The fourth proverb suggests that people
should not act in ways that would leave them with no
alternatives. The mitigating effect of this second group of
proverbs might be responsible for the middle of the road
ranking given to USA on the index.
Methodology
The quantitative research employed Agor’s Intuitive
Measurement Survey (with permission by copyright owner).
The AIM survey was administered to 100 participants from
the US and 100 participants from Malaysia. Questions on
the survey were developed to measure all independent and
dependent variables.
Questions on the survey measured the following
variables: Gender, Management Level, Country of
Operation, and Reported use of intuition in decision
making. Reported use of intuition in decision making was
the only dependent variable in the study, the remaining
three variables were independent variables.
Validity and Reliability of the Instrument
AIM Survey (Agor’s Intuitive Measurement Survey)
is a modifed MBTI (Myers-Briggs Type Indicator). The
instrument, therefore, uses the reliability and validity of
MBTI (Agor, 1984). Studies have found strong support
for construct validity, internal consistency, and test related
reliability of MBTI instrument (Thompson & Borello,
1986). Further, the instrument was designed to best measure
all the variables in this study. The questionnaire was
modifed and simplifed so it contained clear instructions,
questions, and possible answers.
Research Questions
The study’s three research questions explored the
relationship between the study’s independent and
dependent variables, using the following subjects and their
respective management experience: Malaysia’s business
owners, executives, managers and supervisors; US business
owners, managers, executives and supervisors; Malaysia’s
male and female business owners, executives, managers
and supervisors as well as United States male and female
business owners, executives, managers and supervisors.
Research Question 1: What is the relationship between
management level and reported use of intuition in decision
making?
Research Question 2: What is the relationship between
country of operation and reported use of intuition in
decision making?
Research Question 3: What is the relationship between
country of operation and use of intuition in decision
making?
Assumptions and Limitations
The following assumptions were formulated that were
central to the design of this research.
1. Respondents understand the questions and are able
to answer all of them in the questionnaire.
2. Answers to the questionnaire are given with the
respondent’s knowledge and that answers were truthful.
3. Data collection process was reliable.
4. Analysis tools were accurate
Below are some limitations that may infuence the
results. These limitations include:
1. The samples were a combination of criterion,
maximum variation and convenience based selection of
companies in Malaysia and the United States.
2. The United States samples were selected from the
East and West coast of the United States.
Findings
Research Question 1: What is the relationship between
management level and reported use of intuition in decision
making?
Finding 1.1: Malaysia’s male supervisors’ use of
intuition is lower than US male supervisors’. Mean scores
of Malaysia’s male supervisors 6.40; US male supervisors
6.45. Signifcance is at the .000 level.
Finding 1.2: Malaysia’s male supervisors’ use of
intuition is lower than US male managers’. Mean scores
of Malaysia’s male supervisors 6.40; US male managers
7.8889. Signifcance is at the .000 level.
Research Question 2: What is the relationship between
sex and reported use of intuition in decision making?
Finding 2.1: Malaysia’s male supervisors’ use of
intuition is lower than US female business owners. Mean
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
318 Business Intelligence Journal July
scores of Malaysia’s male supervisors 6.40; US female
business owners 7.0476. Signifcance is at the .000 level.
Research Question 3: What is the relationship between
country of operation and use of intuition in decision
making?
Finding 3.1: Malaysia’s male supervisors’ use of
intuition is lower than US female managers’. Mean scores
of Malaysia’s male supervisors 6.40; US female managers
6.8333. Signifcance is at the .002 level.
Finding 3.2: Malaysia’s male supervisors’ use of
intuition is lower than US female executives’. Mean scores
of Malaysia’s male supervisors 6.40; US female executives
7.5200. Signifcance is at the .000 level.
Conclusion
The study discovered that United States managers
were less rational in their managerial decision making
than Malaysian managers in fve of fve categories. These
fndings can be quite signifcant to global business managers
intending to shift more of their activities to Asia in the near
future. Multinational corporations would have to provide
more data for their Malaysian managers in order to make
them more comfortable with their decision making tasks
as they tend to rely more on data based decision-making
techniques.
Business schools in the West may also need to redesign
their curriculum as more business managers feel more
comfortable with intuitive decision making techniques.
This will make business schools more relevant to what
obtains outside school walls and produce graduates that will
be more amenable to top executive position appointments.
The author would also like to recommend more
comparative studies in managerial decision-making so that
a clearer picture of intuitive managerial decision making
can emerge.
References
Aarnio, K., & Lindeman, M. (2005). Superstition,
education and thinking styles. Personality and
Individual DiVerences,39, 1227–1236.
Agor, W.H. (1984). Intuitive management: Integrating
left and right brain management skills. Englewood’s
Cliffs, NJ: Prentice Hall.
Alban Metcalfe B and West M (1991) "Women in
Management" in J Firth-Cozens and M West (eds)
Women at Work Milton Keynes: Open University
Press, 154-171.
Brenner, O.C., & Bromer, J. A. (1981) Sex stereotypes
and leaders behavior as measured by the agreement
scale for leadership behavior. Psychological
Reports, 48, 960-962.
Cappon, D. (1993). The Anatomy of Intuition:
What's your Intuition Quotient? Psychology
Today, Retrieved August 15, 2006 from http://
www.psychologytoday.com/articles/199305/the-
anatomy-intuition.
Clare, D. K. (1999) Workplace devaluation: Learning
from Experience Digital Library and archives (etd-
042699-162003).
Damen, L. (1987). Culture learning: The ffth
dimension on the language classroom. Reading,
MA: Addison-Wesley.
Dane, E., & PRATT M. G.(2007) Exploring intuition
and its role in managerial decision making Academy
of Management Review 32: 33–54.
Eagly, A. H., & Johnson, B. T., (1990). Gender and
leadership style: a meta-analysis. Psychological
Bulletin, 108, 233-256.
Fields A. F. (2003) A Study of Intuition in Decision-
Making using Organizational Engineering
Methodology. Dissertation Abstracts International,
66 (07), 2474. (UMI No. 3182016)
Green, E. & Cassell, C. (1996) Women Managers,
Gendered Cultural Processes and Organizational
Change Gender, Work & Organization, 3: 168–178.
doi: 10.1111/j.1468-0432.1996.tb00057.x
Hayes, R. H., Pisano, G. P., Upton, D. M., &
Wheelwright, S. C. (2004) Operations, Strategy,
and Technology: Pursuing the Competitive Edge
Somerset, New Jersey: Wiley.
Hofstede, G. (1980), Culture’s consequences:
International differences in work related values,
Thousand Oaks, CA: Sage Publications, Inc.
Hofstede, G. (2003) Geert Hofstede cultural dimensions
Itim International Retrieved August 10, 2006 fromhttp://www.geert-hofstede.com/hofstede_china.
shtml.
Kanter, R.M. (1977). Men and women of the
corporation New York: Basic Books.
Khatri, N., & Ng, H. A. (2000). The Role of Intuition
in Strategic Decision Making Human Relations.53:
57-86.
Lau, D.C. (1992). The Analects by Confucius
Angelfre.com Retrieved October 02, 2006 http://
www.angelfire.com/ego/nolimitz/confucius/
analects.html.
2011 319
Loden, M. (1985) Feminine leadership, or, How to
succeed in business without being one of the boys
New York: Times Books.
Lovell, J. (2001). The proverbs of a culture refect
much of its attitudes. Bigpond.net.au Retrieved
August 10, 2006 fromhttp://users.bigpond.net.au/
bwi28/pro/julie.html
Pacini, R., & Epstein, S. (1999). The relation of
rational and experiential information processing
styles to personality, basic beliefs, and the ratio-
bias phenomenon. Journal of Personality and Social
Psychology, 76, 972-987.
Prahlad, A. (2001) Reggae wisdom: Proverbs in
Jamaican music. Jackson, MS: University ' Press of
Mississippi
Thompson, B., and Borello, G. (1986a). "Construct
validity of the Myers-Briggs Type Indicator".
Educational and Psychological Measurement 46:
pp 745-752.
Wajcman, J. (1996), Desperately Seeking Differences:
Is Management Style Gendered? British Journal of
Industrial Relations, 34: 333–349.
Oluwabusuyi I. - Intuitive Managerial Decision Making in Malaysia and the United States
Isola Oluwabusuyi
Business Intelligence Journal - July, 2011 Vol.4 No.2
320 Business Intelligence Journal July
FRANCHISING AND ORGANIZATIONAL PERFORMANCE:
EMPIRICAL INVESTIGATION OF SELECTED FAST FOOD
RESTAURANTS IN NIGERIA
Olu Ojo
Department of Business Administration College of Management
and Social Sciences Osun State University
P. M. B. 2008, Okuku
Osun State, Nigeria
Email: [email protected]
I. A. Irefin
Technology Planning And Development Unit Obafemi Awolowo University
Ile-Ife, Osun State, Nigeria
Email: [email protected]
Abstract
This study tries to gain insight into the relationship between franchising and organizational performance using selected fast food
restaurants in Nigeria as case study. It specifcally investigates the effects of franchising types and franchising ownership on organizational
performance. This research work was undertaken in order to fll the existing gap in literature with regards to the effects of franchising
types and franchising ownership on organizational performance. Thus, this study will add to the existing level of understanding on
the subject of investigation and add to the existing literature on the subject especially in underdeveloped countries where such study
is not popular. The survey research design was employed. Primary data were used and they were collected through the administration
of question to our respondents. Data collected were analyzed using descriptive method and simple percentage. Three hypotheses were
advanced and tested with the aid of correlation coeffcient. The fndings show that there is positive relationship between franchising types
and organizational performance and that positive relationship also exists between franchising ownership and organizational performance.
The author concludes by saying that franchising as a business form is fast becoming the in thing in Nigerian business arena and it is
therefore a good investment opportunity for upcoming entrepreneurs. Key words: Franchising, business organizations, franchising
relationships, organizational performance, franchising types, and franchising ownership.
Franchising occupies a prominent position in
contemporary business world and increasingly provides a
common channel for frm growth, entrepreneurial wealth
creation, and national development. In a competitive
market each business organization must strive to sell its
products and/or services by creating a competitive edge
over its competitors via multi product development, price
variation and market segmentation.
The various franchising types have their cost and beneft
owing to product differentiation, creation and affliation
requirements and therefore, the most suitable type to be
chosen by various interested parties becomes critical.
Therefore people seek to fnd the most suitable and most
productive for their franchise business that will yield returns
and improve quality and effectiveness in a competitive
environment. The ownership of the franchise relationship
is also very important. Countries with centralized and rigid
organizational system like France and China will prefer
that ownership be given to the franchisee and royalties be
paid to the franchiser periodically and this could lead to
lower quality levels or higher prices in franchised outlets.
The other which is ownership with the agent, will also
increase contracting cost and thus costlier to write contracts
and enforce, than those with employed managers by the
franchiser. Therefore, this confict in the moderate, total and
even non ownership of the franchise by either the principal
or the agent has become a problem to be solved.
Based on the above research problems, this study is
expected to provide answers to the following questions:
(i) How does franchising types affect the performances of
2011 321
Ojo O., Irefn I. A. - Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants in Nigeria
Olu Ojo, I. A. Irefn
both the franchiser and franchisee? (ii) Does franchising
ownership model have effect on return on investment of
the franchiser and franchisee? And (iii) What is the effect
of franchising relationship on organizational performance?
In order to answer the above research questions, the
general objective of this research is to explore the extent
to which franchising relationship can affect organizational
performance. The specifc objectives of this study are
(i) To examine the effect of franchise types on franchise
performance within a highly competitive environment. (ii)
To ascertain ownership models of franchising relationship
towards organizational performance. And (iii) To examine
the effect of franchising relationship on organizations
performance.
Franchising has received a great deal of attention over
the past few years. Despite the plethora of research on
franchising, some important gaps exist in our understanding
of this important organizational form. Although some
studies provide a useful starting point for research, there
are other factors associated with success that are unique to
franchising that should be considered. These include the
effect of franchising types on the performances of both
the franchiser and franchisee, and the effect of franchising
ownership model on the return on investment of the
franchiser and franchisee?
