Advanced Topics in Global Information Management Series (Vol. 1) 1930708432, 9781930708433, 9781591400295

Advanced Topics in Information Resources Management features the latest research findings dealing with all aspects of in

303 22 3MB

English Pages 409 Year 2002

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Advanced Topics in Global Information Management Series (Vol. 1)
 1930708432, 9781930708433, 9781591400295

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Advanced Topics in Global Information Management Volume 1

Felix B. Tan University of Auckland, New Zealand

This book is a release of the Advanced Topics in Global Information Management Series

Idea Group Publishing

Information Science Publishing

Hershey • London • Melbourne • Singapore • Beijing

Acquisition Editor: Managing Editor: Development Editor: Copy Editor: Typesetter: Cover Design: Printed at:

Mehdi Khosrowpour Jan Travers Michele Rossi Amy Bingham LeAnn Whitcomb Tedi Wingard Integrated Book Technology

Published in the United States of America by Idea Group Publishing 1331 E. Chocolate Avenue Hershey PA 17033-1117 Tel: 717-533-8845 Fax: 717-533-8661 E-mail: [email protected] Web site: http://www.idea-group.com and in the United Kingdom by Idea Group Publishing 3 Henrietta Street Covent Garden London WC2E 8LU Tel: 44 20 7240 0856 Fax: 44 20 7379 3313 Web site: http://www.eurospan.co.uk Copyright © 2002 by Idea Group Publishing. All rights reserved. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher. ISBN 1-930708-43-2

Advanced Topics in Global Information Management-vol. 1 is part of the Idea Group Publishing series named Advanced Topics in Global Information Management (Series ISSN 1537-9302 )

eISBN 1-59140-029-5 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library.

NEW from Idea Group Publishing • • • • • • • • • • • •

• • • • • • •

• • •

• •

• •

Data Mining: A Heuristic Approach, Hussein Aly Abbass, Ruhul Amin Sarker & Charles S. Newton ISBN: 1-930708-25-4 / eISBN: 1-59140-011-2 / 310 pages / US$89.95 / © 2002 Managing Information Technology in Small Business: Challenges and Solutions, Stephen Burgess ISBN: 1-930708-35-1 / eISBN: 1-59140-012-0 / 367 pages / US$74.95 / © 2002 Managing Web Usage in the Workplace: A Social, Ethical and Legal Perspective, Murugan Anandarajan & Claire A. Simmers ISBN: 1-930708-18-1 / eISBN: 1-59140-003-1 / 386 pages / US$74.95 / © 2002 Challenges of Information Technology Education in the 21st Century, Eli Cohen ISBN: 1-930708-34-3 / eISBN: 1-59140-023-6 / 290 pages / US$74.95 / © 2002 Social Responsibility in the Information Age: Issues and Controversies, Gurpreet Dhillon ISBN: 1-930708-11-4 / eISBN: 1-59140-008-2 / 282 pages / US$74.95 / © 2002 Database Integrity: Challenges and Solutions, Jorge H. Doorn and Laura Rivero ISBN: 1-930708-38-6 / eISBN: 1-59140-024-4 / 300 pages / US$74.95 / © 2002 Managing Virtual Web Organizations in the 21st Century: Issues and Challenges, Ulrich Franke ISBN: 1-930708-24-6 / eISBN: 1-59140-016-3 / 368 pages / US$74.95 / © 2002 Managing Business with Electronic Commerce: Issues and Trends, Aryya Gangopadhyay ISBN: 1-930708-12-2 / eISBN: 1-59140-007-4 / 272 pages / US$74.95 / © 2002 Electronic Government: Design, Applications and Management, Åke Grönlund ISBN: 1-930708-19-X / eISBN: 1-59140-002-3 / 388 pages / US$74.95 / © 2002 Knowledge Media in Health Care: Opportunities and Challenges, Rolf Grutter ISBN: 1-930708-13-0 / eISBN: 1-59140-006-6 / 296 pages / US$74.95 / © 2002 Internet Management Issues: A Global Perspective, John D. Haynes ISBN: 1-930708-21-1 / eISBN: 1-59140-015-5 / 352 pages / US$74.95 / © 2002 Enterprise Resource Planning: Global Opportunities and Challenges, Liaquat Hossain, Jon David Patrick & M. A. Rashid ISBN: 1-930708-36-X / eISBN: 1-59140-025-2 / 300 pages / US$89.95 / © 2002 The Design and Management of Effective Distance Learning Programs, Richard Discenza, Caroline Howard, & Karen Schenk ISBN: 1-930708-20-3 / eISBN: 1-59140-001-5 / 312 pages / US$74.95 / © 2002 Multirate Systems: Design and Applications, Gordana Jovanovic-Dolecek ISBN: 1-930708-30-0 / eISBN: 1-59140-019-8 / 322 pages / US$74.95 / © 2002 Managing IT/Community Partnerships in the 21st Century, Jonathan Lazar ISBN: 1-930708-33-5 / eISBN: 1-59140-022-8 / 295 pages / US$89.95 / © 2002 Multimedia Networking: Technology, Management and Applications, Syed Mahbubur Rahman ISBN: 1-930708-14-9 / eISBN: 1-59140-005-8 / 498 pages / US$89.95 / © 2002 Cases on Worldwide E-Commerce: Theory in Action, Mahesh Raisinghani ISBN: 1-930708-27-0 / eISBN: 1-59140-013-9 / 276 pages / US$74.95 / © 2002 Designing Instruction for Technology-Enhanced Learning, Patricia L. Rogers ISBN: 1-930708-28-9 / eISBN: 1-59140-014-7 / 286 pages / US$74.95 / © 2002 Heuristic and Optimization for Knowledge Discovery, Ruhul Amin Sarker, Hussein Aly Abbass & Charles Newton ISBN: 1-930708-26-2 / eISBN: 1-59140-017-1 / 296 pages / US$89.95 / © 2002 Distributed Multimedia Databases: Techniques and Applications, Timothy K. Shih ISBN: 1-930708-29-7 / eISBN: 1-59140-018-X / 384 pages / US$74.95 / © 2002 Neural Networks in Business: Techniques and Applications, Kate Smith and Jatinder Gupta ISBN: 1-930708-31-9 / eISBN: 1-59140-020-1 / 272 pages / US$89.95 / © 2002 Managing the Human Side of Information Technology: Challenges and Solutions, Edward Szewczak & Coral Snodgrass ISBN: 1-930708-32-7 / eISBN: 1-59140-021-X / 364 pages / US$89.95 / © 2002 Cases on Global IT Applications and Management: Successes and Pitfalls, Felix B. Tan ISBN: 1-930708-16-5 / eISBN: 1-59140-000-7 / 300 pages / US$74.95 / © 2002 Enterprise Networking: Multilayer Switching and Applications, Vasilis Theoharakis & Dimitrios Serpanos ISBN: 1-930708-17-3 / eISBN: 1-59140-004-X / 282 pages / US$89.95 / © 2002 Measuring the Value of Information Technology, Han T. M. van der Zee ISBN: 1-930708-08-4 / eISBN: 1-59140-010-4 / 224 pages / US$74.95 / © 2002 Business to Business Electronic Commerce: Challenges and Solutions, Merrill Warkentin ISBN: 1-930708-09-2 / eISBN: 1-59140-009-0 / 308 pages / US$89.95 / © 2002

Excellent additions to your institution’s library! Recommend these titles to your Librarian! To receive a copy of the Idea Group Publishing catalog, please contact (toll free) 1/800-345-4332, fax 1/717-533-8661,or visit the IGP Online Bookstore at: [http://www.idea-group.com]! Note: All IGP books are also available as ebooks on netlibrary.com as well as other ebook sources. Contact Ms. Carrie Stull at [[email protected]] to receive a complete list of sources where you can obtain ebook information or IGP titles.

Advanced Topics in Global Information Management Table of Contents Preface ........................................................................................................... viii Felix B. Tan, University of Auckland, New Zealand Acknowledgments ..........................................................................................iii Section I: Global Themes Chapter I Multilingual Electronic Commerce in a Global Economy ......................... 1 Aryya Gangopadhyay and Zhensen Huang University of Maryland Baltimore County, USA Chapter II Electronic Commerce and Strategic Change within Organizations: Lessons From Two Cases ................................................ 12 Robert D. Galliers, London School of Economics, UK Sue Newell, Royal Holloway College, University of London, UK Chapter III Trust in Internet Shopping: Instrumental Development and Validation through Classical and Modern Approaches ................... 25 Christy M. K. Cheung and Matthew K. O. Lee City University of Hong Kong, Hong Kong Chapter IV The Integral Role of Information Technology in Achieving Business Strategy Success: Managing the Information Resources of Global Competitors .............................................................. 42 Jeffrey D. Katz, Kansas State University, USA Chapter V The New Realities of Online Trading in the New Global Marketplace ...................................................................................... 63 Bridget Allgood, University College Northampton, UK

Chapter VI The Impact of Culture on the Development of Information Systems: A Case Study ................................................................................ 74 Trevor T. Moores, University of Nevada Las Vegas, USA Frank H. Gregory, Independent Consultant, Thailand Chapter VII Dysfunctional Development Pathways of Information and Communication Technology: Cultural Conflicts .............................. 83 G. Roland Kaye and Stephen Little Open University Business School, UK Chapter VIII Ten Lessons That Internet Auction Markets Can Learn from Securities Market Automation .......................................................... 97 J. Christopher Westland, Hong Kong University of Science and Technology, Hong Kong Chapter IX Transnational Information Systems: Development and Management Issues ........................................................................... 120 Paul Mantelaers and Wander van den Berg Delft University of Technology, The Netherlands Chapter X The Place of ICT in Global Planning ....................................................... 136 Abel Usoro, University of Paisley, UK Chapter XI The Cultural Construction of Information Technology ........................ 150 Vanessa Dirksen, Universiteit van Amsterdam, The Netherlands Section II: Regional Perspectives Chapter XII Survey of IT Outstanding Experiences in US and UK Organizations ..................................................................... 160 Mary Cecelia Lacity, University of Missouri–St. Louis, USA Leslie P. Willcocks, Oxford University, UK