A number of studies provide support for the assumption
of greater franchise success (for example, Castrogiovanni,
Justis, and Julian 1993; Justis, Castrogiovanni, and Chan
1992). However, other recent studies are critical of this
position and argue that franchises exhibit higher rates of
frm discontinuance and lower mean proftability than
independent businesses (Bates 1995). To date the evidence is
mixed. Because of these contradictory results, the question
of whether franchising type and franchising ownership
improves or worsens franchising performance is still worthy
of further research such as the one undertaken in this study.
Besides, most of the existing studies on franchising were
conducted in developed countries using archival data. This
means that there is a major gap in the relevant literature on
developing and underdeveloped countries in which Nigeria
rightly belongs to that has to be covered by research. This
research tries to fll this gap by studying the situation of
Nigerian fast food industry and providing more empirical
evidence on franchising performance.
Literature Review
A number of strategies are available to the entrepreneur
for the expansion of his venture. However, one of these
strategies that are commonly ignored by entrepreneurs
in the new venture expansion is franchising (Ojo, 2008).
Although, franchising is a remarkable organizational
form, in the common economic parlance, it is a hybrid
organizational form of business ownership (Ojo, 2009).
Different defnitions have been given to franchising by
different people over time. Hisrich, Peters and Shepherd
(2005) see franchise as a form of new entry that can
reduce the risk of downward loss for the franchise. They
see franchising as an alternative means by which an
entrepreneur may expand his or her business by having
others pay for the use of the name, process, service etc. It
can be used as a growth mechanism by the organization
(i.e. franchiser).
Franchising dates back to at least 1850’s; when Isaac
Singer, who made improvements to an existing model of a
sewing machine, wanted to increase the distribution of his
sewing machines. His efforts, though unsuccessful in the
long run, was among the frst franchising effort in the United
States. Slightly later, yet much more successful, example
of franchising was John Pemberton’s franchising of Coca-
cola. Early American examples include the telegraph
system, which was operated by various railroad companies
but controlled by western union, an exclusive agreement
between automobile manufactures and operators of local
dealership. Modern franchise came to prominence with the
rise of franchised-based food service establishments. This
trend started as early as 1919 with quick service restaurants
such as A&W Root Beer. In 1935, Howard Deering
Johnson teamed up with Reginald Sprague to establish
the frst modern restaurant franchise. The idea was to let
independent operators use the same name, food, supplies,
logo and even building design in exchange for a fee. The
growth in franchises picked up steam in the 1930”s when
such chains as Howard Johnson started franchising motels.
The 1950”s saw a boom of franchise chains in conjunction
with the development of Americans interstate highway
system. Fast food restaurant, motel chains exploded. In
regards to contemporary franchise chains, Mc Donald’s
is arguably the most successful worldwide with more
restaurant units than any other franchise network (Ike-
Okoh, 2006).
Business format franchising, a type of franchise in 19th
century, took off in 1850’s.
Business format franchising is the form of franchising
most commonly associated with the franchise concept
(Alon, 2004). A franchiser licenses an entire way of doing
business under a brand name. This variety of franchising
is prevalent in accounting services, auto accessories, auto
Business Intelligence Journal - July, 2011 Vol.4 No.2
322 Business Intelligence Journal July
rentals, campgrounds, cleaning systems, fast food, food
retailing, motels/hotels, real estate, and schools. Business
format franchising involves packaging a mode of business,
attracting a supply of capable and dedicated entrepreneurs,
selecting superior prospects, training them in the minute
details of the business operations, providing assistance in
setting up the business at specifc outlets, and maintaining
an ongoing business that is proftable for the franchisor and
the collective franchises. The relationship entails continued
provision of benefcial services such as advertising and
new product development by the franchises and continued
provision of royalties from the franchisees to the franchisors.
Business format franchising represents a complete
package that allows the franchisee to use the format
provided by the franchiser, while retaining independence
as a business. The franchiser typically sells the franchisee
a right to use his intellectual property in return for a lump
sum payment and an annual royalty fee based on sales
for a specifed period of time (Miller and Grossman,
1990). In addition, the franchisee usually agrees to adhere
to franchiser requirements for product mix, operating
procedures and site selection (Rubin, 1978). Business
format franchising is a popular example of a hybrid
organizational format that incorporates elements of both
markets and hierarchies (Williamson, 1991). It is a hybrid
alternative since the franchiser and franchisee both retain a
degree of ownership and authority over the use of the trade
name, operating procedures and the location of outlets and
contracts with independent entrepreneurs to operate the
units (Child, 1987).
Franchising as a business seems to have achieved more
respectability in the fast food industry than ever before as
some franchised brands command noticeable lead today in
terms of awareness, product strength and value. The growth
of the fast food industry in Nigeria has been energetic with
Tantalizer, Captain Cook, Sweet Sensation and Mr. Biggs
are among the most recognized franchise chains featuring
more up from market local menu. It is important to note that
the Nigerian fast food market is still in its infant stage, far
from maturity.
What May be Bought in A Franchise
According to Hisrich et al. (2005), what you may buy in
a franchise as a franchisee includes:
• A product or service with established market ad
favorable market,
• A patented formula or design,
• Trade name or trade marks
• Financial management system for controlling the
fnancial revenue
• Managerial advice from experts in the feld
• Economies of scale for advertising and purchasing
• Head offce services
• A tested business concept
Methods and Materials
This section describes the research methodology used
in the study. Survey research design was utilized in this
study. The theoretical population of study consists of all the
franchised fast food restaurants in Lagos State. The choice
of Lagos State stems from the fact that it is the commercial
centre of Nigeria and that overwhelming majority of
franchised fast food restaurants were concentrated in Lagos
State. The offce in charge of regulating franchise business
in Lagos State reveals that there are about twenty fve
(25) franchised outlets existing in Lagos State. Therefore
these franchised outlets were the theoretical population.
Our study population consists of the entire staff of 25
registered franchised fast food restaurants in Lagos State.
Both probability and nonprobability sampling methods
were used. Stratifed probability sampling method was
used to classify the staff of the franchised restaurants
into three groups: lower level management, middle level
management, and top level management. Combination
of convenience and judgmental nonpropability sampling
methods were used to select our sample elements. These
methods allow a large number of respondents to participate
in the research over a short period of time (Ojo, 2003).
Primary method of data collection was used in this study.
The data consists of a number of items in structured
questionnaire that was administered to the respondents. We
also utilized 5-point Likert scale in our questionnaire. The
decision to structure the questionnaire is predicated on the
need to reduce variability in the meanings possessed by the
questions as a way of ensuring comparability of responses.
The questionnaire is titled “Franchising and Organizational
Performance Questionnaire.” Two hundred and seventy-
fve (275) respondents were selected for the study. However,
only 230 of them flled their questionnaire and returned it
and were used for fnal analysis in the study. This means
that the return rate of completed questionnaire was 83.6%.
To ensure the validity and reliability of the questionnaire
used for the study, even numbers of experts were consulted
to look at the questionnaire items in relations to its ability
to achieve the research objectives, level of coverage,
2011 323
According to the table 3 above, 4.3% of the respondents
strongly disagree with the statement. 17.4% of the
respondents disagree with the statement, and 4.3% of the
respondents are undecided about the statement. Of the
remaining 73.9% of the respondents, 43.5% of them agree
with the statement and 30.4% of them strongly agree with
the statement. This indicates that majority of the respondents
are either agree of strongly agree with the statement.
comprehensibility, logicality, and suitability for prospective
respondents. Data collected from the questionnaire were
analyzed with the aid of descriptive statistical techniques
such as total score and simple percentage, while inferential
statistics such as correlation coeffcients was used to proof
the level of signifcance in testing stated hypotheses.
Data Presentation, Analysis and Results
All the items in the questionnaire were analyzed.
However, only the key ones featured in this section.
Table 1: Age Distribution of Respondents
Age of
Respondents
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid Below 20 Years 20 8.7 8.7 8.7
21- 40 Years 180 78.3 78.3 87.0
41 - 60 Years 30 13.0 13.0 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
From the above table it can be observed that 8.7% of the
respondents fell within the age bracket of less than 20years.
Overwhelming majority of the respondents (78.3%) are in
the age bracket of 21 to 40 years. The remaining 13.0%
of the respondents are within the age bracket of 41 – 60
years. This clearly shows that the larger percentages of the
respondents are young persons in the middle of their career.
Table 2: Respondents’ Position in the Organization
Respondents’ Position in the
Organization
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid Valid Lower Level
Management
40 17.4 17.4 17.4
Middle Level
Management
180 78.3 78.3 95.7
Top Level
Management
10 4.3 4.3 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Table two above presents the position of the respondents
on the organizational hierarchy. The table reveals that 17.4
of the respondents are at the lower level management. The
bulk of the respondents (78.3%) are in the middle level
management, while the remaining 4.3% of the respondents
are in the top level rung of the organization. This shows the
highest responses came from the middle level managers in
the organization.
Table 3: This type of Franchising Relationship Increases profts
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 10 4.3 4.3 4.3
D 40 17.4 17.4 21.7
U 10 4.3 4.3 26.0
A 100 43.5 43.5 69.5
SA 70 30.4 30.4 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Table 4: Type of Franchising has Effect on Money Invested in
the Business
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 30 13.0 13.0 13.0
D 20 8.7 8.7 21.7
U 20 8.7 8.7 30.4
A 100 43.5 43.5 73.9
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Ojo O., Irefn I. A. - Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants in Nigeria
Olu Ojo, I. A. Irefn
Business Intelligence Journal - July, 2011 Vol.4 No.2
324 Business Intelligence Journal July
Table 4 above reveals that 13% of the respondents
strongly disagree that the type of franchising has effect on
money invested in the business, 8.7% of the respondents
disagree with the statement, while another 8.7% of the
respondents are undecided about the statement. Majority
of the respondents (43.5%) agree with the statement while
additional 26.1% of the respondents strongly agree with
the statement. The inference that can be drawn here is that
the type of franchising has effect on money invested in the
business if we go by the respondents’ opinion.
Table 5: The Ownership by Franchisee Agents Increases Return
on Capital Employed
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 10 4.3 4.3 4.3
D 20 8.7 8.7 13.0
U 40 17.4 17.4 30.4
A 100 43.5 43.5 73.9
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
In table 5 above, 4.3% of the respondents strongly
disagree that ownership by franchise agent’s increases return
on capital employed. A total of 8.7% of the respondents
disagree with the statement. 17.4% of the respondents are
undecided about the statement. 43.5% of the respondents
agree with the statement. The remaining 26.1% strongly
agree with the statement. Based on the above data, we
can confdently say that ownership by franchise agents
increases return on capital employed.
According to the table 6 above, 4.3% of the respondents
strongly disagree that proft is maximized when higher
ownership is vested with the franchisee. In addition,
another 4.3% of the respondents disagree with the
statement. 13% of the respondents are undecided about the
statement. However, overwhelming majority (52.2%) of the
respondents agree with the statement while the remaining
26.1% of the remaining respondents strongly agree with the
statement. Thus, we can conclude that proft is maximized
when higher ownership is with the franchisee.
Table 7: Franchising Leads to Positive Corporate Performance
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 30 13.0 13.0 13.0
D 20 8.7 8.7 21.7
U 20 8.7 8.7 30.4
A 100 43.5 43.5 73.9
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
Source: Field Survey, 2009.
Table 7 above presents the respondents feedback to the
effect of franchising on corporate performance. 13.0% of
the respondents strongly disagree that franchising leads
to corporate performance, while another 8.7% of the
respondents disagree with the statement. Another 8.7% of
the respondents are undecided about the statement. Among
the remaining respondents, 43.5% of them agree with the
statement while 26.1% of them strongly agree. Thus, we
can inferred that franchising leads to positive corporate
performance.
Testing of Hypothesis
Our three hypotheses were tested using correlation
coeffcient.
Table 6: Proft is Maximized when Higher Ownership is with the
Franchisee
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
Valid SD 10 4.3 4.3 4.3
D 10 4.3 4.3 8.6
U 30 13.0 13.0 21.6
A 120 52.2 52.2 73.8
Source: Field Survey, 2009.
F
r
e
q
u
e
n
c
y
P
e
r
c
e
n
t
V
a
l
i
d
P
e
r
c
e
n
t
C
u
m
u
l
a
t
i
v
e
P
e
r
c
e
n
t
SA 60 26.1 26.1 100.0
Total 230 100.0 100.0
2011 325
Table 10: Correlation Between Franchising Relationship and
Organizational Performance
Types Performance
Franchising
Relationship
Pearson
Correlation
1 .535(**)
Sig. (2-tailed) .009
N 230 230
Performance Pearson
Correlation
.535(**) 1
Sig. (2-tailed) .009
N 230 230
Table 8: Correlation Between Franchising Types and
Organizational Performance
Types Performance
Types Pearson
Correlation
1 .363(**)
Sig. (2-tailed) .089
N 230 230
Performance Pearson
Correlation
.363(**) 1
Sig. (2-tailed) .089
N 230 230
(**) Correlation is signifcant at the 0.01 level (2-tailed).
The analysis above shows that there is a positive
correlation between the two variables: franchising types
and organizational performance [r = .363(**); N = 230],
meaning that the relationship between them is positive.
Hypothesis 2
H
1
: The franchisee ownership has effect on organizational
performance.
Here franchisee ownership was correlated against
corporate performance. The outcome is presented in table
9 below.
Table 9: Correlation Between Franchisee Ownership and
Organizational Performance.
Types Performance
Owners hip Pearson
Correlation
1 .570(**)
Sig. (2-tailed) .005
N 230 230
Performance Pearson
Correlation
.570(**) 1
Sig. (2-tailed) .005
N 230 230
(**) Correlation is signifcant at the 0.01 level (2-tailed).
The analysis above shows that there is a positive
correlation between the two variables, franchisee ownership
and organizational performance [r = .570(**); N = 23],
meaning that the relationship is positively related.
Hypothesis 3
H
1
: Franchising relationship has effect on organizational
performance.
In order to test the signifcant relationship between
franchising relationship and organizational performance
the Pearson product moment correlation was used. Our
data were combined and analyzed to check the relationship
and strength of the relationship between franchising and
organizational performance. The analysis is as presented
below:
Hypothesis 1
H
1
: There is a signifcant relationship between franchise
types and organizational performance.
(**) Correlation is signifcant at the 0.01 level (2-tailed).
From the above table, the analysis shows that there is a
positive correlation between the two variables, franchising
and organizational performance [r = .535(**); N = 230),
meaning that franchising and organizational performance
are positively correlated.
In all the three instances, we accept our hypotheses.
Discussion of Findings
As said earlier, this study focused on a research into
franchising and how it affects the profts of an organization
operating it. Based on analyzed data, the fndings in this
study include the followings:
i. The larger percentage (78.3) of the respondents
are young persons in the middle of their career. Their age
bracket is 21to 40 years.
ii. The bulk of the respondents (78.3%) are in the
middle level management of the organization..
Ojo O., Irefn I. A. - Franchising and Organizational Performance: Empirical Investigation of Selected Fast Food Restaurants in Nigeria
Olu Ojo, I. A. Irefn
Business Intelligence Journal - July, 2011 Vol.4 No.2
326 Business Intelligence Journal July
iii. Majority of the respondents either agree (43.5%)
or strongly agree (30.4%) with the statement that business
format type of franchising relationship has effect on money
invested in the business and also increases profts.
iv. In addition, 43.5% of the respondents agree while
another 26.1% of the respondents strongly agree that
ownership by franchisee agent’s increases return on capital
employed.
v. It is also discovered that 52.2% of the respondents
agree and 26.1% of the respondents strongly agree that
proft is maximized when higher ownership is with the
franchisee.
vi. Besides, 43.5% of the respondents agree and
another 26.1% of the respondents strongly agree that
franchising leads to positive corporate performance.
vii. Finally, all the three hypotheses tested reveal
positive correlation between business format type of
franchising and organizational performance, ownership
type and organizational performance, and franchising
relationship and organizational performance.
Conclusion
Recent trends in the business environment have brought
about innovative ways in which frms can take the lead
in their industry even in the face of great competitors.
The conclusion that is drawn from this study is that
organizations that are currently operating franchising
as a business arrangement are making large profts and
are gaining even stronger brand names through it. By
maintaining the same standards operated in the parent
company, in all their franchises they stand a greater chance
of maintaining their customer loyalties. Franchising is fast
becoming the thing in the business arena in Nigeria as more
and more companies are being drawn to it as a good strategy
for distributing their products and raising better awareness
for their companies.
References
Alon, I. (2004). Global Franchise and Development
in Emerging and Transitioning Market”, Journal of
Macro-Marketing, 24 (2), pp. 156-167.
Bates, T. (1995). Analysis of Survival Rates among
Franchise and Independent Small Business
Startups. Journal of Small Business Management
33(2), pp. 26-36.
Castrogiovanni, G. J., R. T. Justis, and S. D. Julian
(1993). Franchise Failure Rates: An Assessment of
Magnitude and Infuencing Factor. Journal of Small
Business Management 16: pp.105-114.
Child, J. (1987). Information Technology, Organization
and the Response to Strategic Challenges, California
Management Review, 30, pp. 35-50.
Hisrich, R. D., M. P. Peters, D. A. Shepherd (2005).
Entrepreneurship, New York: McGraw-Hill/Irwin.
Ike-Okoh, C. (2006). Franchising: Forwarding-
Looking Form of Entrepreneurship, Businessday
March 28, pp. 2-3.
Justis, R. T., G. J. Castrogiovanni, and P. Chan
(1992). Examination of Franchise Failure Rrates.
Proceedings of the Society of Franchising.
Minneapolis, MN: University of St. Thomas.
Miller, A., and T. Grossman (1990). Business Law,
Glenview: Scott Foresman:
Ojo, O. (2003). Fundamentals of Research Methods,
Lagos: Standard Publications.
Ojo, O. (2008). Franchising: Hybrid Organizational
Arrangement for Firm Growth and National
Development, Lex et Scientia International Journal,
Nr.XV, Vol. 2, pp. 113-120.
Ojo, O. (2009). Fundamentals of Business
Management, Lagos: Standard Publications.
Rubin, P. H. (1978). The Theory of The Firm and The
Structure of The Franchise Contract, Journal of
Law and Economics, 21, pp. 223-234.
Welsh, D.H.B., Alon, I. and Falbe, C.M. (2006), “An
Examination of International Retail Franchising
In Emerging Markets” Journal of small Business
management, 44 (1), pp. 130-149.
Williamson, O. E. (1991). Comparative Economic
Organization: The Analysis of Discrete Structural
Alternatives, Administrative Science Quarterly, 36,
pp. 269-296.
2011 327
RELATIONSHIP BETWEEN REWARDS AND EMPLOYEE’S
MOTIVATION IN THE NON-PROFIT ORGANIZATIONS OF
PAKISTAN
Nadia Sajjad Hafiza, Syed Sohaib Shah, Humera Jamsheed
MS Scholar, Department of Management Sciences, COMSATS Institute of Information Technology,
Abbottabad, Pakistan.
Khalid Zaman (Corresponding author)
MS Scholar, Department of Management Sciences, COMSATS Institute of Information Technology,
Abbottabad, Pakistan.
Email: [email protected], [email protected]
Abstract
This study empirically examines the relationship between rewards and employee’s motivation in the non-proft organizations
of Pakistan. Employees of three organizations (PERRA, World Vision and SUNGI Development Foundation) working in Khyber
Pakhtonkhuwa province of Pakistan were taken as sample of the study. Self designed questionnaire was used for data collection. 125
questionnaires were distributed and 107 were returned, hence the response rate was 85.6%. The data was analyzed using the techniques
of rank correlation coeffcient and multiple regression analysis. All the fndings were tested at 0.01 and 0.05 level of signifcance. The
result concludes that there is a direct relationship between extrinsic rewards and the employee’s motivation. However, intrinsic rewards
found an insignifcant impact on employee motivation. Key words: Rewards, Employee motivation, Non-Proft Organization (NGO),
Correlation, Regression, KPK province, Pakistan.
In today’s competitive business environment companies
are facing many challenges and among those challenges
acquiring right workforce and retaining it, is of utmost
importance. Nowadays, human asset is considered to be
the most important asset of any organization. In order to
get the effcient and effective result from human resource,
employee motivation is necessary.
Employee will give their maximum when they have a
feeling or trust that their efforts will be rewarded by the
management. There are many factors that effect employee
performance like working conditions, worker and employer
relationship, training and development opportunities, job
security, and company’s over all policies and procedures
for rewarding employees, etc. Among all those factors
which affect employee performance, motivation that comes
with rewards is of utmost importance. Motivation is an
accumulation of different processes which infuence and
direct our behavior to achieve some specifc goal (Baron,
1983).
This study will examine the employee’s motivation of an
organization with the rewards (both extrinsic and intrinsic)
given to them. For the study, particularly employees
of non-proft organizations of Pakistan will be selected,
and as a sample employees would be taken from three
non-proft organizations of Abbottabad, KPK, Pakistan,
namely, PERRA, World Vision and SUNGI Development
Foundation.
Rewards can be extrinsic or intrinsic, extrinsic rewards
are tangible rewards and these rewards are external to the job
or task performed by the employee. External rewards can
be in terms of salary/pay, incentives, bonuses, promotions,
job security, etc. Intrinsic rewards are intangible rewards or
psychological rewards like appreciation, meeting the new
challenges, positive and caring attitude from employer, and
job rotation after attaining the goal. Frey (1997) argues that
once pay exceeds a subsistence level, intrinsic factors are
stronger motivators, and staff motivation requires intrinsic
rewards such as satisfaction at doing a good job and a sense
of doing something worthwhile.
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
328 Business Intelligence Journal July
Desired performance can only be achieved effciently
and effectively, if employee gets a sense of mutual gain
of organization as well as of himself, with the attainment
of that defned target or goal. An organization must
carefully set the reward system to evaluate the employee’s
performance at all levels and then rewarding them whether
visible pay for performance or invisible satisfaction. The
concept of performance management has given a reward
system which contains; needs and goals alignment between
organization and employees, rewarding employees both
extrinsically and intrinsically. The system also suggests
where training and development is needed by the employee
in order to complete the defned goals. This training or
development need assessment of employee gives them an
intrinsic motivation.
The objective of this study is to fnd out the relationship
between rewards and employees’ performance in non-proft
organization in Pakistan. More specifc objectives are to
fnd out:
• The effect of intrinsic rewards on employee’s
performance.
• The effect of extrinsic rewards on employees'
performance and
• The effect of both intrinsic and extrinsic rewards on
employees' performance
Based on the above objectives, the present study seeks
to test the following hypothesis:
H1: There is a direct relationship between intrinsic rewards
and employee motivation.
H2: There is a direct relationship between extrinsic rewards
and employee motivation.
The signifcance of the study would be an idea about
relative importance of extrinsic and intrinsic rewards.
Management can get a better idea while preparing its reward
system that what kind of reward would be given the most
importance and at what stage through such a type of study.
This paper is organized as follows: after introduction
in section 1, literature review is carried out in section 2.
Research framework and methodology is mentioned in
section 3. Result and discussion is provided in section 4.
Final section concludes the study.
Literature Review
Human resource is one of the most important resources of
gaining competitive advantage over competitors for a frm.
And this resource can be retained and optimally utilized
through motivating it using different techniques among
which reward is of signifcance importance. Carraher et al
(2006) advocates that there should be an effective reward
system to retain the high performers in the organization
and reward should be related to their productivity. A lot of
work has been done on evaluating the relationship between
rewards and employee motivation and there exist a large
number of studies in the literature describing impact of
reward on employee motivation. In order to maximize the
performance of the employees organization must make
such policies and procedures and formulate such reward
system under those policies and procedures which increase
employee satisfaction and motivation. Bishop (1987)
suggested that pay is directly related with productivity and
reward system depends upon the size of an organization.
Organizations in today’s competitive environment want to
determine the reasonable balance between employee loyalty
and commitment, and performance of the organization.
The existing literature on individuals’ innovative
performance reveals a wide array of individual and
organizational antecedent factors. Among many individual
antecedents that infuence employees’ innovative
performance are attitudes (Williams, 2004), cognitive styles
(Scott and Bruce, 1994), personality and demographic
characteristics such as age, education background, and
prior R&D experience (Roberts, 1991 and Rothwell, 1992).