Chapter XIII Management Integration through Software Applications: Japanese Manufacturing Firms in the UK Exert Control .................... 190 John Kidd and Tessa Yuk Lan Yau Aston Business School, Birmingham, UK Chapter XIV Can National Information Infrastructures Enhance Social Development in the Least Developed Countries? .................... 207 Peter Nelson Meso and Nancy Bogucki Duncan Kent State University, USA Chapter XV Contract, Control and ‘Presentiation’ in IT Outsourcing: Research in Thirteen UK Organisations ................................................ 227 Thomas Kern, Erasmus University Rotterdam, The Netherlands Leslie Willcocks, University of Oxford, UK Chapter XVI A Systematic Model to Integrate Information Technology into Metabusinesses: A Case Study in the Engineering Realms ................................................................................... 250 Luiz Antonio Joia, Brazilian School of Public Administration, Getulio Vargas Foundation, Brazil Chapter XVII Facial Social Risks in IT Development in South Africa– Learning from Scandinavia ........................................................................ 268 Helana Scheepers, University of Pretoria, South Africa Lars Mathiassen, Aalborg University, Denmark Chapter XVIII IS Project Characteristics and Performance: A Kuwaiti Illustration ................................................................................. 290 Adel M. Aladwani, Kuwait University, Kuwait Chapter XIX Information Systems Leadership Roles: An Empirical Study of Information Technology Managers in Norway ....................... 304 Petter Gottschalk, Norwegian School of Management, Norway

Chapter XX The Problem of Networked Organizations in India: A Case Study ............................................................................................... 320 Gurpreet S. Dhillon and Trevor T. Moores University of Nevada Las Vegas, USA Ray Hackney, Manchester Metropolitan University, UK Chapter XXI Knowledge Assets in the Global Economy: Assessment of National Intellectual Capital ................................................................. 329 Yogesh Malhotra, Florida Atlantic University, USA Chapter XXII The Nature of E-Loyalty in B2C E-Commerce ...................................... 346 Daniel Tomiuk and Alain Pinsonneault McGill University, Canada Chapter XXIII Generic Attributes of IS Graduates: An Analysis of Australian Views .............................................................. 370 Robert Snoke and Alan Underwood Queensland University of Technology, Australia About the Authors ...................................................................................... 385 Index .......................................................................................................... 393

viii

Preface A growing body of knowledge is being accumulated in the area of global information management (GIM). Research in this field has grown significantly in the 1990’s. Not only are established information systems (IS) journals publishing an increasing amount of GIM research, but there are now journals like the Journal of Global Information Management that publish research specific to the field. What exactly is global information management? GIM research can be broadly considered to be of two types–one that is global and the other regional. GIM is a field of study that examines the development, implementation, management and use of IS in a global/international context (for example, global sourcing/supply chain management, global planning in multinational companies, global e-business strategies, and cultural/language issues). At the same time, GIM research also deals with management, technological and use of IS issues in a single- or multi-country organizational environment (for example, outsourcing experiences in U.S. and U.K. organizations, national information infrastructures in less developed countries, and IS project characteristics and performance in Kuwait). Topics of study in GIM research are therefore highly diverse. This book contains a selection of research papers on a wide range of GIM topics. It is hoped that the book will advance our understanding of the field. It is the first of a series of books on the subject. The book is organized into 2 sections–global themes and regional perspectives.

Global Themes The chapters in this section explore various IS issues in a global/international setting. The chapter by Gangopadhyay and Huang examine the issues relating to multilingual electronic commerce, in particular, the nature of user interactions in multilingual electronic catalogs. Galliers and Newell review and contrast the experiences of two multinational companies in attempting significant change projects incorporating information and communication technologies. Cheung and Lee develop and test a model on trust in the context of Internet shopping. Katz considers the impact of culture and business strategy on the organizational and information technology infrastructures of global competitors. Allgood discusses how the global nature of Internet-based share trading impacts the players in the competitive stock brokering environment. Moores and Gregory report on a case study that highlights the difficulties in applying Soft Systems Methodology in the development of IS in an Eastern culture. Kaye and Little argue that the assumptions of universality and

ix common linear pathways of development are flawed because of cultural conflicts. Westland explores the lessons Internet auction markets can learn from securities market automation by considering exchanges in Chile, Russia, and China. Mantelaers and van den Berg investigate issues relating to the development and management of transnational information systems. Usoro probes how global planning is achieved in multinational companies in the U.K. and South Africa. Dirksen calls for more cultural considerations in the study of information systems.

Regional Perspectives The chapters in this section delve into various IS issues in a single- or multi-country organizational setting. The chapter by Lacity and Willcocks detail the results of a survey of IT outsourcing experiences in U.S. and U.K. organizations. Kidd and Yau discuss the nature of computer software that has been imported from Japan to the UK to control imported production systems and /or to control management data flows. Meso and Duncan study the relationship between national information infrastructure and social development, suggesting ways governments of least developed countries may enhance their nations’ growth by developing strategic plans for national information infrastructure. Kern and Willcocks present findings from thirteen UK organizations on the role of outsourcing contract and its purpose for ensuring control over the client’s outsourcing destiny. Joia develops and tests a model to link effectively different information technologies in order to coordinate a metabusiness using a case study of a major engineering firm in Brazil. Scheepers and Mathiassen consider the development and implementation of information technology in South Africa based on experiences in Scandinavia. Aladwani discusses why managers cannot rely on prescriptions suggested by IS projects research in developed countries to understand IS projects in developing countries. Gottschalk presents results from a survey done in Norway on IS managers’ leadership roles. Dhillon, Moores and Hackney argue that many emerging economies, like India, may face potential misalignment of interests with the advent of networked organizations. Malhotra contends that there is a need for assessing knowledge capital at the national level and discusses current and future assessment methods. Tomiuk and Pinsonneault offer a conceptual framework to better understand the assess the impact of information technology on customer loyalty in retail banking. Snoke and Underwood investigate the differences between academic and industry views of the desired generic attributes of IS graduates in Australia using a Delphi study.

x

Acknowledgments I am grateful to the authors. Without their contributions this book would not have been possible. I would also like to thank Ms. Jan Travers, Dr. Mehdi Khosrowpour, and the team at Idea Group Publishing for their efforts and assistance in getting this book to print. Felix B. Tan August 2001

Section I Global Themes

Multilingual Electronic Commerce in a Global Economy 1

Chapter I

Multilingual Electronic Commerce in a Global Economy Aryya Gangopadhyay and Zhensen Huang University of Maryland Baltimore County, USA

The purpose of this research is to further the knowledge required for building electronic commerce systems that operate in multiple languages in global settings. The issues in multilingual electronic commerce are presented in two parts. First we describe a bilingual electronic catalog that can be used by online retailers for selling products and/or services to customers interacting in either English or Chinese that was developed to investigate into the nature of user interactions in multilingual electronic catalogs. Second, we discuss issues in developing multilingual electronic catalogs.

INTRODUCTION As all of commerce is converging on the Internet, the nature of business is changing rapidly. One of the main features of business on the Internet is the ability to transcend geographic boundaries. But along with the benefits of the widespread outreach of the virtual marketplace come many challenges. An example of such a challenge is to provide costeffective interchange across language and culture. Many organizations are confronted with the requirement of making their products or services available in multiple languages, particularly in their Asian and European markets. Internationalization of systems may involve enabling the input and display of non-English characters, changing default formats for date, time, currency, and measuring units, and using Unicode to handle the mix of European and Asian characters for complex operations such as rolling up data from multiple sites in many languages around the world1 . While language technology (Nirenburg, 1992; Onyshkevych & Nirenburg, 1995; Sheremetyeva & Nirenburg, 1996) is making rapid progress, much research is needed in managing and accessing multilingual information in order to reach full potential of global electronic commerce (e.g., Malhotra, 1997, 1998). Copyright © 2002, Idea Group Publishing.

2 Gangopadhyay & Huang

The purpose of this research is to further the knowledge required for building information systems that operate in multiple languages. Specifically, we focus on studying user behavior in performing various tasks in a multilingual system. In order to study user behavior and performance in a multilingual electronic commerce setting, we have designed a bilingual electronic catalog which can be used by online retailers for selling products and/ or services to customers interacting either in English or Chinese. An electronic catalog is a graphical user interface that presents product and/or service information to users, typically using the World Wide Web. An electronic catalog is a key component of electronic commerce that has been used for business-to-consumer commerce as well as business-to-business commerce (Adam et al., 1998). Although the term “electronic catalog” might sound like an electronic extension of chapter catalogs, it offers features that are far beyond those found in chapter catalogs. Such features include computational services such as efficient browsing and searching, online order processing such as checking out products using shopping carts and secure payment mechanisms, and backend processing such as integration with company databases (Segev et al., 1995). These features have extended the role of electronic catalogs to the point of being used as electronic storefronts. With the rapid proliferation of electronic commerce both in local and global markets, there is an increasing need to provide support for internationalization such as foreign currencies, different date and time formats, sort order, and multiple languages (Broin, 1999). The need for providing multilingual support is echoed by the rapid increase of non-English speaking users in the Internet. The rest of the chapter is organized as follows. In the next section we describe the electronic catalog and its components. Next, we discuss issues related to language preferences by bilingual users, based on an experimental study. Next we discuss various issues in designing multilingual systems. The last section contains our conclusions and future research directions.

A BILINGUAL ELECTRONIC CATALOG Description of the Catalog A prototype electronic catalog has been implemented in the World Wide Web using ColdFusion 4.0 as the front end, which is connected to a Microsoft Access database at the back end, using an ODBC driver. The catalog is composed of two identical interfaces in two languages: English and Chinese. Following the unified content model (Doherty, 1999), the English interface has been translated element by element into the Chinese interface, with the only difference being the order in which the products are sorted. The purpose of using the unified content model was to eliminate any presentation bias in user preferences. The front-end interface is shown in figure 1, which shows two language options (English and Chinese) and two separate applications (Office Supplies and Food Market). Figures 2a-2b show the second-level interface that is invoked once a user selects Figure 1: The front-end interface

Multilingual Electronic Commerce in a Global Economy 3

Figure 2a: Three modes of search in English

Figure 2b: Three modes of search in Chinese

the Food Market application in the English and Chinese versions, respectively. There are three modes of operations that a user can select in order to interact with the system: browsing mode, searching mode, and matching mode. In browsing mode, the user is looking at the products available in the catalog without having any specific item in mind, which is supported by the list box “Select the category:” in Figure 2a. In searching mode, the user is searching for a product class without having a specific item in mind, which is supported by the list box “Select the subcategory:” in Figure 2a. In matching mode, the user has a specific item in mind,

4 Gangopadhyay & Huang

which is supported by the text box “Search by keyword:” in Figure 2a. Once the user selects a category, all products in that category are listed at the next level interface, an example of which is shown in Figures 3a and 3b for the English and Chinese versions, respectively.

Figure 3a: Category selection in English

Figure 3b: Category selection in Chinese

Multilingual Electronic Commerce in a Global Economy 5

When an item is finally selected, the product information is displayed. At this point the user has the option to include the product in the shopping cart, continue to shop, or go back to the initial interface, as shown in Figures 4a and 4b for the English and Chinese versions, respectively. The second level interfaces for the Office Supplies application are shown in Figures 5a and 5b for the English and Chinese versions respectively.