In terms of organizational antecedents, expenditure on
R&D (Hadijimanolis, 2000), cooperation with external
technology provider, leader’s infuence (Hage and Dewar,
1973), and reward system (Eisenberger and Cameron,
1996; Janssen, 2000; Mumford, 2000) are commonly cited
as factors that affect individuals’ innovative performance.
Effcient reward system can be a good motivator but an
ineffcient reward system can lead to demotivation of the
employees. Reio and Callahon (2004) concludes that both
intrinsic and extrinsic rewards motivates the employee
resulted in higher productivity.
Most of the organizations have gained the immense
progress by fully complying with their business strategy
through a well balanced reward and recognition programs
for employee. Deeprose (1994) argued that the motivation of
employees and their productivity can be enhanced through
2011 329
providing them effective recognition which ultimately
results in improved performance of organizations. The
entire success of an organization is based on how an
organization keeps its employees motivated and in what
way they evaluate the performance of employees for job
compensation.
Sometimes management pays more attention to extrinsic
rewards but intrinsic rewards are equally important in
employee motivation. Intangible or psychological rewards
like appreciation and recognition plays a vital role in
motivating employee and increasing his performance.
Andrew (2004) concludes that commitment of employees is
based on rewards and recognition. Lawler (2003) argued that
prosperity and survival of the organizations is determined
through how they treat their human resource. Ajila and
Abiola (2004) examine that intrinsic rewards are rewards
within the job itself like satisfaction from completing a
task successfully, appreciation from the boss, autonomy,
etc, while extrinsic rewards are tangible rewards like pay,
bonuses, fringe benefts, promotions, etc. Filipkowski
and Johnson (2008) examined the relationships between
measures of job insecurity, organizational commitment,
turnover, absenteeism, and worker performance within a
manufacturer. A positive relationship was found between job
insecurity and intentions to turnover, and a small negative
correlation was found between measures of job insecurity
and organizational commitment. Tosti and Herbst (2009)
discuss about behavior systems approach which can be
used to achieve a customer centered organization through
examples and reports from consultation cases. Johnson et
al (2010) establish the effects of presenting organizational
information through implicit and explicit rules on sales-
related target behaviors in a retail setting. Results indicated
that when organizational information was presented in a
specifc form, productivity was increased and maintained
longer than when presented in other forms.
Research Design and Methodology
As this study examines the impact of extrinsic and
intrinsic rewards on employee motivation, the employees
of the non-proft organizations working in KPK has been
taken as population. Employee motivation is taken as
dependent variable and extrinsic and intrinsic rewards are
taken as independent variable. The framework of the study
is given in Figure 1, 2 and 3 respectively.
Employee
Motivation
Challenging
Tasks
Recognition and
Appreciation
Empowerment
and Autonomy
Figure 1: Intrinsic Rewards and Employee´s Motivation
Source: Self Construct
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
330 Business Intelligence Journal July
Employee
Motivation
Pay/Salary
Fringe Benefits
Bonuses
Promotions
Figure 2: Extrinsic Rewards and Employee’s Motivation
Source: Self Construct
Sampling
Using convenience method of sampling, 125
questionnaires were distributed among the employees of
three selected non-proft organizations namely, PERRA,
World Vision and SUNGI working in Abbottabad. 107
questionnaires were returned, so the response rate was
85.6%.
Data Collection
Self Designed questionnaire has been developed
for data collection. Self-designed questionnaire was
divided into two parts; one containing socio-demographic
questions and the second part containing questions related
to variables that are extrinsic and intrinsic rewards and
employee motivation. Parameters for measuring extrinsic
rewards are pay, bonuses, fringe benefts, and promotions
while parameters for measuring intrinsic rewards are
empowerment and autonomy, recognition and appreciation,
and challenging tasks. Five point Likert Scale ranging
from 1 (strongly disagree) to 5 (strongly agree) was used
to measure responses. The respondents’ scores for each
construct were obtained by summing across all the item
scores of the individual variables. The Cronbach’s Alpha
reliability coeffcients for the sample are given in Table 1.
Intrinsic Rewards
Extrinsic Rewards
Employees’ Motivation
Figure 3: Research Framework
Table 1: Cronbach’s Alpha Reliability Coeffcients
Items Cronbach’s Alfa
Extrinsic Rewards 0.83
Intrinsic Rewards 0.90
Results and Discussion
Frequency Distribution
Table 2 shows the demographic information of the
respondents. Most of the respondents are falling in the age
group of 26-30 year with the percentage of 50% and then 31-
35 years of age with 21%. The demographic characteristics
also show a gender division of the respondents, majority
of the respondents’ are males, i.e. 80% representing a
bigger part of the sample group. However, 20% percent
respondents are females.
2011 331
Table 2: Frequency Distribution of Demographic Variables
Variables Frequency Percentage
Age
21-25 10 9.3
26-30 55 51.4
31-35 25 23.3
36-40 09 8.4
41-45 05 4.6
46-50 02 1.8
51-55 01 1.2
Gender
Males 77 71.9
Females 30 28.1
Education
Under graduate 03 2.8
Graduate 42 39.2
Masters 61 57.0
Others 01 1.0
Descriptive Statistics
Table 3 explains that majority of the responses regarding
pay, fringe benefts, appreciation and challenging tasks
were neutral i.e., mean value is less than 3.5. The responses
of promotions, bonuses, empowerment and motivation
show that employees consider this factor slightly more
important than other factors as mean value is greater than
3.5. Standard deviation of challenging tasks and pay shows
that these variables have extensive responses than its mean
as value indicates 1.33 and 1.24 respectively. One of the
reasons for this deviation could be the selection of three
different organizations in the sample having different pay
packages.
Variables Mean Std. Deviation
Pay (PAY) 3.2857 1.2411
Fringe Benefts (FB) 3.3571 .8934
Promotions (PRO) 3.5714 .7086
Bonuses (BON) 3.3714 .8456
Empowerment (EMP) 4.0500 .7668
Appreciation (APP) 2.9464 .8511
Challenging Task (CT) 2.9089 1.3373
Motivation (MOT) 3.9857 .4761
Table 3: Descriptive Statistics
Correlation Analysis
Spearman’s Correlation was performed to study the size
and magnitude of the relationship between the variables.
The relationship between extrinsic, intrinsic and employee
performance are shown in the Table 4.
Table 4 shows that all extrinsic rewards i.e., pay,
fringe benefts, promotions, bonuses are positive and
signifcant related with employee motivation, whereas
intrinsic rewards i.e., empowerment appreciation and
challenging task have a negative but insignifcant impact
with employee motivation. Correlation coeffcient between
fringe benefts and motivation (0.319) is the highest among
all the variables and is signifcant at 99%. Pay (0.281) and
promotions (0.254) are also signifcant at 99%. Bonuses
have a weakest correlation among extrinsic rewards but
it is signifcant at 95%. Among intrinsic rewards all are
insignifcant relationship with employee motivation.
V
a
r
i
a
b
l
e
s
M
O
T
P
A
Y
F
B
P
R
O
B
O
N
E
M
P
A
P
P
C
H
A
MOT 1.000
PAY .281** 1.000
FB .319** .342** 1.000
PRO .254** .328** .317** 1.000
BON .190* .076 .423** .330** 1.000
EMP -.042 .070 -.114 .074 .240* 1.000
APP -.019 -.179 -.172 .155 .116 .451** 1.000
CT -.023 -.250* -.289** -.166 .100 .683** .829** 1.000
Table 4: Correlation Matrix
** Correlation is signifcant at the .01 level.
* Correlation is signifcant at the .05 level.
Incremental Regression
The incremental regression is performed by removing
individual independent variables from the model and by
checking the effect on the value of R-squared. Among all
the variables removed, fringe beneft has altered the value
of R-squared to a highest degree i.e., 8.4% decreases in the
portion of the dependent variable explained by independent
variables as the value for the R-squared changes from
54.2% to 45.8%. This importance is also highlighted in the
regression result as the value of coeffcient of the variable
is highest among all the variables in their three models
respectively. The result is presented in Table 5.
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
332 Business Intelligence Journal July
Hypothesis Testing
H1: There is a direct relationship between intrinsic rewards
and employee motivation.
The results of correlation matrix rejects the hypothesis as
it gives a negative and weaker relationship of empowerment
(-0.042), appreciation (-0.019) and challenging tasks
(-0.023) with employee motivation. Intrinsic rewards are
insignifcant relationship with employee motivation.
Incremental regression also rejects this hypothesis, which
shows that empowerment and appreciation has a negative
impact on employee motivation. And only empowerment
is signifcant among intrinsic variables. Chris and Awonusi
(2004) supports that extrinsic rewards have a signifcant
impact on employee motivation while intrinsic rewards
don’t have any signifcant impact on employee motivation.
Table 5: Incremental Regression Dependent Variable: Employee Motivation
Variables OLS1 OLS2 OLS3 OLS4 OLS5 OLS6 OLS7 OLS8
Pay .145 _ .240*** .188 .131 .077 .140 .108
FB .347* .387* _ .364* .380* .374* .355* .346*
PRO .203 .236*** .235*** _ .220*** .180 .154 .134
BON .097 .077 .214*** .129 _ .039 .076 .060
EMP -.286** -.246*** -.324** -.268** -.263** _ -.272** -.205
APP -.217 -.205 -.263 -.071 -.184 -.163 _ .069
CH .377 .316 .374 .233 .337 .154 .176 _
R square .542 .528 .458 .516 .535 .494 .533 .519
F-value 3.680* 4.056* 2.786* 3.817* 4.221* 3.381* 4.162* 3.877*
D-W 1.773 1.799 1.611 1.753 1.789 1.716 1.740 1.712
*, ** and *** indicates signifcance at 0.01, 0.05 and 0.09 percent level.
The result further shows goodness of ft in the
incremental regression after removing fringe benefts in
Table 6. This decrease in the value of the R-squared shows
the importance of fringe benefts in the model.
Table 6: Results of Incremental Regression removing Fringe
Benefts
Models Values
R-squared (original) 0.542
R-squared (after the removal) 0.458
H2: There is a direct relationship between extrinsic rewards
and employee motivation.
The results of correlation matrix have supported the
hypothesis that there exist a positive/direct relationship
between extrinsic rewards and employee motivation.
Incremental regression analysis also confrms the
hypothesis, as the most signifcant variable is fringe beneft
which is an extrinsic reward.
Conclusion
Human resource is considered to be the most important
resource of an organization to remain competitive in today’s
competitive business world. Acquiring the right workforce
and then retaining that force is one of the challenges faced
by organizations and their management. The results from
this study reveal that there is a signifcant and positive
relationship between extrinsic rewards and employee
motivation but it has been observed that organizations are
not offering right amount of fnancial rewards (extrinsic
rewards) to their employees in this sector. Pay is a signifcant
factor which affects employee motivation but the results
moderately supports the hypothesis due to difference
between the pay packages of three different organizations.
Fringe benefts are very important in motivating employees
according to this study, so organizations must have to
provide all the essential fringe benefts to their employees,
it also increase their job effciency.
2011 333
On the other hand, intrinsic rewards have a weaker
impact on employee motivation. Empowerment has
negative effect on employee motivation, due to the lack
of trust between employee and his/her boss through which
employee thinks that his/her boss has over burden him
instead of thinking himself empowered. There is an indirect
relationship between appreciation and employee motivation
as employees of the organizations are not satisfed with their
pay packages. So in the absence of extrinsic rewards which
is the basic source of motivation for employee, intrinsic
rewards like recognition, appreciation and empowerment
is of little importance. Pay is potentially powerful tool
to employee motivation so the employees can only be
intrinsically motivated to perform an activity when they are
fully satisfed with the pay they are getting.
There are certain limitations of the study which can
be taken into account for further studies in the future, like
sample size was too small and only organizations working in
Abbottabad were considered. Another important limitation
was that the pay structure of the three organizations was
different from one another, so responses regarding pay
variable were deviating, generating slightly signifcant
impact of pay on employee motivation. These limitations
can be avoided in the future studies carried out in this feld,
and a more clear picture can be obtained regarding impact
of extrinsic and intrinsic rewards on employee motivation.