Language Preferences The system described above was used in an experiment to study how bilingual users exhibit language preferences in interacting with system. The details of the experimental design and test results are described in Gangopadhyay and Huang (2000). From these experiments we developed several implications for future research and practice in global electronic commerce. Firstly, users clearly indicated that they preferred to use their ethnic language (Chinese, in this case) when searching for ethnic products because it is difficult to translate them into another language such as English. Hence although the unified content

Figure 4a: Product information in English

Figure 4b: Product information in Chinese

6 Gangopadhyay & Huang

Figure 5a: Office supplies application–English

Figure 5b: Office supplies application–Chinese

model (Doherty, 1999) is easier to implement in a global electronic commerce system, it may not be a good strategy, from a user interface standpoint, for all product categories. Another closely related issue is the level of understanding of the information presented for bilingual users. The experimental results indicate that the subjects performed tasks equally well in both Chinese and English when dealing with structured information. However, when dealing with unstructured information, there was a significant improvement in their performance when the language of interface was Chinese as opposed to English. While it will take more research to establish a relationship between language preference and

Multilingual Electronic Commerce in a Global Economy 7

information complexity, such research can render significant insights into the design of multilingual interfaces for global electronic commerce. In multilingual systems that depend on word-for-word translation from one language to another, a lot of information may be lost during the translation process. For example, it is hard to tell the differences between “table” and “desk” in Chinese. In these cases, images may be helpful to present product information. It is also worthwhile to study the effect of multimedia information on user performance and satisfaction in global electronic commerce.

DESIGN ISSUES IN MULTILINGUAL SYSTEMS One of the major problems in global electronic commerce stems from differences in language and format, disparity in catalogs, lack of flexibility in the way information is organized, and case-to-case translations (Leger et al., 2001). Typically, there are several groups of users interacting with a multilingual system in global electronic commerce: manufacturers, content providers, and users. Manufacturers and content providers add new products, delete and modify existing product information, and maintain articles, descriptions, and other information about products. Users are the customers that interact with the system to gather information, search for specific products, and perform business transactions, including purchases and returns. In a system that supports multilingual electronic commerce, the content providers would like to manage the content in their own language but reach customers from many different nationalities, without having to deploy large translation resources. The users, likewise, would like to interact with the system in their own native language without having to worry about the location of the organization housing the system and their language preferences. There are several functions that have to be supported in this context: 1. Natural language interface: Since both content providers and users interact with the system in their native language, the system contains a natural language processor. In addition to extracting the meaning of the requests made by the users and content providers, the system must also be able to integrate and classify new product information using domain knowledge. 2. Maintaining multilingual catalogs: Since the products and services that are offered must be made available in many languages, an automated language translator must be available to create and maintain catalogs in multiple languages. 3. Business support: In addition to the above, support for special business transactions must be provided in multiple languages. Examples of such transactions include contract negotiation, product liability information, and copyright protection.

Ontology-Based Multilingual Support Most of the recent developments on machine translation and natural language support in multiple languages make use of ontologies for modeling domain knowledge (e.g., Agnesund, 1997, Leger et al., 2000) as well as linguistic knowledge (Mahesh et al., 1995; Bateman et al., 1994; Simons et al., 1995). In addition to natural language support, ontologies are also being used in adding semantics to structured languages such as XML (e.g., Erdmann et al., 2001). The main advantage of using ontologies is that knowledge can be captured in a languageindependent manner. Agnesund (1997) raises two fundamental questions for developing

8 Gangopadhyay & Huang

ontologies: how to choose the concepts in an ontology, and how to relate these concepts to lexical terms. There are several approaches to address these issues. For example, it is possible to choose concepts based on the lexicon of the langauage (Bateman et al., 1994). In this implementation, if a concept has no lexical counterpart, it will not be represented in the ontology. On the other hand, concepts that have lexical counterparts will have a one-to-one mapping. This can lead to the problem that if a concept does not have a lexical representation in a language, then it cannot be translated into it. The other approach is to divide the semantics of a concept between the lexicon and the ontology (Mahesh et al., 1995), where the language-specific knowledge is represented in the lexicon and language-neutral knowledge is represented in the ontology. The problem with this approach is that without a formal basis for choosing concepts the development of the ontology becomes unpredictable as new languages are added to the system. The solution suggested in Agnesund (1997) is to include language-specific knowledge into the lexicon, which then serves as the ontology. In our methodology, choice of concepts does not pose any great challenge since the domain is fairly specific. However, linking concepts with lexical entries in different languages is still a problem, as there may not be a lexical counterpart of a concept in every language that the system supports. We solve this problem by adding a semantic entry to every lexical term. When a term or a phrase is translated from one language into another, instead of direct translation, it is transliterated through the ontology. This can provide a contextual foundation for the translation and can also accommodate situations where a certain word or phrase in one language may not have a counterpart in another language. To illustrate, let us assume that a retailer has a product called “Kung Pao Sauce” that is being sold through a multilingual electronic commerce system. The ontologial entry corresponding to this product would be as shown in Figure 6. With this ontological entry, we can translate “Kung Pao Sauce” into Spanish as “Un listo utilizar sichuanestilo revolver-fríe la salsa,” instead of just “Salsa de Kung Pao,” which would have been a word-for-word translation. Furthermore, if the user searches for “Especias de Sichuan,” or “Sichuan Spices,” the system identifies “Kung Pao Sauce” as one of the possible products that the user is looking for. Once again, the search is facilitated with the help of the ontology.

System Architecture The system for multilingual electronic commerce would typically consist of a natural language processor for each language supported by the system, a machine language translator, an ontology server, and a series of agents to facilitate interactions among the various components. The agents are used for mapping a natural language query into an ontological representation, integrating product and service offerings from multiple vendors Figure 6: An ontology entry Term: Kung Pao Sauce

Semantics: A ready -to-use Sichuan-style stir-fry sauce. Superclass: Sichuan Spices Subclass: None

Multilingual Electronic Commerce in a Global Economy 9

into one composite system, and providing relevant feedback to the end user. One of the major architectural issues in designing such systems is to decide the extent and scope of ontologies used in the system. For example, it is possible to create language specific local ontologies, or language independent global ontologies, or a hybrid architecture. The advantage of having local ontologies is to be able to capture linguistic knowledge as well as emphasize language-specific terms and phrases for which there may not be any counterpart in other languages. Global ontologies, on the other hand, are more effective in capturing languageindependent knowledge, and linguistic domain knowledge is less emphasized. Wache et al. (2001) argue in favor of hybrid ontologies that can be designed to capture both languagespecific as well as language-independent domain knowledge. The system architecture we propose for multilingual support is shown in Figure 7, which is based on the hybrid approach. In this architecture, each source is described by its own ontology. The domain ontology servers combine the local ontologies using a shared vocabulary. Users specify their requirements in any of the languages supported by the system. The natural language processor extracts the intent of the query and passes it onto the appropriate domain ontology server, which then retrieves the result from the corresponding data source through the local ontology. Content providers, on the other hand, specify content specification through the appropriate data source in their preferred language. The local ontology is either created or modified and the shared vocabulary is modified accordingly. The system is currently under development to support five languages: English, Chinese, Spanish, French, and German. Figure 7: System architecture

10 Gangopadhyay & Huang

CONCLUSION In this chapter we have described the design and implementation of multilingual electronic commerce. We discussed experimental results for testing language preferences for users based on the task and product characteristics. A bilingual electronic catalog has been implemented in English and Chinese and two applications have been developed: office supplies and food market. For each application, there are three methods that a subject can use to search for a certain item: browsing, directed search, and exact match using keywords. Through these experiments, we have established a linkage between language preferences and task and product characteristics in using electronic catalogs. We have also described issues in multilingual electronic commerce and suggested a methodology for using ontologies to provide multilingual support. We have described a prototype that is currently under development. Future plans include conducting larger experiments with multilingual electronic catalog and identifying how the differences in language and culture affect the usability of such systems.

ENDNOTE 1 http://www.sybase.com/detail/1,3693,1009232,00.html

REFERENCES Adam, N. R., Dogramaci, O., Gangopadhyay, A. and Yesha, Y. (1998). Electronic Commerce: Business, Technical, and Legal Issues. New York: Prentice-Hall. Agnesund, M. (1997). Representing culture-specific knowledge in a multiligual ontology. In Proceedings of the IJCAI-97 Workshop on Ontologies and Multilingual NLP. Bateman, J., Magnini, B. and Rinaldi, F. (1994). The generalized Italian, German, English upper model. In Proceedings of the ECAI-94 Workshop on Implemented Ontologies, 35-45. Amsterdam. Broin, U. Ó. (1999). International aspects of user interface design. MultiLingual Computing & Technology, 10(3), 50-54. Computer Economics. (2001). English will Dominate Web for Only Three More Years. Retrieved from the World Wide Web: http://www.computereconomics.com/new4/pr/ pr990610.html. Doherty, W. (1999). Creating multilingual Web sites. MultiLingual Computing & Technology, 10(3), 34-37. Erdmann, M. and Studer, R. (2001). How to structure and access XML documents with ontologies. Data and Knowledge Engineering, 36, 317-335. Gangopadhyay, A. and Huang, Z. (2000). Online user interaction with electronic catalogs: Language preferences among global users. Journal of Global Information Management (Special Issue on Knowledge Management), 8(3), 16-23, July-September. Leger, A., Lehtola, A. and Villagra, V. (2001). MKBEEM—Developing multilingual knowledge-based marketplace. ERCIM News, July, (46). Mahesh, K. and Nirenburg, S. (1995). A situated ontology for practical NLP. In Proceedings of the IJCAI-95 Workshop on Basic Ontological Issues in Knowledge Sharing, 19-21, Montreal.

Multilingual Electronic Commerce in a Global Economy 11

Malhotra, Y. (1997). Knowledge management in inquiring organizations. In the Proceedings of 3rd Americas Conference on Information Systems (Philosophy of Information Systems Mini-track), 293-295. Indianapolis, IN, August 15-17. Malhotra, Y. (1998). TOOLS@WORK: Deciphering the knowledge management hype. Journal for Quality & Participation (Special issue on Learning and Information Management), 58-60. July-August, 21(4). Nirenburg, S. (1992). Text planning with opportunistic control. In Kittredge, R. (Ed.), Machine Translation Special Issue on Natural Language Generation, 7(1-2), 99-124. Onyshkevych, B. and Nirenburg, S. (1995). A Lexicon for knowledge-based MT. Machine Translation, 10(1-2). Segev, A., Wan, D. and Beam, C. (1995). Designing electronic catalogs for business value: Results of the commercenet pilot. Technical Report. The Fisher Center for Management and Information Technology, October, Haas School of Business, University of California, Berkeley. Simons, G. and Thomson, J. (1995). Multilingual data processing in the CELLAR environment. Paper presented at Linguistic Databases, March 22-23, University of Groningen, Centre for Behavioral and Cognitive Neurosciences. Sheremetyeva, S. O. and Nirenburg, S. (1996). Empirical modeling in computational morphology. Nauchno-Technicheskaja Informacija (Scientific and Technological Information), Series 2, Issue 7. Wache, H., Vögele, Visser, U., Stuckenschmidt, H., Schuster, G., Neumann, H. and Hübner, S. (2001). Ontology-based integration of information—A survey of existing approaches. Submitted to IJCAI-01 Workshop on Ontologies and Information Sharing. Retrieved from the World Wide Web: http://www.informatik.uni-bremen.de/~heiner/ paper-pages/ois01a.html.