References
Ajila, C and Abiola, A. (2004). Infuence of Rewards
on Workers Performance in an Organization,
Journal of Social Science, 8(1), pp.7-12
Andrew, D and Kent, R. (2007). ‘The impact of
perceived leadership behaviors on satisfaction,
commitment, and motivation: An expansion
of the multidimensional model of leadership’,
International Journal of Coaching Science, 1(1), p
35-56.
Baron, R. A. (1983). Behavior in organizations, p. 123,
New York: Allyn & Bacon, Inc.
Bishop, J. (1987). The recognition & Reward
of Employee Performance, Journal of Labor
Economics Vol. 5, No. 4 Part 2: The New Economics
of Personnel pp. S36-S56.
Carraher, R, Gibson, A. & Buckley R (2006).
.Compensation in the Baltic and the USA, Baltic
Journal of Management Vol. 1, pp 7-23.
Chris A., Awonusi A. (2004). Infuence of Rewards on
Workers Performance in a Organization, Journal of
Social Science, 8(1): 7-12
Collins, J. (1995). Motivating Your Organization, 1st
edition. McGraw-Hill International, p 187.
Deeprose, D. (1994). How to recognize and reward
employees. New York: AMACOM.
Eisenberger, R., and Cameron, J. (1996). Detrimental
Effects of Reward, American Psychologist, Vol.51,
No.11, pp.1153-1166.
Filipkowski, M. and Johnson, C. M. (2008).
Comparisons of Performance and Job Insecurity
in Union and Nonunion Sites of a Manufacturing
Company, Journal of Organizational Behavior
Management, Vol.28, No.4, pp.218 – 237.
Frey. B. (1997). On the Relationship between Intrinsic
and Extrinsic Work Motivation. International
Journal of Industrial Organization, 15, p 427 – 439.
Gaertner, K. & Gaertner, G. (1985). Performance-
contingent Pay for Federal Managers.
Administration & Society, 17, 1, p 7-20.
Hadjimanolis, A. (2000). An Investigation of
Innovation Antecedents in Small Firms in the
Context of a Small Developing Country, R&D
Management, Vol.30, No.3, pp.235-245.
Hage, J., and Dewar, R. (1973). Elite Values Versus
Organizational Structure in Predicting Innovation,
Administrative Science Quarterly, Vol.18, No.3,
pp.279-290.
Janssen, O. (2000). Job Demands, Perceptions of Effort-
Reward Fairness and Innovative Work Behavior,
Journal of Occupational and Organizational
Psychology, Vol.73, No.3, pp.287-302.
Johnson, R.A., Houmanfar, R. and Smith, G.S.
(2010). The Effect of Implicit and Explicit Rules
on Customer Greeting and Productivity in a Retail
Organization, Journal of Organizational Behavior
Management, Vol.30, No,1, pp38- 48
Lawler, E. E. (2003). Treat people right. San Francisco:
Jossey-Bass Inc. McGraw-Hill Irwin.
Mumford, M.D. (2000). Managing Creative People:
Strategies and Tactics for Innovation, Human
Resource Management Review, Vol.10, No.3,
pp.313-351.
Reio, G, T. & Callahon, J. L. (2004). Affect, Curiosity,
and socialization-related Learning; a path analysis
of antecedents to job performance, Journal of
Business and Psychology, Vol.19, pp3-22.
Hafza N. S., Shah S. S., Jamsheed H., Zaman K. - Relationship Between Rewards and Employee’s Motivation in the Non-Proft Organizations of Pakistan
Nadia Sajjad Hafza, Syed Sohaib Shah, Humera Jamsheed, Khalid Zaman
Business Intelligence Journal - July, 2011 Vol.4 No.2
334 Business Intelligence Journal July
Roberts, E.B. (1991). Entrepreneurs in High
Technology: Lessons from MIT and Beyond, New
York: Oxford University Press.
Rothwell, R. (1992). Successful Industrial Innovation:
Critical Factors for the 1990s, R&D Management,
Vol.22, No.3, pp.221-239.
Scott, S.G. and Bruce, R.A. (1994). Determinants of
Innovative Behavior: A Path Model of Individual
Innovation in the Workplace, Academy of
Management Journal, Vol.37, No.3, pp.580-599.
Tosti, D. and Herbst, S.A. (2009). Organizational
Performance and Customer Value, Journal of
Organizational Behavior Management, Vol.29,
No.3, pp.294- 314.
Williams, S.D. (2004). Personality, Attitude, and
Leader Infuences on Divergent Thinking and
Creativity in Organizations, European Journal of
Innovation Management, Vol.7, No.3, pp.187-204.
2011 335
COMPARING BANKRUPTCY PREDICTION MODELS IN IRAN
Ali Ebrahimi Kordlar, Nader Nikbakht
Faculty of management, Tehran university, Tehran, Iran
Email: [email protected]
Abstract
This paper compares the predictive power of a number of previous research models on bankruptcy prediction in Tehran Stock
Exchange (TSE) from 2001 to 2009. To compare the predictive power of these models, the adjusted R
2
(as In-sample metric) and
root mean squared error (RMSE), mean absolute error (MAE) and mean absolute percent error (MAPC) (as out-of-sample prediction
metrics). Finally, based on the estimation results of previous models in Iran, we present a fnal version logit type model that has higher
performance than other models. The empirical results of In-sample and out-of-sample prediction power indicate that our presented model
has higher adjusted R
2
and lower RMSE, lower MAE and lower MAPC, too. Key words: Bankruptcy prediction model, Logit model,
Voung test, Tehran Stock Exchange
Since Beaver (1966), an expanded literature on
bankruptcy prediction has emerged, and its impact has
spilled into the commercial world, where it has been used
in the development of several commercially employed
bankruptcy prediction models. Out of this literature have
come a number of competing empirical models with
alternative explanatory variables and alternative statistical
methodologies for model estimation.
The dependent variable in these models is commonly a
dummy variable where ''frm fled for bankruptcy" is set to
1 and other is set to 0. The independent variables are often
accounting ratios extracted from fnancial statements and
include measures of proftability, liquidity, and leverage.
Some studies also include market-based variables such
as the volatility of stock returns and past excess returns.
The accounting-based models developed by Altman (1968)
and Ohlson (1980) have emerged as the most popular
bankruptcy prediction models and are often used by
empirical accounting researchers as indicators of fnancial
distress.
Altman (1968) employs multivariate discriminate
analysis (MDA) on a list of fnancial ratios to identify
those ratios that are statistically associated with future
bankruptcy. Ohlson (1980) uses a logit ?model, which uses
less restrictive assumptions than those taken by the MDA
approach. Zmijewski (1984) adopts a probit approach that
is also based on accounting data but uses a different set
of independent variables. All of these approaches predict
future bankruptcy based on accounting ratios drawn from
frm’s fnancial statements.
More recently, Shumway (2001) has proposed a discrete-
time hazard ? model to predict a frm's bankruptcy using
both accounting and market variables. The main difference
between this model and the static logit ? model is that the
hazard model can be estimated within the logit ?framework
while using the entire life span of information (all frm-
years) for each frm. By contrast, the static logit ? model
can only incorporate one frm-year for each observation
(i.e., each observation consists of a single set of variables
observed at a single point in time).
Another stream of the bankruptcy prediction literature
focuses on market-based information. For example,
Hillegeist et al. (2004) have developed a BSM-Prob
bankruptcy prediction model that is based on the Black-
Scholes-Merton option-pricing model. Their results
indicate that the BSM-Prob model outperforms the models
of Altman (1968) and Ohlson (1980) in a series of tests.
There are also a number of papers that propose various
frm-characteristics that may be useful to future bankruptcy
prediction. For instance, Rose (1992) presents a model of
frm diversifcation in which managers use diversifcation
to reduce the risk of bankruptcy, particularly where the
ratio of the manager's frm-specifc human capital to his
non-frm-specifc human capital is high. Denis et al. (1997)
measure corporate diversifcation by the number of business
segments. Beaver et al. (2005) propose that, other things
equal, large frms have a smaller probability of bankruptcy
and that a part of this explanation is related to corporate
diversifcation. That is, corporate diversifcation and frm-
size are two frm-characteristics that may help to predict
future bankruptcy.
Hillegeist et al. (2004) compare the performance of their
BSM-Prob model against the Altman and Ohlson models in
a series of in-sample and out-of-sample tests, concluding
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
336 Business Intelligence Journal July
that the BSM model outperforms the accounting-based
models. Similarly, Chava and Jarrow (2004) examine the
relative performance of Shumway's hazard model against
the Altman and Zmijewski models, concluding that the
hazard model outperforms static logit models.
Wu, Gaunt and Gray (2010) build a new model
comprising key variables from each of the previous models
(Altman 1968; Ohlson 1980; Zmijewsky 1984; Shumway
2001; and Hillegeist et.al 2004) and add a new variable that
proxies for the degree of diversifcation within the frm.
The degree of diversifcation is shown to be negatively
associated with the risk of bankruptcy. This more general
model outperforms the existing models in a variety of in-
sample and out-of-sample tests.
Base on Zmijewski and Shirata models, Mehrani,
Mehrani, Monsef and Karami (2005) present a new
prediction model to bankruptcy prediction in Tehran
Stock Exchange (TSE). To reach this model, research
hypotheses are divided to two groups. The frst group is on
the power of classifying corporations in to bankrupt and
non-bankrupt frms. These hypotheses confrm the power
of correct classifying corporations in two bankrupt and
non-bankrupt groups. The second group is on the difference
between importances of fnancial ratios as an independent
variable. These hypotheses confrm the difference between
importances of independent variables in predicting
corporate failures.
Etemadi, Anvarirostami and Dehkordi (2009)
investigate the application of Genetic Programming
(GP) for bankruptcy prediction modeling. They use GP
to classify 144 bankrupt and non-bankrupt Iranian frms
listed in Tehran stock exchange (TSE). Then a multiple
discriminant analysis (MDA) was used to benchmarking
GP model. Genetic model achieved 94% and 90% accuracy
rates in training and holdout samples, respectively; while
MDA model achieved only 77% and 73% accuracy rates in
training and holdout samples, respectively.
Rahnamai Roud Poshti, Alikhani and Maranjouri (2009)
compare the results of prediction power of Altman and
Fulmer models and fnd that there is a signifcant difference
between the performances of two models and show that
Altman model has done more conservatively than Fulmer
model as well.
Makian, Almodaresi and Karimi (2010) fnd that an
Artifcial Neural Network (ANN) model performs much
better than the discriminant analysis and logistic regression
techniques. Moreover, the results confrm that the accuracy
of ANN model is higher than the discriminant analysis and
logistic regression techniques for predicting of bankruptcy.
The analysis also shows that none of the frms will bankrupt
in the year after the period covered in this study. Komijani
and Saadatfar (2007) fnd that in comparison with other
models, applying neural network models can improve
the potentials of fnancial managements to stand against
economic fuctuations and bankruptcy.
The rest of paper is organized as follows. Research
models and data collection process are described in section
2. Section 3 presents descriptive statistics, correlation
analysis and regression results. The last section of study
provides conclusion remarks.
Methodology
Research models
There are a number of key models that have been
developed by various researchers and presented in the
bankruptcy prediction literature over the last three decades.
A number of the most prominent models in bankruptcy
prediction models are as follows:
a. Altman (1968) multiple-discriminate analysis model
it it it it it it it
X X X X X Failing ? ? ? ? ? ? ? + + + + + + =
5 5 4 4 3 3 2 2 1 1
(1)
Where:
1
9 8 7
6 5 4
3 2 1
exp 1
?
¦
)
¦
`
¹
¦
¹
¦
´
¦
|
|
|
.
|
\
|
(
(
(
¸
(
¸
+ + + +
+ + +
+ + +
? + =
it it it it
it it it
it it it
it
CHIN INTWO FUTL
NITA OENEG CLCA
WCTA TLTA OLSIZE
Failing
? ? ? ?
? ? ?
? ? ? ?
(2)
Failing: is 0 for failed frm-years and 1 for other frm-years.
X1: Net working capital/total assets.
X2: Retained earnings/total assets.
X3: Earnings before interest and taxes/total assets.
X4: Market value of equity/book value of total liabilities.
X5: Sales/total assets.
b. Ohlson (1980) logit model
2011 337
Where:
OLSIZE: Log (total assets/GNP price-level index). The index
assumes a base value of 100 for 1988.
TLTA: Total liabilities divided by total assets.
WCTA: Working capital divided by total assets.
CLCA: Current liabilities divided by current assets.
OENEG: 1 if total liabilities exceed total assets, 0 otherwise.
NITA: Net income divided by total assets.
FULT: Funds provided by operations (income from operation
after depreciation) divided by total liabilities.
INTWO: 1 if net income was negative for the last 2 years, 0
otherwise.
CHIN: (NI
t
-NI
t-1
)/(|NI
t
|+|NI
t-1
|), where NIt is net income for
the most recent period. The denominator acts as a level
indicator. The variable is thus intended to measure the
relative change in net income.
The variable of Failing is defned similar to previous
models.
c. Zmijewski (1984) probit model
( )
it it it it it
CACL TLTA NITL Failing ? ? ? ? ? + + + + ? =
3 2 1
(3)
Where:
NITL: Net income divided by total liabilities
CACL: Current assets divided by current liabilities.
and TLTA is defned in Ohlson’s model.
d. Shumway (2001) hazard model
Now, Based on previous models, we present a logit ?type
combined model as follows:
e. Combined model
1
6 5
4 it 3
3 2 1
1
2
exp 1
?
¦
¦
)
¦
¦
`
¹
¦
¦
¹
¦
¦
´
¦
|
|
|
|
.
|
\
|
(
(
(
¸
(
¸
+ + +
+ +
+ +
? + =
? it
it it
it
it it
it
CACL NITL
INTWO OENEG
X X
Failing
? ? ?
? ?
? ? ?
(5)
Where, all variables are defned in previous models. We
expect that the predictive power of this model is higher than
that of other models.
To compare the predictive power of research models
to future bankruptcy, we use the adjusted R
2
and related
Voung (1989) test (as In-sample prediction metric) and root
mean squared error (RMSE), mean absolute error (MAE)
and mean absolute percent error (MAPC) (as out-of-sample
prediction metrics)
Data
We use the 2009 version of Tadbirpardaz (the Iranian
database of Tehran Stock Exchange) annual data fles and
sample all frms in Tehran Stock Exchange between 2001
and 2009 with suffcient data available to calculate the
research variables. In some cases whereby the required
data is incomplete, we use the manual archive in the TSE’s
library. We eliminate banks and fnancial institutions from
sample. Imposing all the data-availability requirements
yields 1,532 frm-years over the period 2001–2009. This is
the full sample that we use for testing research hypotheses.
Research results
Descriptive statistics
Table 1 provides descriptive statistics of research
variables in failed and continuous frm-years. There are
142 failed frm-years and 1390 continuous frm-years. In
our available statistical population the mean (median) of
X1, X2, X3, X4 and X5 in failed frms are -0.50 (-0.38),
-0.63 (-0.42), -0.07 (-0.05), 0.40 (0.27) and 0.61 (0.53). The
mean and median of all these variables for failed frms are
lower than the mean and median of mentioned variables
for continuous frms. The mean (median) of OLSIZE for
failed frms, -0.71 (-0.77) is lower than that of OLSIZE
1
1 5
1 4
it 3 2 1
RESIZE
exp 1
?
?
?
¦
)
¦
`
¹
¦
¹
¦
´
¦
|
|
|
.
|
\
|
(
(
(
¸
(
¸
+ +
+
+ + +
? + =
it it
it
it it
it
LAGSIGMA
LEXRETURN
TLTA NITL
Failing
? ?
?
? ? ? ?
(4)
Where:
RESIZE: Log (the number of outstanding shares multiplied
by year-end share price then divided by total market
value).
LEXRETURN: Cumulative annual return in year t-1 minus the
value-weighted TSE index return in year t-1.
LAGSIGMA: Standard deviation of the residual derived from
regressing monthly stock return on market return in
year t-1.
and other variables are defned in previous models.
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
338 Business Intelligence Journal July
for continuous frms, -0.07 (-0.15), too. Failed frms on
average have a higher debt ratio (1.31) than continuous
frms (0.65). Also, the mean of ratio of current assets on
current liabilities (CLCA) for failed frms (1.77) is greater
than this ratio for continuous frms (0.95). On average,
failed frms (-3.72) are smaller than the continuous frms
(-3.10). The mean of WCTA, CLCA and TLMTA for failed
frms (0.67, 1.77 and 0.77, respectively) is higher than the
mean of these variables for continuous frms (0.60, 0.95,
and 0.50, respectively).
The mean of NITA, FULT, CHIN, NITL and CACL (-0.14,
0.01, -0.05, -0.10 and 0.68) for failed frms is lower than the
mean of mentioned variables for continuous frms (0.13,
0.22, 0.03, 0.23 and 1.19, respectively). Furthermore, the
mean (median) of LEXRETURN and LAGSIGMA are -0.05
(-0.02) and 0.10 (0.03), respectively and are lower than
the mean (median) value of these variables for continuous
frms, 0.18 (0.15) and 0.12 (0.10), respectively.
Correlation analysis
The Pearson correlation coeffcients between research
variables are provided in Table 2. In this table, the
coeffcients that are marked with asterisk (*), are not
signifcant at the ordinary signifcance level.
Table 1. Descriptive statistics
Failed frms: 142 Obs. continuous frms: 1390 Obs.
Mean Median Maximum Minimum St.Dev Mean Median Maximum Minimum St.Dev
X1 -0.50 -0.38 0.21 -2.63 0.52 0.04 0.04 0.49 -0.69 0.17
X2 -0.63 -0.42 -0.05 -2.97 0.59 0.11 0.08 0.50 -0.20 0.11
X3 -0.07 -0.05 0.17 -0.43 0.10 0.16 0.14 0.55 -0.19 0.10
X4 0.40 0.27 5.47 0.03 0.64 1.68 0.93 15.51 0.03 2.05
X5 0.61 0.53 1.85 0.02 0.34 0.83 0.78 2.59 0.04 0.37
OLSIZE -0.71 -0.77 2.69 -3.20 1.16 -0.07 -0.15 4.41 -3.20 1.32
TLTA 1.31 1.19 3.34 0.57 0.51 0.65 0.66 1.20 0.19 0.15
WCTA 0.67 0.70 0.93 0.20 0.17 0.60 0.62 0.94 0.11 0.19
CLCA 1.77 1.52 5.49 0.64 0.91 0.95 0.88 4.95 0.26 0.39
NITA -0.14 -0.12 0.19 -0.62 0.13 0.13 0.11 0.53 -0.19 0.10
FUTL 0.01 0.00 0.41 -0.26 0.11 0.22 0.17 1.53 -0.36 0.26
CHIN -0.05 -0.02 0.94 -1.00 0.43 0.03 0.05 1.00 -1.00 0.31
NITL -0.10 -0.10 0.25 -0.26 0.09 0.23 0.16 1.52 -0.24 0.23
CACL 0.68 0.66 1.57 0.18 0.27 1.19 1.14 3.84 0.20 0.42
RESIZE -3.72 -3.76 -2.25 -4.67 0.48 -3.10 -3.13 -1.35 -4.70 0.66
LEXRETURN -0.05 -0.02 0.57 -1.00 0.22 0.18 0.15 1.46 -1.77 0.33
LAGSIGMA 0.10 0.03 0.68 0.00 0.13 0.12 0.10 0.74 0.00 0.09
TLMTA 0.77 0.79 0.97 0.15 0.15 0.50 0.52 0.97 0.06 0.21
Table 2. Pearson correlation coeffcient
X
1
X
2
X
3
X
4
X
5
O
L
S
I
Z
E
T
L
T
A
W
C
T
A
C
L
C
A
N
I
T
A
F
U
T
L
C
H
I
N
N
I
T
L
C
A
C
L
R
E
S
I
Z
E
L
E
X
R
E
T
U
R
N
L
A
G
S
I
G
M
A
X2 0.74
X3 0.42 0.60
X4 0.18 0.26 0.54
X5 0.17 0.15 0.29 0.01*
OLSIZE -0.04* 0.15 0.09 -0.08 -0.15
TLTA -0.80 -0.86 -0.56 -0.35 -0.07 -0.07
2011 339
This table presents Pearson correlation coeffcient. The coeffcients that are marked with asterisk (*), are not signifcant but others are signifcant at
the 5% level or lower.
The results of Pearson correlation coeffcients show
that the most of coeffcients between research variables are
signifcant at the 5% level or better.
Model estimation
The regression results of models and In-sample and
out-of-sample prediction metrics to compare the models
are presented in Table 3. The estimation results of Altman
(1968) model indicate that X1 (-0.12), X2 (-0.48) and X3
(-0.52) have a negative and signifcant relationship with
dependent variable (Failing) at the 1% level and X4 (0.01)
has a positive relation with dependent variable at the 5%
level but X5 has no signifcant relationship with dependent
variable. The F-statistic of this model (468.16) is signifcant
at the 1% level.
Altman (1968)
model
Ohlson (1980) model
Zemijewsky (1984)
model
Shumway (2001)
model
Combined model
Panel A: Model estimation results
Intercept 0.18** -12.87** -8.00** -16.94** -2.22**
X1 -0.12** -0.95
X2 -0.48** -13.40**
X3 -0.52**
X4 0.01*
X5 0.00
OLSIZE -0.14
TLTA 14.58** 8.21** 17.75**
WCTA -2.92
CLCA -0.60
OENEG -0.17 1.73**
NITA -1.40
FUTL -2.60
INTWO 3.79** 0.25
Table 3: The results of regression models
X
1
X
2
X
3
X
4
X
5
O
L
S
I
Z
E
T
L
T
A
W
C
T
A
C
L
C
A
N
I
T
A
F
U
T
L
C
H
I
N
N
I
T
L
C
A
C
L
R
E
S
I
Z
E
L
E
X
R
E
T
U
R
N
L
A
G
S
I
G
M
A
WCTA 0.31 -0.12 -0.11 -0.24 0.26 -0.17 0.24
CLCA -0.90 -0.61 -0.37 -0.18 -0.23 0.08 0.67 -0.39
NITA 0.51 0.68 0.91 0.55 0.20 0.12 -0.68 -0.18 -0.44
FUTL 0.19 0.36 0.61 0.57 0.12 0.02* -0.43 -0.32 -0.20 0.56
CHIN 0.06 0.09 0.29 0.16 0.09 0.01* -0.07 0.00* -0.07 0.33 0.10
NITL 0.39 0.52 0.81 0.64 0.13 0.04* -0.59 -0.23 -0.37 0.86 0.69 0.28
CACL 0.76 0.44 0.35 0.29 0.16 -0.15 -0.57 0.31 -0.77 0.39 0.28 0.07 0.51
RESIZE 0.09 0.28 0.45 0.43 -0.07 0.76 -0.26 -0.27 -0.04* 0.48 0.30 0.10 0.40 -0.01*
LEXRETURN 0.10 0.16 0.27 0.21 0.01* 0.15 -0.09 0.04* -0.09 0.29 0.03* 0.09 0.20 0.06 0.34
LAGSIGMA 0.10 0.05 0.05 0.02* 0.03* 0.05 -0.06 0.03* -0.10 0.07 -0.04* 0.06 0.01* 0.04* 0.11 0.09
TLMTA -0.34 -0.41 -0.67 -0.79 -0.07 0.10 0.54 0.28 0.30 -0.71 -0.56 -0.16 -0.73 -0.37 -0.51 -0.28 -0.08
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
340 Business Intelligence Journal July
Altman (1968)
model
Ohlson (1980) model
Zemijewsky (1984)
model
Shumway (2001)
model
Combined model
CHIN -0.25
NITL -3.21* -9.88* -0.91
CACL -0.03 0.01
RESIZE 0.01
LEXRETURN -1.25
LAGSIGMA -0.60
TLMTA
Panel B: In-sample prediction power metric
Adj. R
2
51.25 Pse. R
2
83.21 72.71 76.07 87.83
F stat. 468.16** LR stat. 904.61** 1272.92** 1083.25** 1384.16**
Voung Z 11.48** 2.21* 4.81** 3.22**
Panel C: Out-of-sample prediction power metrics
RMSE 0.21 0.14 0.16 0.14 0.12
MAE 0.14 0.03 0.06 0.05 0.03
MAPE 4.96 1.64 2.69 2.05 1.41
*, ** signifcant at the 5% and 1%, respectively
In the Ohlson (1980) model, TLTA (14.58) and the
dummy variable of INTWO (3.79) are signifcant at the
1% level and in Zemijewsky (1984) model TLTA (8.21)
and NITL (-3.21) are signifcant at the 1% and 5% level,
respectively. The results of Shumway (2001) model
indicates that TLTA (17.75) and NITL (-9.88) are signifcant
at the 1% and 5% level respectively. Finally, our presented
model show that X2 (-13.40) and OENG (1.73) are both
signifcant at the 1% level.