12 Galliers & Newell

Chapter II

Electronic Commerce and Strategic Change within Organizations: Lessons from Two Cases Robert D. Galliers London School of Economics, UK Sue Newell Royal Holloway, University of London, UK

This article reviews and contrasts the experiences of two major companies in attempting significant change projects incorporating information and communication technologies. It does so by utilizing and critiquing the MIT “Management in the 1990’s” model and by reflecting on socio-technical approaches to organizational change. It makes the point that while much of current attention is on electronic commerce as it pertains to industry transformation and interorganizational relations, it is nonetheless a phenomenon that can impact complex internal relations and communication. Additionally, conclusions are drawn with respect to the process of change and the need for further longitudinal studies when researching change projects of this kind.

INTRODUCTION It goes without saying that electronic commerce has been a major topic of interest in recent years, with considerable importance being placed on the opportunities provided by information and communication technologies (ICTs) to improve coordination between businesses and with customers. Electronic commerce resonates as a potential means of finding solutions to some of the inter-organizational communication issues that confront modern-day businesses. However, with globalization, companies are themselves increasAppeared in Journal of Global Information Management, vol. 9, no. 3, 2001. Reprinted by permission.

Electronic Commerce and Strategic Change within Organizations

13

ingly widely dispersed geographically so that internal communications are also more difficult and complex. ICTs can therefore also potentially improve communication between individuals and groups within an organization. This suggests that we might usefully broaden and unpack the concept of electronic commerce to include intraorganizational collaboration and partnerships. For example, corporate intranets, ERP systems, corporate databases, and even internal email systems are all examples of electronic commerce that are increasingly used within a company. These ICTs enable data, knowledge, and information to be shared even when the individuals involved are widely distributed. However, while intraorganizational electronic commerce can provide a communication link between people who are functionally, hierarchically, or geographically separated, this does not necessarily mean that the social and psychological barriers between groups will be broken. The cases reported in this paper demonstrate very clearly that technology per se will not automatically improve communication between groups where long-standing barriers exist. The importance of breaking down functional and hierarchical barriers has long been recognized in the information systems world. In academia, our roots in systems thinking provide us with a conceptual base for viewing organizational issues from a process orientation, with information requirements being associated with those activities necessary to achieve a desired purpose (e.g., Checkland, 1981). This so-called “infological approach” has been paralleled by what might be called a datalogical perspective, where the focus is on an analysis of the data entities–and their relationships–required to provide necessary information (Martin, 1982). These concepts have found practical form in such phenomena as database technologies and approaches, and more recently in business process reengineering (BPR), knowledge management, and knowledge management systems (KMS) (see, for example, Davenport, 1993; Alavi & Leidner, 1999). This paper describes research concerning two companies that have been seeking to improve collaboration and communication internally across functional and departmental boundaries through the use of ICT. In one case the development and introduction of the ICTbased system leads to unintended, negative effects; in the other, there is preliminary evidence to suggest that the results have been much more positive. The experiences of the two companies help to reinforce lessons that have been known for some time,1 as well as provide new insights. The fact that these two cases are contemporary, and that there appears to be evidence that some of the lessons of the past have been forgotten or have remained unheard, suggests that the comparison may be enlightening.2 The paper is structured as follows. Following this introduction, we will describe the methodology followed in the research upon which the case studies are based. Then we will briefly outline important aspects of each case, the first concerning the experiences of a major multinational bank in implementing KMS, and the second describing the somewhat more emergent and holistic approach adopted by a major international airline. In the next section, we will attempt to extract from the two cases those features that appear to be key. The concluding section provides something of a reflection on more general lessons, in terms of what remains to be done in the current research effort and potentially useful directions for future research in this area.

METHODOLOGY Qualitative methods were used to explore the impact of ICTs on intraorganizational processes since these methods allow the researcher to examine the phenomenon of interest in its natural setting. Case studies were considered most appropriate, since they allow for the

14 Galliers & Newell

adoption of multiple data collection methods (Yin, 1984), which was thought to be important in order to develop the rich case descriptions needed to build theoretical understanding. Case studies generate insightful stories, rather than statistical information, and permit a better understanding of organizational complexity from an insider’s viewpoint (Mitchell, 1983), allowing the researcher to formulate a more holistic perspective on the studied phenomenon (Van Maanen, 1979). The cases reported in this paper were drawn from a larger pool of cases that have focused on investigating innovation processes in both manufacturing and service organizations across Europe. These two cases have been selected because they allow some interesting comparisons in relation to intra-organizational electronic commerce. Exploring the contextual similarities and differences between the cases is useful for examining their impacts on intra-organizational electronic commerce (Orlikowski, 1993). Methods of data collection included interviews, on-site observation, and documentation. Adopting multiple methods is important not only to enhance the richness of the findings, but also to ensure validity through the process of triangulation (Stake, 1995). The research in each case was conducted over a period of approximately 18 months. This included four visits to the bank during this period and two to the airline. The interviews, in both cases, were conducted primarily with project team members and sponsors and included individuals from various hierarchical levels and divisions. The interviews were conducted face-to-face, using a semi-structured interview schedule, and were tape recorded and later transcribed. The interviews varied in length, but most were between one-and-a-half to two hours. The on-site observation took place during visits to the companies and allowed informal conversations with project members during coffee and lunch breaks. In the bank, a two-day workshop was attended which greatly increased the opportunity for informal as well as formal research material to be collected. This allowed the researchers to make sense of the situation from an insider’s point of view (Evered & Louis, 1981). The on-site interviews at the airline were supplemented by telephone discussions, and further clarification regarding specific issues was provided via e-mail communication. Documents, such as minutes, reports, and intranet sites related to the projects were used to further enhance the richness of the material collected. Observation of subsequent developments has taken place through study of media articles, press releases, and the like. As is typical in inductive studies, writing the case studies was an iterative process in which the data were constantly revisited. Triangulation across the different sources of primary and archival case material revealed a high level of consistency, so that the “story” described in each case can be said to be valid. While there were no preconceived hypotheses at the outset of the inquiry, patterns emerged from the data which suggested the potential for using the MIT model (Scott Morton, 1991, p. 20) to compare the two cases in relation to their electronic commerce projects (see below).

CASE 1: THE BANK The bank has around 70,000 employees and operates in approximately 70 countries worldwide, with its headquarters in Europe. The bank was formed from a merger of two banks from the same country and has subsequently grown via the acquisition of banks in the various countries in which it now operates. ICT has always been an important priority in the bank, as evidenced by the fact that it is the fifth highest spender on IT in Europe. Its structure and functioning is highly decentralized, with resources allocated to the independent divisions and with very few resources retained at the center. It consists of a number of different product divisions including domestic, international, and investment banking.

Electronic Commerce and Strategic Change within Organizations

15

A problem that was exercising the minds of top management in this bank was that they had recently lost a key account thanks to the perceived inability of the bank to adopt similar procedures and provide similar services in the countries in which the global client company was operating. Exasperated by the bank’s apparent inability to present a “common face” worldwide (despite this being a stated strength in its marketing literature), the client company took its business to a rival. The bank’s reaction to this event was to set up a pilot intranet project. This pilot project was a key part of its “Vision 2000” strategy, which was aimed at creating “The Networked Bank.” The strategic intent of this project was thus to create a network across the bank so that knowledge and information could be shared more effectively across functions and geographically dispersed sites. This, it was assumed, would lead to the common adoption of defined “best practices” and so stimulate the integration of procedures and services across the bank. However, in direct contrast to the stated strategy, the actual impact was that the existing boundaries between functions and dispersed business units were reinforced. This was because, during the 18 month life of the pilot, in excess of 150 known intranets had been set up by individual departments in individual countries. Once this became evident, the bank called together representatives (both banking and IT) from its major national sites worldwide with a view to investigating how to coordinate this emerging web of intranets and achieve the stated strategic objectives. The idea was that this might help reduce response variety in the different countries. The meeting took place over two days at the bank’s headquarters. Two problems emerged almost instantaneously. First, the banking representatives were “too busy” to attend both days - having attended the first “strategic focus” day, they left their IT colleagues to “sort out the technical details” on the second day. Second, the IT representatives on the second day focused exclusively on technical solutions and came to the conclusion that the way forward was to create a corporate portal through which individuals could navigate the myriad of intranets. However, these national IT representatives were so energized as a result of these discussions that many of them set about designing their very own “corporate” portal. Within ten days the Bank was the “proud owner” of six or seven “corporate” portals, each with its own characteristics and idiosyncracies–and there were more on the way. In other words, the bank’s knowledge management system (KMS) turned out to be many KMSs and the original objective of presenting a common client interface was lost. Additionally, while making available a considerable quantity of apparently usable and useful data, few of the bank’s employees found their KMS to be particularly useful in their dealings with clients–or anything else for that matter. In one department in one country, and this appeared to be the most established of the intranets, the best that anyone could come up with when asked how it was being used was “to look up the company bus timetable.” A key problem in this case was that the vision of a “global” bank was in stark contrast to the existing culture and structure of the bank. The bank had grown largely through acquisition and merger, and each of these acquired banks had been left very much to operate using their home-grown procedures, offering the particular services they had historically provided. There had been no attempt to standardize, and indeed the culture of decentralization was built into the distribution of resources and the performance management measures used. Thus, when the pilot intranet project was started, each nationally-located bank and department recognized the potential usefulness of this technology for improving its own internal efficiencies. However, there was no reason for each to consider the potential of the technology for communicating and sharing information and knowledge across the existing internal boundaries, since there was no real incentive for them to do so. Thus, while all were aware of the stated vision, and indeed during interviews related their own intranet initiatives

16 Galliers & Newell

to this globalizing vision, each unit was most concerned with using its own resources for establishing internal efficiencies. Collaborating with other units would take time and resources away from this focus, at least in the short term. The result was that a strategy aimed at coordinating knowledge and information across the bank actually resulted in a great deal of reinvention, as each unit developed its own applications for sharing across its own uniquely developed intranet. There were examples given during interviews of how significant sums of money (c. $500,000) had been spent on developing intranet applications for knowledge sharing in one department, only for those involved to later find that similar sums had been spent by other units developing very similar if not identical applications elsewhere.