The results of likelihood ratio (LR) for Ohlson model
(904.61), Zemijewsky model (1272.92), Shumway model
(1083.25) and our presented combined model (1384.16)
show that all models are signifcant at 1% level, generally.
The In-sample prediction power (adjusted R
2
) of our
combined model (87.83%) is higher than that of Altman
model (51.25%), Ohlson model (83.21%), Zemijewsky
model (72.71%) and Shumway model (76.07%). the results
of Voung (1989) test show that the differences between the
prediction power of combined model and Altman model
(11.48), Ohlson model (2.21), Zemijewsky model (4.81)
and Shumway model (3.22) are all signifcant at the 1%
level.
The results show that the root mean squared error
(RMSE) of our combined model (0.12) is lower than that
of Altman model (0.21), Ohlson model (0.14), Zemijewsky
model (0.16) and Shumway model (0.14). The results
also indicate that the mean absolute error (mean absolute
percent error) of our combined model, 0.03 (1.41) is lower
than (or equal to) that of Altman model, 0.14 (4.96), Ohlson
model, 0.03 (1.64), Zemijewsky model, 0.06 (2.69) and
Shumway model, 0.05 (2.05). Thus, the results indicate that
the In-sample and out-of-sample prediction power of our
presented model in predicting failed frms is higher than that
of other previous models. Thus, our logit type bankruptcy
prediction model signifcantly outperforms other models.
Conclusion
In this paper, we examine the predictive power of a
number of bankruptcy prediction models. The models use
a range of different independent variables (accounting
information and market and frm-characteristic data) and
a range of different econometric specifcations (multiple-
discriminate analysis, logit ?, probit models). We fnd that
frms are more likely to experience bankruptcy if they
have relatively lower earnings before interest and tax to
total assets, a larger decline in net income, relatively low
working capital to total assets, or high total liabilities to
total assets.
We compare the empirical performance of a range of
bankruptcy prediction models using a series of in-sample
and out-of-sample performance metrics. Across all of these
metrics, both in-sample and out-of-sample, we fnd that, our
combined model signifcantly outperforms models from the
extant literature.
2011 341
References
Agarwal, V., and Taffer, R. (2008), "Comparing the
performance of market-based and accounting-
based bankruptcy prediction models", Journal of
Banking & Finance 32, PP: 1541-1551.
Altman, E. I. (1968), "Financial ratios, discriminant
analysis and the prediction of corporate bankruptcy",
Journal of Finance 23, PP: 589-609.
Beaver, W. H. (1966), "Financial ratios as predictors
of failure", Journal of Accounting Research 4, PP:
71-111.
Beaver, W. H., McNichols, M.F., and Jung-Wu, R.
(2005), "Have fnancial statements become less
informative? Evidence from the ability of fnancial
ratios to predict bankruptcy", Review of Accounting
Studies 10, PP: 93-122.
Begley, J., Ming, J., and Watts, S. (1996), "Bankruptcy
classifcation errors in the 1980s: an empirical
analysis of Altman's and Ohlson's models", Review
of Accounting Studies 1, PP: 267-284.
Burgstahler, D., James, J., and Noreen, E. (1989),
"Changes in the probability of bankruptcy and equity
value", Journal of Accounting and Economics 11,
PP: 207-224.
Campbell, J. Y., Hilscher, J., and Szilagyi, J. (2008),
"In search of distress risk", Journal of Finance 63,
PP: 2899-2939.
Chava, S., and Jarrow, R. A. (2004), "Bankruptcy
prediction with industry effects", Review of
Finance 8, PP: 537-569.
Denis, D. J., Diane, K., and Denis, A. S. (1997),
"Agency problems, equity ownership, and corporate
diversifcation", Journal of Finance 52, PP: 135-
160.
Dichev, I. D. (1998), "Is the risk of bankruptcy a
systematic risk?", Journal of Finance 53, PP: 1131-
1147.
Etemadi, H., Anvari Rostami, A. A., and Dehkordi,
H. F. (2009), "A genetic programming model for
bankruptcy prediction: empirical evidence from
Iran", Expert Systems with Applications March,
36, PP: 3199-3207.
Francis, J. (1990), "Corporate compliance with debt
covenants", Journal of Accounting and Research
28, PP: 326-347.
Hillegeist, S. A., Keating, E., Donald P., Cram, J. G.,
and Lunstedt. (2004), "Assessing the probability of
bankruptcy", Review of Accounting Studies 9, PP:
5-34.
Komijani, A., and Saadatfar, J. (2007), "Application
of "artifcial neural network models", in predicting
economic bankruptcy of the registered corporations
in stock market", Journal of Economic Essays 3,
PP: 11-44.
Lyandres, E., and Alexei, Z. (2007), "Investment
opportunities and bankruptcies prediction.
Working Paper, Rice University, Houston. Ohlson,
James, 1980. Financial ratios and the probabilistic
prediction of bankruptcy", Journal of Accounting
Research 18, PP: 109-131.
Makian S.N.A.D., Almodaresi S.M.T., and Karimi
Takalou S. (2010), "A comparison among artifcial
neural network, discriminant analysis and logistic
regression techniques for bankruptcy: a case study
of Kerman's frms", The Economic Research 10,
PP: 141-161.
Mehrani, S., Mehrani, K., Monsef, Y., and Karami,
Gh. R. (2005), " Implication Zmijewski and Shirata
bankruptcy prediction models in Tehran stock
exchange", The Iranian Accounting and Auditing
Review 12, PP: 105-131.
Rahnamai Roud Poshti, F., Alikhani, R., and
Maranjouri, M. (2009), "Applicational investigation
of Altman and Fulmer bankruptcy prediction
models in Tehran stock exchange", The Iranian
Accounting and Auditing Review 16, PP: 19-34.
Rose, D. (1992), "Bankruptcy risk, frm-specifc
managerial human capital, and diversifcation",
Review of Industrial Organization 7, PP: 65-73.
Shumway, T. (2001), "Forecasting bankruptcy more
accurately: a simple hazard model", Journal of
Business 74, PP: 101-124.
Stone, M. (1991), "Firm fnancial stress and pension
plan continuation/replacement decisions", Journal
of Accounting and Public Policy 3, PP: 175-206.
Vassalou, M., and Xing, Y. (2004), "Default risk in
equity returns", Journal of Finance 59, PP: 831-868.
Vuong, Q. H. (1989), "Likelihood ratio tests for
model selection and non-nested hypotheses",
Econometrica 57, PP: 307-333.
Kordlar A. E., Nikbakht N. - Comparing Bankruptcy Prediction Models in Iran
Ali Ebrahimi Kordlar, Nader Nikbakht
Business Intelligence Journal - July, 2011 Vol.4 No.2
342 Business Intelligence Journal July
Wu, Y., Gaunt, C., and Gray, s. (2010), "A comparison of
alternative bankruptcy prediction models", Journal
of Contemporary Accounting and Economics 6,
PP: 34-45.
Zmijewski, M. E. (1984), "Methodological issues
related to the estimation of fnancial distress
prediction models", Journal of Accounting Research
22 (Suppl.), PP: 59-82.
2011 343
STRATEGIC INFLUENCE OF PROMOTIONAL MIX ON
ORGANISATION SALE TURNOVER IN THE FACE OF STRONG
COMPETITORS
Adebisi Sunday. A (Ph.D)
Department of Business Administration, Faculty of Management sciences
University of Ado Ekiti, P.M.B 5363, Ekiti State Nigeria.
Email: [email protected]
Babatunde Bayode .O
Department of Business Administration, College of Management and Social Sciences,
Osun State University, P.M.B 2008, Okuku, Osun State, Nigeria
Email: [email protected]
Abstract
This paper aim at study strategic infuence of promotional mix on organisation sale turnover in manufacturing organisation.
The research data were gathered through the use of secondary data and primary data, secondary data included 6years annual report
comprising the sales turnover (2005-2009) and questionnaire which is an instrument of primary data collection. The questionnaires
were administered to the workers of 7up Company and some customers in Solebo Estate in Lagos. The researcher adopted the simple
percentage and regression model for the analysis of the collected data.. The result of the fndings revealed that strategic promotional
mix infuences the sale turnover with little 25% while other variable not included in the variable tested takes the larger 75% that will
rapidly lead to organisation growth. Since promotional mix constitute few % of variable that can push an organisation to the highest
level, therefore other factors of marketing mix such as product development, effective pricing, distribution of right quality and quantity
to the consumers should be appropriately considered. Key words: Strategic promotional mix, marketing mix, regression model, push
and pull strategy.
It is not enough for a business to have good products
sold at attractive prices. To generate sales and profts, the
benefts of products have to be communicated to customers.
In marketing, this is commonly known as “promotion”.
Although promotion is not done only for these factors but
for other such as to build brand loyalty, to reminds and
reassure costumers, to launch a new product and maybe
to defend market share by responding to competitors’
campaigns with their own advertising A business’ total
marketing communications programme is called the
“promotional mix” and consists of a blend of advertising,
personal selling, sales promotion, public relation stool sand
direct marketing.
The organization has to convey the message about the
product on offer to its consumers. This helps in sustaining a
perennial demand for the product and in suitably positioning
it among the target audience. The process of communicating
the message is called promotion. It infuences the purchase
decision of the consumer. The different channels available
to the organization for communicating the message
constitute the promotion mix. It includes advertising, sales
promotion, and public relations. The paper examines the
different elements of the promotion mix and discusses the
issues involved therein. It also delves into the advantages of
using the Internet as a selling medium.
It is very important for every product to be promoted,
that is to say it needs to be drawn to the attention of the
market place and it’s beneft be identifed. The aim of an
organization promotional strategy is to bring existing and
potential customers to a state of relative awareness of the
organization’s product and a not just that but also to a state
of adoption.
The promotional mixes (sales promotion, publicity,
personal selling, advertising, public relation) have a stage
at which it will be most effective. Advertising and publicity
are suitable for all stages while the remaining mix can be
effective from stage three.
Sunday A. A., Bayode O. B. - Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors.
Adebisi Sunday A., Babatunde Bayode O.
Business Intelligence Journal - July, 2011 Vol.4 No.2
344 Business Intelligence Journal July
Literature review
Cole (1996) defnes promotional mix as “the means use
in bringing customers from a state of relative unawareness
to a state of actively adopting the product”. It means of
communicating with individuals, groups, or organizations
to directly or indirectly facilitate exchange of informing and
persuading one or more audience to accept an organization’s
product.
Ross (2001) sees promotional mix as “the total marketing
communication programme of a particular product”.
Adebisi (2006) defned promotional mix as “any
marketing effort whose function is to inform or persuades
actual or potential consumers about the merit a product
possess for the purpose of inducing a consumer to either
start buying or continue to purchases the frm’s product.”
Elements of Promotional Mix
Every product needs to be drawn to the attention of
the target market, and its beneft identifed. The principal
methods are;
• Advertising
• Personal Selling
• Sales promotion
• Publicity
• Public relation
The aim of an organization’s promotional strategy
is to bring existing or potential from a state of relative
unawareness of the organization’s product to a state of
actively adopting them. Several stages of customer’s
behaviours have been identifed. This has been described
in several ways, but in summary can be stated as follows;
Stages Behaviours
One Unawareness of product
Two Awareness of product
Three Interest in product
Four Desire for product
Five Conviction about value of product
Six Adoption/purchases of product
Source: Ross (2001) Stages of Product Awareness
Advertising
Advertising is the process of communication, persuasive
information about a product to the markets by means of the
written and spoken word. There are fve principal media of
advertising as follows; The press, commercial television,
direct mail, commercial radio and outdoor.
Objective of Advertising
A company’s objective for embarking on advertisement
can therefore be on any of the following objectives
enumerated below;
i. To introduce a new product or service; here
advertisement attempt to present to the prospective buyers
a new product or service and this usually near a costly and
dramatic launching of a new product or service.
ii. To expand the market to new buyers; advertisement
is done to introduce a product to new buyers who might
fnd interest or usefulness of it.
iii. To announce modifcation; a product that is already
in the market might want to be given a new face, and then
there is need for advertisement to highlight to the consumers
that modifcations had been done.
iv. To announce a price change; advertisement can
be done if the organization’s product or service price
is increased or decreased so that the consumer might be
informed.
v. To introduce a new packaging; packaging
identifcation at the point of sales is always important and
is a reason why packages are shown in advertisement.
vi. To stimulate sales promotion; advertising maybe
make special offer when business environment is static. It
could be a gift of cup or biro etc at the purchase of such
product.
vii. To educate consumers; the educational
advertisement is necessary when a commodity, service or
an offer needs careful explanation.
viii. To maintain sales; customers product which had
been in the market use advertisement to maintain sales.
ix. To challenge competition; the purpose of some
advertisement are to challenge the competitors in the
presence of buyers.
2011 345
Personal Selling
This is the process by which the seller sells to the
consumer face to face.The personal selling consists of a
selling process, which is illustrated below.Personal selling
is the most expensive form of promotion. Company that
use more of personal selling are said to be adopting push
strategy while that of advertising are using pull strategy.
Sales Promotion
Sales promotion activities are a form of indirect
advertisement, designed to stimulate sales mainly by the use
of incentives; Free sample, Twin-pack bargain, Temporary
price reduction, Special discount bonus.
Publicity
Publicity differs from other promotional mix in that it is
costless most of the time. Publicity according to Cole (1996)
is “news about the organization or its products reported in
the press”. Publicity sometimes cost but its cost is always
related with advertisement. Publicity is a very necessary
tool because it creates the good will of an organization.
Use of publicity
Publicity when properly managed by the Public Relation
Offcer of an organization can serve the following purposes;
it can be used to attract public attention, it can also be used
to maintain public visibility and used for the provision of
information to the public. Publicity often takes the form of
news released or press conferences, appearance or event
sponsorship.
Public Relation
Public relation is another form of promotion. It is the
means by which the organization related or communicates
with the environment. Public relation is aimed at better
customer relations and immediate feedback.
Personal Selling Advertisement Sales Promotion Direct Marketing Public Relation
A
d
v
a
n
t
a
g
e
s
Permit measurement of
efectiveness.
Elicit an immediate
response.
Tailors the message to ft
the customer.
Personal contact
Reaches a large group
of potential buyers for a
relative low price.
Allows strict control over the
fnal message.
Can be adapted to either
mass audience or specifc
audience segment.
Produces an immediate
consumer response.
Attracts attention and
creates product awareness.
Allows easy measurement
of result.
Provides short time sales
increase.
Generates an immediate
response.
Covers a wide audience with
targeted advertising.
Allows complete, customized,
personal message.
Produces measurable results.
Creates a positive attitude
towards a product or
company.
Enhances credibility of a
product or company.
D
i
s
a
d
v
a
n
t
a
g
e
s
Relies almost exclusively
upon the ability of the
sales person.
Involve high cost per
contact.
Does not permit total
accurate measurement of
result.
Usually cannot close sales
Is non-personal in nature.
Is difcult to diferentiate
from eforts.
Sufers from image problem.
Involves a high cost per reader.
Depends on quality and
accuracy of mailing lists.
May not permit accurate
measurement of efect on
sales.
Involves much efort directed
towards non-marketing
oriented goals.
Promotional Mix Strategy
Marketing managers may choose between two
alternative strategies to use when promoting their product,
which are; Push strategy and Pull strategy.
a. Push strategy: When a market uses the pull promotion,
it means that the product involves “pushing” through
distribution channel till it gets to the fnal consumers.
The strategy involves the producer directing his
marketing activities towards channels members to
induce them to bring the product or promote the product
to the fnal consumers. One major promotional mix use
in this strategy is advertising and sales promotion.
b. Pull strategy: In this strategy the produce the fnal
consumers to induce them to buy the product. If the
strategy is effective, the consumers demand for the
product from channel members (middlemen). This is
the most used strategy.
In this strategy, consumer’s demand pulls the product
through the channel.
The two strategies can be applied simultaneously.
However the B2C (business to consumer) use more of pull
strategy while the B2B (business to business) use more of
the push strategy.
Sunday A. A., Bayode O. B. - Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors.
Adebisi Sunday A., Babatunde Bayode O.
Business Intelligence Journal - July, 2011 Vol.4 No.2
346 Business Intelligence Journal July
Push strategy versus Pull strategy
Producer’s marketing Resellers’ Marketing
Activities Activities
Producer Middlemen Consumers
Push Strategy
Source: Cole (1996) “Push Strategy”.
Demand Demand
Producer Middlemen Consumers
Producer’s marketing activities.
Pull Strategy
Source: Cole (1996) “Pull Strategy”.
Developing an optimal Promotional Mix
The blending of all mix of promotion (personal selling,
advertisement, public relation, sales promotion and
publicity) is what leads to the achievement of marketing
objectives. Qualitative measures can be used to determine
the effectiveness of a mix components in a given market
segment. The choice of a proper mix of promotion elements
present one of the most diffcult tasks for a marketer.
Steps to developing optimal Promotional Mix
An organization must make sure that is promotional mix
is effective, if it will beneft from it. The following are the
steps involves in having an optimal promotional mix.
Step 1; identify the audience: a marketing promotion
starts with clear target audiences in mind. The audience may
be potential buyers or current users, it may be individuals,
group, special public or the general public. The target
audiences are one of the major determinants of what mix
is to be used?
Step 2; determine the needs: after having identifed the
target audience, the next thing to do is to know what their
basic needs are and is the product relevant for the audience?
Step 3; determine the promotional objectives: the
promotional activities should have a target to be meant
which is referred to as objective. Is it to increase sales? Or
is it to just create awareness? It should be clearly stated.
Step 4; choose the mix: after the expected response is
determined, then the marketer can look critically at all mix
and identify which can best satisfy their objectives.
For a promotional mix to lead to purchase, it must have
pass through some stages which is illustrated in the diagram
below:
2011 347
The buying readiness stages
Brand Ignorance Awareness Knowledge
Conviction Purchase Liking Preference
Source: Cole (1996) “The Buying Readiness Stage”
Measuring The Effectiveness Of Promotional
Mix
If an organization cannot measure it’s effectiveness in
terms of promotion, it cannot say precisely if the activities
had been successful. For an organization to measure the
effectiveness of it promotional activities, this can be done
in either of the following ways;
• Direct sales result; this method reveals the sales revenue
for each amount input into promotion. That is, it
measures the rate of sales to the expense on promotion.
• Indirect evaluation; this method focus on quantitatiable
indicators of effectiveness. For instance, the
effectiveness is measured based on the organization
study of the number of audience that actually heard
about the product during the promotional activities.
• Returns method; this method is the work of Professor
Don Schutz. He said promotional effectiveness should
be measured based on the returns of the period of
promotion. What is the proft like during promotion
and when there is no promotion?
• Direct response method; this method is concentrating
on having a way of getting response from the targeted
audience and this response should be used to measure
the effectiveness of promotion.
Methodology
Methodology is a vital process of carrying out empirical
study. It forms the background in which the procedures
employed in carrying out a research are designed. It is the
procedure that follows a step after one another of which
data gather for a research is being analyzed”.
Population of the Study
The case study of this organization has many plants in
Nigeria in places like Ibadan, Ijora, Kano, Ikeja and so on.
The researcher therefore picked the Ikeja plant as the case
study. Marketing department personnel were sampled in the
organization since it is related with the topic under study.
The number of employees in the marketing department is
forty two as given by Mr. Wemimo the sales manager.
Sampling Procedure
Sampling method is the study of a fragment of the entire
population when it is not feasible to carry out the study on
the entire population. The study used the random sampling
method because it allows equal chance for every element
of the population and therefore the sample size was thirty
while the consumers were ffty.
Data Collection Instrument
This research work is both qualitative and quantitative
in nature, and as such data were collected via the use of
both primary data and secondary data.
Regression Method of Analysis
Adebisi (2006) defned regression as “the statistical
method of predicting and determining the probable value
of dependent (?) given the value of independent (?).The
hypothesis gathered for this study will be tested at 5% level
of signifcance.
The regression formular is:
? =a + b? + ui
Sunday A. A., Bayode O. B. - Strategic Infuence of Promotional Mix on Organisation Sale Turnover in the Face of Strong Competitors.
Adebisi Sunday A., Babatunde Bayode O.
Business Intelligence Journal - July, 2011 Vol.4 No.2
348 Business Intelligence Journal July
Where;
? = dependent variable
? = independent variable
a = intercept
b = slope
ui = error term.
The regression model will build on,
Hypothesis One
? (dependent variable) sales turnover
? (independent variable)Strategic promotional mix.
Hypothesis Two
? (dependent variable) market share and growth
? (independent variable)Strategic promotional mix
Table 1.
Years Turnover
Selling and Distribution
Expenses
2004 14937371 2931311
2005 17346662 2207731
2006 22071731 4557690
2007 27309123 5576791
2008 30572218 6364557
2009 34022650 8230830
Source: 2003-2009 7up Nig Plc statement of income and expenditure
bulletin,
Hypothesis one
Ho- Strategic promotional mix does not infuence sale
turnover rate positively in the face of full competition.
H1- Strategic promotional mix infuence sale turnover
rate positively in the face of full competition.
For this hypothesis the six years fnancial report of
7up was used and 45% of selling expenses was taken as
promotional mix.
Years Turnover
Selling and
Distribution
Expenses
% taken as
promotional
expenses
2004 14937371 2931311 1319089.95
2005 17346662 2207731 993478.95
2006 22071731 4557690 2140960.5
2007 27309123 5576791 2509555.95
2008 30572218 6364557 2864050.65
2009 34022650 8230830 3212302.50
Source: Data analysis 2010.
Model B Std. Error T Sig. t R R
2
F
C
o
n
s
t
a
n
t
1.2E +08 8.6E + 07 1.416 .230
.173
a
.030 .123
S
t
r
a
t
e
g
i
c
P
r
o
m
o
t
i
o
n
a
l
m
i
x
-6.752 19.268
-.350 .744
Source: Data analysis 2010
Regression analysis showing the strategic infuence of
promotional mix on sales turnover.
?=f(?)
? = a + b? + ui
=0.000000012+6.752 ? + ui
Std. error = (0.00000086) (19.268)
t =(1.416)(.350)
The estimated regression model is given as:
The result of the regression analysis shows that the
regression coeffcient (R) is .173
a
. It implies that there is
positive and strong infuence between strategic promotional
mix and sales turnover in 7up manufacturing company. The
result also reveals that the coeffcient of determinant (R
2
) is
.030, it connotes that about 3% variation in sales turnover
could be explained by promotional mix. The remaining 97%
were largely due to other variables outside the regression
model that also affects sales turnover.
Testing the effect of independent variable ( strategic
promotional mix) on dependent variable (sales turnover)
2011 349
Regression analysis showing the impact of promotional mix on
sales volume.
Model B
Std.
Error
T Sig. t R R
2
F
C
o
n
s
t
a
n
t
2.050 3.898 .526 .692
.500
a
.250 .333
S
t
r
a
t
e
g
i
c
P
r
o
m
o
t
i
o
n
a
l
m
i
x
.500 .866 .577 .667
Source: Data analysis 2010
?=f(?)
? = a + b? + ui
=2.050+.500? + ui
Std.error= (3.898)(.866)
t=(.526)(.577)
The result of the regression analysis shows that the
correlation coeffcient (R) is .500
a
. It implies that there is
positive and strong infuence between strategic promotional
the result shows that t-value is .350 at 0.05 level of
signifcant (t=.350: p