CASE 2: THE AIRLINE The second case is of one of the world’s largest international passenger airlines based in Europe. It had 356 aircraft in December 1999, operating over one thousand flights a day, and flying 45 million passengers in 1998-99. It has approximately 64,000 staff worldwide, 80% of them based in Europe. Originally a publicly-owned company, it was privatized in 1987. It is now owned entirely by private investors with around 265,000 shareholders, including 71% of the company’s own employees. The airline market has become increasingly fierce and had led to a drop in profits for the company in both 1998-99 and 1999-2000. The company’s response was to target profitable passenger segments and to work more closely with other airline companies in global alliances. These partnerships give the airline a presence in all major world markets. The issue confronting this airline was that many of its central departments were housed in different buildings with consequent inefficiencies and complex communications. A new headquarters was therefore planned, presenting an opportunity to upgrade systems and communications, utilizing the very best in modern ICT, and to house the c. 3000 employees involved. A project team was set up, with a representative from each of the departments affected by the proposed office move. The chair of the project team reported to a steering committee comprising senior executives and chaired by a main board member. The initial major objective was for the move–a $320m investment–to reap annual savings of at least $25m through streamlined procedures, reduced paper usage, and reductions in headcount. The project was quickly perceived (by both the steering committee and the project team members) as representing an opportunity not only to save money but to improve communications across departmental boundaries and to enable creative, innovative thinking to take place. This did result in significant savings with, for example, 89 tons of paper having been saved in the first five months of operation in the new building. In addition, however, further benefits ensued. Primarily a partnership between those responsible for office design and the IT and Human Resource Management departments, the project’s scope soon expanded to include new ways of working, streamlined procedures, and extensive training programs for all involved in the move–including the CEO. Everyone was expected to participate, and ideas about potential innovations were actively sought: the very process of preparing for the move set the tone for the new working environment, with knowledge sharing and knowledge creation being both encouraged and rewarded. Those involved recognized that the potentially disruptive and unsettling experience of a major office move, combined with new procedures, new technology and systems, not to mention downsizing, could be perceived in a much more positive light, were the approach to be inclusive and participative. Furthermore,

Electronic Commerce and Strategic Change within Organizations

17

excitement and pride might actually be generated in looking forward to the new opportunities available in the new working environment. The fact that the media also took considerable interest in the project added to the sense of expectancy being generated by the move. Communication was seen to be key throughout and this was achieved through face-to-face briefings, discussion groups, training programs, videos, newsletters, and the like. The new headquarters was designed to enable informal, ad hoc meetings to take place in, for example, coffee bars located on either side of the main thoroughfare. Individual offices and meeting rooms are few and far between (even the CEO did not have his own office), with “hot desking” being the norm, and mobile cell-phones being used to ease mobility and communication. The intranet is used extensively, for example, for booking travel (and for claiming the related travel expenses), for gleaning information on both corporate and individual customers (either in response to inquiries or proactively for direct marketing purposes), and for communicating with company offices and major client organizations worldwide. It is too early to judge the longer term benefits (and drawbacks) of the move, which has not been without its teething problems, of course. For example, the hot desking concept has not been entirely successful, with individuals deliberately leaving their belongings behind on “their” desk in an attempt to reserve their own spaces. In addition, others have been somewhat unsettled by the lack of a “home,” given their need to “belong,” while the continuing need to drive down costs even further has raised tensions. From the systems perspective, it is clear that further modifications to user interfaces may be required since infrequent use leads to forgetfulness, which in turn has a tendency to raise frustrations about “the technology.” Further, the initial euphoria attached to the move to the new building, which, as already indicated, had attracted considerable media interest and critical acclaim, died down considerably over time. This perfectly natural decline in enthusiasm was exacerbated, for example, by adverse market and media reaction to certain of the airline’s highprofile strategic decisions and the subsequent replacement of the CEO. Despite this and some of the problems mentioned above, many of the sought-after benefits have been achieved, and communication appears to be more open and streamlined, with informality having been retained, despite the new formal procedures. Morale remains high, despite the company’s somewhat disappointing financial results and the continuing intense competition it faces.

REFLECTIONS: KEY ISSUES While considerably more research has to take place in both organizations in order properly to reflect on the widely differing experiences of the two companies, some initial comments can be made. The two cases are clearly rather different, but in both the focus was on using ICT to facilitate communication and knowledge sharing across their dispersed operations. In other words, both organizations were attempting to use electronic commerce to transform intraorganizational relations, in particular by encouraging greater coordination and collaboration across departments and business units. In both cases, top management was committed to change and saw each project as being strategically important from the point of view of the business. While both change projects were to be enabled by ICT, both were essentially “business-driven.” In the case of the bank, however, it would seem that the project was initially conceived, and remained, one of superimposing an ICT on to an existing organizational structure, with little in the way of changed work practices. For example, departmental boundaries and

18 Galliers & Newell

authority remained untouched, with each department remaining responsible for its own “slice of the action.” An ICT-based KMS was perceived to be the answer, with it being “parachuted” on top of the existing organization. Making apparently useful data available for everyone would lead to the common client interface that was considered essential to retain key customers. For the airline, what was initially an office move, conceived for the purposes of cost reduction and greater efficiency, soon became an opportunity radically to streamline office procedures, reduce paperwork, and improve communication–both internally and externally. What emerged was what we might view as a socio-technical design for the new building and the people in it. Perhaps the irony in this comparison is that while the bank actively sought improved customer relationship management and internal communication, the airline achieved it without planning to do so. We might characterize the bank’s approach as a case of “top-down,” business-driven planning, while the airline’s management gave their project team broad guidelines (and tough financial targets) but the freedom to be creative and the ability to allow ideas to emerge from knowledge sharing. It has often been said that it is easier to effect change when there is a sense of crisis. Paradoxically, however, it was the bank that saw itself as facing a crisis, while the airline simply wanted to become more competitive by being more efficient–seeing the issue as simply the on-going business of competing in what is a cut-throat industry. In addition, it is clear that while the bank expected that the mere provision of a KMS would lead to the improvements they sought, the airline took considerable steps to inform, motivate, and energize those concerned, both before the move and afterwards. Utilizing the MIT “Management in the 1990’s” model (Scott Morton, 1991), we can map the alternative approaches adopted by the two case companies. Figure 1 illustrates the MIT model in outline, while Figure 2 provides a comparison of the paths followed by the bank and the airline. Essentially, the original model attempts to illustrate the nature of strategic change and fit. The model emphasizes (after Leavitt, 1965) that a change in any of the five elements will have an impact on the other four elements. It also reminds us that there is more to the management of change than the early advocates of business process reengineering would have had us believe (Galliers & Swan, 1999), given their emphasis on process innovation on the back of ICT alone (e.g., Davenport, 1993). While Figure 1 might be seen as an idealized version of an holistic approach to organizational change and fit, Figure 2 represents the stark reality of the approaches adopted in the two case companies.

Figure 1: The MIT model of strategic change and fit Structure (intra- & interorganizational)

Strategy

Management Processes Individuals (roles, values, capabilities …)

Socio-, economic, technological environment

Technology

Electronic Commerce and Strategic Change within Organizations

19

The model in Figure 1 depicts all the elements as being interrelated, but in practice the dominant assumption has been that technology must be aligned to the established strategystructure fit (Scott Morton, 1991). In other words, a firm must first develop its strategy, then design the structure to support this strategy, and finally build new management processes, align IT, and ensure that appropriate roles are created and individuals are trained (McGee & Prusak, 1993). Comparing the two cases against these normative assumptions, it is clear that neither case company operated according to this “best practice” prescription. Essentially the bank’s approach was little more than identifying a technological solution to a perceived business problem. It was assumed that, having articulated the strategy, all that was necessary was to align the selected ICT with this strategy. Unfortunately, this ignored the existing management processes and individual roles and skills, as well left the organizational structure unchanged. Recall, too, that the individual national banks were essentially operating in opposition to the newly defined strategy. Hence, the approach ignored the organizational environment, let alone the wider socio-economic environment. In this situation, as the MIT model would suggest, the impact of the ICT (in this case intranet technology used as a KMS) was limited and certainly did not result in the strategic changes intended. In contrast, the airline, taking into account the socio economic environment in which it was operating, initially converted its strategy (i.e., to improve efficiency and cut costs by moving disparate functional units into a single location) into required changes in structure and business processes. Shortly after its formation, members of the project team saw the wider opportunities afforded by the move, more specifically in terms of a new working environment, enabled by ICT (in this case an intranet/KMS), and the pressing need to develop new roles and skills. Importantly, the development of the technology support for the new strategy as well as the roles and management processes were much more emergent than is suggested by the MIT model. There was never any explicit attempt to align the ICT available with the strategy. What is more, and this is not adequately illustrated by the MIT model, the process by which they went about the change project was key, with participation and inclusiveness being the watchwords. In other words, the process of IT implementation management in the airline appeared to be crucial to the success experienced, especially in terms of involving potential users as widely as possible and letting solutions emerge and

Figure 2: The two approaches compared

Environment Structure Stru cture

Strategy

Management Processes

Stra teg y

Technology

Ma nag ement P ro ces ses

Indiv iduals

Individuals

Environment

Technolo gy

20 Galliers & Newell

evolve as individuals interacted and learned about the available technologies. Similarly, Yetton (1997) argues that successful IT management is often more evolutionary than the strategic alignment approach advocated.

CONCLUSIONS AND FUTURE RESEARCH In many ways, the message from this comparison is not new–i.e., that we should not expect business improvements to come from ICT alone, but we should see such change projects as being essentially about human beings, about social systems … albeit with a technological component. After all, this has been the mantra of the socio-technical school for the past 50 years or so. Having said that, the fact that we can observe such differing experiences in implementing change on the back of ICT at the beginning of the 21st century suggests that much has still to be learned in this respect. This is perhaps particularly true now, given that such flexible ICT as intranet technology is both commonplace and accessible, and its promise is well known among even the most troglodyte of business executives. In selling solutions as the IT industry is prone to do, it is not unreasonable to expect that executives might be lulled into a false sense of security when investing in those solutions. We have seen here that a business-driven, top-down approach, even with top management commitment, may well be necessary but not sufficient to effect change that brings about business benefits. We have also seen that a broader conception of an IS change project (taking into account people and their modes of working and communicating) seems more likely to bring about beneficial results–even unexpected ones. Creating the conditions for emergence and serendipity appears to hold considerable promise in this context. The related concepts of “bricolage” (Ciborra, 1992) and improvisation (Orlikowski, 1996; 1997) have recently been introduced in the context of ICT management, and the airline case certainly appears to support the argument of letting ICT applications evolve as people use and experiment with them. The use of the term “evolve” is both deliberate and important in this context. Often, debate in the field of business strategy has focused on the two schools of thought concerned with emergent and deliberate strategies (cf., Mintzberg, 1989). This finds form in the IS literature, in addition (e.g., Orlikowski, 1996). In both cases considered here, however, the strategy was evidently deliberate and emanated from “the center.” In the case of the bank, the strategy was limited to the development of a technological solution imposed on an “alien” culture. It therefore failed, at the time of writing at least, to bring about the desired results. Conversely, in the case of the airline, a more broadly based, but nonetheless deliberate, strategy was formulated. In contrast to the bank, however, an appreciative climate was fostered, enabling greater acceptance of the proposed strategy. Additionally, a climate of mutual learning developed, thus enabling the strategy to evolve in the light of experience. While the findings here support a socio-technical view of organizational change, they also take us beyond this, since even here the process of change has been underexamined. For example, both companies involved multifunctional project teams in implementing the change programs. However, there was a significant difference in how these teams operated (i.e., in their processes). In the bank, the different functions worked essentially independently, even when they were supposedly working on a joint project. This was demonstrated most clearly at the two-day workshop. Knights and Wilmot (1997) refer to this as mechanistic pooling, which they define as occurring where “each member of the pool takes a different ‘slice’ of the project and the work then proceeds with the minimum of communication between

Electronic Commerce and Strategic Change within Organizations

21

its members” (p. 19). In the airline, on the other hand, the project team was much more clearly engaged in joint knowledge production, where the knowledge from the different functional specialists was shared and integrated to produce new solutions which were accepted by those involved. These micro-level processes appeared to be essential to the relative success of the two cases but are not captured by static and more messo-level models of organizational change. The MIT model may provide a useful starting point for examining the impact of intraorganizational electronic commerce, but what is now needed is longitudinal research that tracks and monitors the messy and sometimes tortuous process by which this change takes place. In this vein, ongoing research in the two case study companies is taking place with a view to ascertain more clearly the repercussions of the two initiatives, over time, both in terms of intended and unintended effects. We cannot assume that the only outcomes in the case of the bank are entirely negative, nor that in the case of the airline, they are entirely positive. Pennings (1998), for example, has highlighted the lagged effects of ICT investments, and Pettigrew (1990) counsels longitudinal research in the context of change projects. While the research described here was clearly longitudinal (being conducted over a period of 18 months), the projects which are being considered here are ongoing in the sense that in neither case has the technology been fully appropriated. Appropriation occurs when a new technology that has been implemented has become routine (Clark & Staunton, 1989). With complex technologies implemented in large global companies, this is likely to take a very long time, if indeed it ever happens. So, here we have considered the initial impacts only of two ICT implementation projects. However, with the pressure on academics to publish, and with the limits to research funding available, it is often very difficult to conduct longitudinal research over extended periods, covering the entire innovation process from agenda formation (when the idea for implementing a particular new technology is first tabled) through to appropriation. This means that many of our research findings are likely to provide only a snapshot (or at least only a small movie extract) of the ICT management process. Given the importance noted here of emergence and evolution, this suggests a problem for those interested in this area and poses a significant challenge for IS research more generally. Discussion and debate as to how to resolve this problem is urgently needed so that we prevent oversimplification of ICT-based change which tends to highlight the failures rather than the successes. Successes are likely to take considerable time to emerge and not be captured by the snapshots that commonly get taken through the research process conducted in real time. On the other hand, retrospective accounts suffer from post-hoc rationalizing and justification so that much of the emergent and messy process of ICT-based change is concealed (Lanzarra, 1999). There is some additional learning that might potentially be gleaned from this comparison. The lack of a cumulative tradition in IS has been highlighted and criticized for some time (e.g., Keen, 1980). More recently, and in a similar vein, one of the authors of this paper has expressed his concern regarding our propensity in the IS field to study emerging phenomena in isolation. In doing so, he highlighted, as examples, KMS and electronic commerce at this point in the development of our subject area (Galliers, 1999). It would seem, certainly on the face of it, that the experience of both the airline and the bank would lend some weight to further research in which electronic commerce and KMS are considered as related phenomena in the context of the kind of strategic change reported on here. Finally, in relation to intraorganizational electronic commerce per se, there is an important difference between the two cases that needs to be considered. Specifically, while the bank relied wholly on a virtual space (the intranet) to facilitate improved internal

22 Galliers & Newell

communication and collaboration, the airline incorporated an improved physical as well as virtual space as part of its change program. The new building designed by the airline not only colocated individuals from the different functional areas but also created places where people could meet, both formally and informally. Nonaka and Konno (1994) develop the concept of “Ba” to highlight the need for “space” in the knowledge creating company. While some of these “Bas” are virtual, others require a shared physical space. It is tempting to argue that the airline was relatively more successful precisely because its change strategy that encouraged greater interdepartmental collaboration, perhaps serendipitously, gave employees access to both a virtual and a physical space in which to interact with colleagues. This suggests that intraorganizational electronic commerce will need to be carefully integrated with face-to-face opportunities for interaction for the benefits of such virtual communication to be more fully exploited.

ENDNOTES 1 Through, for example, the work of those associated with the socio-technical school of thought (e.g., Mumford, 1983; 1987), particularly in relation to Information Systems (IS) failures (e.g., Bostrom & Heinen, 1977a,b) and a more balanced, non-technologically deterministic and emanicapatory approach to IS development and implementation. (e.g., Bjerknes et al., 1987). 2 The fact that a new book (Coakes et al., 2000) has recently been published on new thinking and contemporary accounts of the socio-technical approach suggests that we are not alone in this belief.

REFERENCES Alavi, M. and Leidner, D. E. (1999). Knowledge management systems: Issues, challenges, and benefits. Communications of the Association for Information Systems, 1(7), February. http://cais.isworld.org/articles/1-7/article.htm. Bjerknes, G., Ehn, P. and Kyng, M. (Eds.). (1987). Computers and Democracy: A Scandinavian Challenge. Aldershot, UK: Avebury. Bostrom, R. P. and Heinen, J. S. (1977a). MIS problems and failures: A socio-technical perspective, Part 1: The causes. MIS Quarterly, 1(3), September. Bostrom, R. P. and Heinen, J. S. (1977b). MIS problems and failures: A socio-technical perspective, Part 2: The application of socio-technical theory. MIS Quarterly, 1(4), December. Checkland, P. B. (1981). Systems Thinking. Systems Practice. Chichester, UK: Wiley. Ciborra, C. U. (1992). From thinking to tinkering: The grassroots of strategic information systems. The Information Society, 8, 297-309. Clark, P. and Staunton, N. (1989). Innovation in Technology and Organization. London, UK: Routledge. Coakes, E., Willis, D. and Lloyd-Jones, R. (Eds.). (2000). The New Socio-Tech: Graffiti on the Long Wall. London, UK: Springer-Verlag. Davenport, T. H. (1993). Process Innovation: Reengineering Work through Information Technology. Boston, MA: Harvard Business School Press.

Electronic Commerce and Strategic Change within Organizations

23

Evered, R. and Louis, M. R. (1981). Alternative perspectives in the organizational sciences: “Inquiry from the inside” and “inquiry from the outside.” Academy of Management Review, 6, 385-395. Galliers, R. D. (1999). Towards the integration of e-business, knowledge management and policy considerations within an information systems strategy framework. Journal of Strategic Information Systems, 8(3), September, 229-234. Galliers, R. D. and Swan, J. (1999). Information systems and strategic change: A critical review of business process re-engineering. In Currie, W. L. and Galliers, R. D. (Eds.) Rethinking Management Information Systems: An Interdisciplinary Approach, 361-387. Oxford, UK: Oxford University Press. Keen, P. G. W. (1980). MIS research: Reference disciplines and a cumulative tradition. Proceedings: 1st International Conference on Information Systems,9-18. Philadelphia, PA, December. Knights, D. and Wilmott, H. (1997). The hype and hope of interdisciplinary management studies. British Journal of Management, 8, 9-22. Lanzarra, G. (1999). Between transient constructs and persistent structures: Designing systems in action. Journal of Strategic Information Systems, 8(4), 331-350. Leavitt, H. J. (1965). Applying organizational change in industry: Structural, technological and humanistic approaches. In March, J. G. (Ed.), Handbook of Organizations. Stokie, IL: Rand McNally. Martin, J. (1982). Strategic Data Planning Methodologies. Englewood Cliffs, NJ: PrenticeHall. McGee, J. and Prusak, L. (1993). Managing Information Strategically. New York, NY: Wiley. Mintzberg, H. (1989). Strategy formation: Ten schools of thought. In Fredricson, J. (Ed.), Prospectus on Strategic Management. New York, NY: Ballinger. Mitchell, J. C. (1983). Case and situation analysis. Sociological Review, 31, 186-211. Mumford, E. (1983). Designing Human Systems: The ETHICS Approach. Manchester, UK: Manchester Business School. Mumford, E. (1987). Sociotechnical Systems Design: Evolving Theory and Practice. In Bjerknes et al. (Eds.), 59-76. Nonaka, I. and Konno, N. (1998). The concept of ‘Ba’: Building a foundation for knowledge creation. California Management Review, 40(3), 40-54. Orlikowski, W. J. (1993). CASE tools as organizational change: Investigating incremental and radical changes in system development. MIS Quarterly, 17, 309-340. Orlikowski, W. J. (1996). Improvising organizational transformation over time: A situated change perspective. Information Systems Research, 7(1), 63-92. Orlikowski, W. J. (1997). An improvisational model of change management: The case of groupware technologies. Sloan Management Review, Winter. Pennings, J. (1998). Innovations as precursors of organizational performance. In Galliers, R. D. and Baets, W. R. J. (Eds.). Information Technology and Organizational Transformation: Innovation for the 21st Century Organization, 153-178. Chichester, UK: Wiley. Pettigrew, A. M. (1990). Longitudinal field research on change: Theory and practice. Organization Science, 1(3), 267-292. Sauer, C., Yetton, P. W. & Associates. (1997). Steps to the Future: Fresh Thinking on the Management of IT-Based Organizational Transformation. San Francisco, CA: Jossey-Bass. Scott Morton, M. S. (Ed.). (1991). The Corporation of the 1990s: Information Technology and Organizational Transformation. New York, NY: Oxford University Press.

24 Galliers & Newell

Stake, R (1995). The Art of Case Study Research. London, UK: Sage. Van Maanen, J. (1979). Reclaiming qualitative methods for organizational research. Administrative Science Quarterly, 24, 520-526. Yetton, P. W. (1997). False Prophecies, Successful Practice, and Future Directions in IT Management, 27-54. Sauer, Yetton & Associates. Yetton, P. W., Johnston, K. D. and Craig, J. F. (1994). Computer-aided architecture: A case study of IT and strategic change. Sloan Management Review, 35(4), 57-67. Yin, R. K. (1984). Case Study Research–Design and Methods. London, UK: Sage.

Trust in Internet Shopping 25

Chapter III

Trust in Internet Shopping: Instrumental Development and Validation through Classical and Modern Approaches Christy M. K. Cheung and Matthew K. O. Lee City University of Hong Kong, Hong Kong

Despite the phenomenal growth of Internet users in recent years, the penetration rate of Internet shopping is still low, and one of the most often cited reasons is the lack of consumers’ trust (e.g., Hoffman et al., 1999). Although trust is an important concept in Internet shopping, there is a paucity of theory-guided empirical research in this area. In this paper, a theoretical model is proposed for investigating the nature of trust in the specific context of Internet shopping. In this model, consumers’ trust in Internet shopping is affected by propensity to trust and two groups of antecedent factors, namely, “trustworthiness of Internet vendors” and “external environment.” Trust, in turn, reduces consumers’ perceived risk in Internet shopping. As an important step towards the rigorous testing of the model, the necessary measurement instrument has been developed with its reliability and validity empirically tested. The psychometric properties of the measurement instrument have been investigated using both a classical approach (based on Cronbach’s alpha and exploratory factor analysis) and a contemporary approach (based on structural equation modeling techniques), as a way of methods triangulation for validating instrument properties. The resulting instrument represents a rigorously developed and validated instrument for the measurement of various important trust-related constructs. This research contributes to the development of trust theory in e-commerce and add to the repository of rigorous research instruments for IS survey researchers to use. Appeared in Journal of Global Information Management, vol. 9, no. 3, 2001. Reprinted by permission.

26 Cheung & Lee

INTRODUCTION The growth of the Internet and its user base in recent years has been truly phenomenal. As of November 2000, the estimated number of people online has exceeded 407 million1 and the annual growth rate remains high. Enormous potential therefore exists for the use of the Internet for the purchase of goods and services (“Internet Shopping”). However, this potential remains largely untapped, as market surveys2 have confirmed that many Internet users are still reluctant to make purchases on the Internet. One of the most often cited reasons for consumers not purchasing from Internet shops is the lack of trust, which stops or discourages consumers from entering into exchange relationships with Internet shops (e.g., Hoffman et al., 1999). Despite the importance of trust in consumer-based electronic commerce, little theory-guided empirical research has been undertaken to understand the nature of trust, its antecedents, and its consequences in the specific context of Internet shopping. The few research papers available (e.g., Jarvenpaa et. al., 1999) tend to focus on very small models, ignoring many potentially important constructs (e.g., trust propensity, privacy, and security) suggested by the rich but distant literature on trust. Thus, more research is called for. In addition, the study of e-commerce trust has been hampered by a lack of validated measurement instruments in the literature. As a step towards bridging this gap, this paper proposes a research model of trust in Internet shopping and presents the development of an empirically validated measurement instrument for testing the trust model. This paper is presented as follows. The first section reviews literature on trust. Then we present the research model and propositions and describe the instrument development, validation process, and results. Finally, conclusions of the findings and suggestions for future research are provided.

TRUST The notion of trust has been examined under various contexts over the years, for example, in bargaining (Schurr & Ozanne,1985), industrial buyer-seller relationships (Doney & Cannon, 1997), distribution channels (Dwyer, Schurr & Oh, 1987), partner cooperation in strategic alliances (Das, 1998), and the use of market research (Moore et al., 1993). Different theoretical perspectives have been used in these studies, which may be aggregated into three categories (Lewicki & Bunker, 1995): 1. The views of personality theorists, conceptualizing trust as a belief, expectancy, or feeling which is deeply rooted in the personality and has its origins in the individual’s early psychological development. 2. The views of sociologists and economists, conceptualizing trust as a phenomenon within and between institutions, and as the trust individuals put in those institutions. 3. The views of social psychologists, characterizing trust in terms of the expectation and willingness of the trusting party engaging in a transaction, the risks associated with and acting on such expectations, and the contextual factors that serve to either enhance or inhibit the development and maintenance of that trust. Although the social-psychological perspective appears to be most relevant in the understanding of consumer trust in Internet shopping as this perspective focuses on transactions, the other perspectives also contribute to our understanding of trust in this context in their unique ways. For example, it would be insufficient to consider consumers’ trust in Internet shopping without examining the trust propensity (which is a personality trait) of the consumers concerned.

Trust in Internet Shopping 27

The distant literature on trust across a wide variety of disciplines provides a useful basis on which to investigate consumer trust, its antecedents, and its consequences in the context of Internet shopping. However, most of the literature still suffers from the problems identified by Mayer and Davis (1995). Confusions between trust and its antecedents still abound. For example, in Kini and Choobineh (1999) the definition of trust included the sources of trust itself. In addition, most of the literature does not contain empirical supporting evidence. In the context of consumer trust in Internet shopping, theory-guided empirical study is rare. This severe lack of theory-guided empirical studies is critically impeding our understanding of consumers’ trust in Internet shopping.

RESEARCH MODEL AND PROPOSITIONS This research synthesizes the diverse literature on trust in order to develop an integral research model of consumer trust in Internet shopping (CTIS). According to Hardin (1992), trust is a three-party relation involving properties of a trustor, attributes of a trustee, and a specific context in which trust is conferred. As depicted in Figure 1, trust in Internet shopping is affected by a consumer’s propensity to trust, the trustworthiness of an Internet vendor, and relevant external environmental factors impacting Internet shopping transactions. Limerick and Cunnington (1993) also argued that trust can reduce uncertainty about the future and is a necessity for a continuing relationship with participants who have opportunistic behavior. The essence of risk is uncertainty about the future. Thus, the formation of trust, in turn, reduces consumers’ perceived risk of Internet shopping.

Figure 1: A conceptual model of trust in Internet shopping Trustworthiness of Internet Vendor Perceived Security Control Perceived Privacy Control Perceived Integrity Perceived Competence External Environment

Trust in Internet Shopping

Third Party Recognition Legal Framework Propensity to Trust

Cultural Environment

Experience

Perceived Risk

28 Cheung & Lee

Trustworthiness of Internet Vendor The perceived trustworthiness of a party is often suggested as an important antecedent of trust. There is a long line of research (summarized in, for example, Mayer et al., 1995) examining the influence of perceived trustworthiness on the building of trust. Mayer et al. (1995) found that three factors–ability, integrity, and benevolence–are consistently related to trust in most previous studies. Hence, these factors are included in our model. In addition, in the specific context of this study, two new factors are added to the model to reflect the specific nature of Internet shopping. These two factors are perceived security control (PSC) and perceived privacy control (PPC). PSC and PPC are critical characteristics of Internet shopping transactions affecting the development of Internet users’ trust in Internet shopping. Previous studies find that these two factors are the major concerns of Internet users. In particular, privacy is the number one consumer issue facing the Internet (Benassi, 1999; Hoffman et al., 1999; Wang et al., 1998).

Perceived Security Control (PSC) In this study, perceived security control refers to the Internet users’ perception on the Internet vendors’ ability to fulfill security requirements, such as authentication, integrity, encryption, and nonrepudiation. Consumers tend to have a better trust in Internet shopping if a higher level of security is believed to exist. Therefore, the proposition is: Proposition 1: The perceived security control of Internet vendors is positively related to CTIS.

Perceived Privacy Control (PPC) In this study, perceived privacy control is conceived as the Internet users’ perception of the ability of Internet vendors in protecting consumers’ personal information collected from its electronic transactions from unauthorized use or disclosure. Consumers tend to have a better trust in Internet shopping if they perceive their privacy information as being well protected. Therefore, the proposition is: Proposition 2: The perceived privacy control of Internet vendors is positively related to CTIS.

Perceived Integrity (PI) In this study, perceived integrity refers to the perception of Internet users on the honesty of Internet vendors. For instance, whether they have consistent actions, whether their actions are congruent with their own words, and whether their transactions with consumers are fair. Integrity gives rise to trust. Therefore, the proposition is: Proposition 3: The perceived integrity of Internet vendors is positively related to CTIS.

Perceived Competence (PC) Perceived competency, in this study, is defined as the Internet consumers’ perception on the skills, abilities, and expertise of Internet vendors. Consumers tend to have a higher trust in Internet shopping if they think Internet vendors are competent. Therefore, the proposition is: Proposition 4: The perceived competence of Internet vendors is positively related to CTIS.

Trust in Internet Shopping 29

External Environment According to Lewicki and Bunker (1996), trust is context specific. In the faceless world of electronic commerce, third party recognition and a legal framework are two key environmental and contextual factors affecting the formation of consumers’ trust. Benassi (1999) argued that third party recognition, such as that provided by organizations such as TRUSTe, can help build consumers’ trust on the Internet and, in turn, accelerate the growth of the Internet. The Graphics, Visualization, & Usability Center’s (GVU) 10th WWW User Survey 3 also reported that a majority (over 70%) of Internet users worldwide wanted more new laws to protect their privacy online.

Third Party Recognition (TPR) In this study, third party recognition refers to the assurance of the trustworthiness of Internet vendors by third party recognition bodies. Thus, the proposition is: Proposition 5: The perceived effectiveness of the third party recognition is positively associated with CTIS.

Legal Framework (LF) In this study, legal framework refers to the law and code of practice established to protect Internet shoppers during electronic transactions. An effective legal framework can enhance consumers’ trust in Internet shopping. Thus, the proposition is: Proposition 6: The perceived effectiveness of the legal framework is positively associated with CTIS.

Propensity to Trust (PTT) Propensity to trust is a stable, within-party factor that affects the likelihood that a party will trust another party. People with different cultural backgrounds, personality types, and developmental experiences vary in their propensity to trust (Hofstede, 1980). People living in an environment with a strong culture of trust tend to have a higher propensity to trust. People who have a positive experience with the Internet also tend to be more trusting of Internet transactions. This propensity to trust is viewed as a personality trait that leads to generalized expectations about the trustworthiness of others. Mayer et al. (1995) have suggested that trust propensity has a direct impact on the formation of trust. Thus, the propositions are: Proposition 7: The strength of a trusting cultural environment is positively associated with PTT. Proposition 8: Prior positive personal experience in Internet usage is positively associated with PTT. Proposition 9: PTT is positively associated with CTIS.

Consequence of Trust: Perceived Risk (PR) Perceived risk is very powerful in explaining consumers’ behavior, since consumers tend more often to avoid mistakes than to maximize utility in purchasing (Mitchell, 1998). In particular, perceived risk is higher in Internet shopping than in the traditional mode of shopping because, for instance, a consumer will not be able to physically examine the appropriateness of a product before a purchase decision is made. Peter and Ryan (1976)

30 Cheung & Lee

argued that perceived risk generally consists of two components, one related to an uncertainty or probability of loss notion and the other related to a consequence or the importance of the notion of loss. In this study, perceived risk refers to the Internet users’ perception on the possibility of yielding unexpected outcomes with undesirable consequences. Many prior studies (Dion et al., 1995; Doney & Cannon, 1997; Morgan & Hunt, 1994) have discovered a strong relation between risk and the concept of trust. As suggested by Selnes (1998), perceived risk in a buyer-seller relationship is reduced by trust. Therefore, the proposition is: Proposition 10: CTIS is negatively associated with perceived risk in Internet shopping.

INSTRUMENT DEVELOPMENT A systematic and rigorous approach of developing measurement instruments is strongly advocated in the IS discipline. IS scholars (e.g., Bailey & Pearson, 1983; Ives et al., 1983; Doll & Torkzadeh, 1988; Davis, 1989) have claimed that constructs with strong theoretical justification and measures with high degrees of reliability and validity are prerequisites to cumulative knowledge in IS research. Since this study is one of the first empirical studies of trust in Internet shopping, a large part of the measurement instrument had to be developed from scratch, rather than be borrowed from the literature. The process of instrument development in this study is mainly divided into three stages. They are: (1) item creation, (2) scale development, and (3) instrument testing.

Stage 1: Item Creation Churchill (1979) recommended a series of techniques in generating measurement items, including literature searches, experience surveys, critical incident analysis, focus groups, and in-depth interviews. The use of these methods can enable the generation of measurement items with a relatively high degree of content validity (Moore & Benbasat, 1991). In this study, forty-one items were generated using the following three methods and all these measurement items are listed in Appendix 1. 1. Literature searches–five validated measurement items were obtained from the literature. 2. Focus groups–the items were generated through the discussion of the research topic with eight academic staff in the business faculty of a large university. 3. In-depth interviews–the items were obtained from personal interviews with six research students in the IS department of a local university.

Stage 2: Scale Development In this study, four judges who were experts in IS were requested to do the card sorting. As there were eleven constructs and forty-one items in the sample item pool, the card-sorting process was simplified with labels and definitions of constructs provided for categorization. Theoretically, if an item was correctly placed in a particular category, it was considered to display initial convergent validity with the related construct and discriminant validity with the others. Two measurements, Cohen’s (1960) Kappa and item placement ratio, were computed to assess the reliability of the sorting procedures and the construct validity of the scales. With

Trust in Internet Shopping 31

reference to previous studies (Vessey, 1989; Jarvenpaa, 1989; Todd & Benbasat, 1989), an acceptable score of Kappa should be greater than 0.65. In addition, high degree of “correct” placement of items within them can be considered to demonstrate a high degree of construct validity. In this study, the average inter-judge agreement score is 0.96, and the overall placement ratio of items to target constructs is 95.73%. In sum, a high level of agreement is found among the judges, and a high degree of convergent and discriminant validity of the scale is obtained. (Details of the two measurements are reported in Appendix 2.)

Stage 3: Instrument Testing Pilot Test A self-administrated questionnaire was distributed to 40 research students and academic staff (who were not involved in the previous stages of instrument development) in the faculty of business of a large university. Item-to-total score correlation and the effects of deleting items on Cronbach’s alpha were worked together to determine candidate items for further studies. Synthesizing all the previous procedures and iteration results, five items were removed from the item pool. As illustrated in Table 1, one item was removed from each of the five constructs, including “perceived security control,” “perceived privacy control,” “perceived integrity,” “perceived competence,” and “cultural environment.” In accordance with Hair et al. (1998), Cronbach’s alpha for each construct should be greater than 0.70. As shown in Table 1, most of these constructs have a Cronbach’s alpha larger than 0.70, except the construct “perceived security control” with an alpha of 0.68.

Field Test A refined self-administered questionnaire was distributed to the management information systems (MIS) students from the school of business in a large university, and a total of 405 usable questionnaires were collected. Before the scales were subjected to any statistical analysis, a thorough examination of the psychometric properties of measuring items was performed. The instrument validation process consisted of both classical approach (item-to-total score correlation, Cronbach’s alpha, and exploratory factor analysis) and contemporary approach (structural equation modeling) (Bagozzi, Yi, & Phillips, 1991). The aims of using these approaches as a way of method triangulation were to ensure the development of a rigor measurement instrument.

Classical Approach Similar to the pilot test, item-to-total score correlations and Cronbach’s alpha were applied to evaluate the reliability. In accordance with Table 2, all Cronbach’s alphas exceed 0.70 and all item-to-total correlations are greater than 0.70 (except one item in the construct “cultural environment”). For assessing the validity of the instrument, the 36 items were subjected to exploratory factor analysis (EFA). In an attempt to improve the interpretation and to obtain some theoretically meaningful factors, orthogonal rotational, EQUIMAX rotation,4 was applied to these items and an eleven-construct solution with a total of 74.63% variance extracted was obtained. Except for one item in the construct “perceived risk” that loaded higher on the construct “trust,” all the remaining items loaded on their respective theoretical constructs correctly. The factor loading matrix is shown in Appendix 3.

32 Cheung & Lee

Contemporary Approach Structural equation modeling (SEM) techniques are very useful in validating measurement instruments (Steenkamp & Trijp, 1991). LISREL (LInear Structural RELations) and PLS (Partial Least Squares) are two commonly used second-generation multivariate techniques to test psychometric properties of measurement instrument through SEM. Because of the ability to estimate latent model under conditions of non-normality and small sample sizes, PLS has been widely adopted in IS research (Chin, 1998; Compeau & Higgins, 1995; Vandenbosch & Higgins, 1996). In this study, we will evaluate the measurement model of PLS for instrument validation. Composite reliability and average variance extracted are used to assess the reliability of the constructs, and they are reported in Table 3. The acceptance value of composite reliability is 0.70 (Hair et al., 1998), and it is found that all constructs have a high degree of internal consistency. Another reliability measure, average variance extracted, reflects the overall amount of variance in the items accounted for by the latent construct. According to Fornell & Larcker (1981), average variance extracted is a more conservative measure than composite reliability, and their suggested acceptable level of average variance extracted is 0.50 or above for a construct. As shown in Table 3, all average variance extracted of constructs are greater than 0.50. Fornell and Larcker (1981) suggested that average variance extracted can also be used to evaluate discriminant validity. To demonstrate the discriminant validity of the constructs, average variance extracted for each construct should be greater than the squares of the Table 1: Cronbach’s Alpha of the eleven constructs in pilot study Construct

No. of Items

Cronbach’s Alpha

Item(s) Removed

Perceived Security Control

3

0.68

Internet vendors have the ability to verify Internet shoppers’ identity for security purpose.

Perceived Privacy Control

3

0.81

Perceived Integrity

3

0.73

Internet vendors will sell my personal information to the third parties without my permission. I will not be overcharged by Internet vendors during sales transactions.

Perceived Competence

3

0.90

Most Internet vendors have a good reputation.

Propensity to Trust Cultural Environment

4 3

0.93 0.74

None A high degree of trust exists in

Experience

3

0.89

my family. None

Third Party Recognition Legal Framework

3 3

0.85 0.85

None None

Trust Perceived Risk

4 4

0.70 0.82

None None

Trust in Internet Shopping 33

Table 2: Psychometric properties of measurement instrument (classical approach) * Item does not load into its target construct. Item

Item-to-total score correlation

Factor loading to “target” construct

a1a a1b a1c a1d

0.88 0.87 0.86 0.83

0.83 0.83 0.82 0.81

a2a a2b a2c

0.89 0.86 0.68

0.89 0.85 0.59

a3a a3b a3c

0.88 0.92 0.90

0.86 0.92 0.89

b1a b1b b1c

0.84 0.85 0.78

0.85 0.83 0.49

b2a b2b b2c

0.84 0.86 0.86

0.72 0.78 0.70

b3a b3b b3c

0.83 0.87 0.83

0.73 0.73 0.75

b4a b4b b4c

0.84 0.89 0.90

0.63 0.82 0.86

c1a c1b c1c

0.82 0.85 0.86

0.70 0.81 0.75

c2a c2b c2c

0.78 0.91 0.89

0.63 0.88 0.87

d1a d1b d1c d1d

0.86 0.88 0.83 0.75

0.68 0.64 0.60 0.70

d2a d2b d2c d2d

0.89 0.85 0.76 0.86

-0.71 -0.73 -0.41* -0.66

Propensity to Trust (α = 0.88)

Cultural Environment (α = 0.75)

Experience (α = 0.88)

Perceived Security Control (α = 0.76)

Perceived Privacy Control (α = 0.81)

Perceived Integrity (α = 0.79)

Perceived Competence (α = 0.85)

Third Party Recognition (α = 0.79)

Legal Framework (α = 0.83)

Trust (α = 0.85)

Perceived Risk (α = 0.86)

correlations between the constructs and all other constructs. By examining Table 4, it is obvious that all squared correlations between constructs are smaller than the average variance extracted of their respective constructs. In view of the results obtained from classical and contemporary approaches, two problematic items are found in the constructs “cultural environment” and “perceived risk.” First, a relatively low item-to-total scores correlation (0.68) is found in the item “I am living

34 Cheung & Lee

Table 3: Psychometric properties of measurement instrument (contemporary approach)

Propensity to Trust ρ=0.92 σ=0.73 Cultural Environment ρ=0.85 σ=0.66 Experience ρ=0.92 σ=0.80 Perceived Security Control ρ=0.85 σ=0.65 Perceived Privacy Control ρ=0.88 σ=0.72 Perceived Integrity ρ=0.88 σ=0.71 Perceived Competence ρ=0.91 σ=0.76 Third Party Recognition ρ=0.88 σ=0.71 Legal Framework ρ=0.90 σ=0.74 Trust ρ=0.90 σ=0.70 Perceived Risk ρ=0.91 σ=0.71

Item

Factor Loading

Standard Error

tstatistic

a1a a1b a1c a1d

0.88 0.88 0.85 0.81

0.02 0.01 0.02 0.03

43.78 62.92 43.78 29.91

a2a a2b a2c

0.88 0.88 0.66

0.03 0.02 0.08

33.30 38.92 8.54

a3a a3b a3c

0.89 0.94 0.85

0.26 0.22 0.25

3.45 4.35 3.38

b1a b1b b1c

0.77 0.79 0.86

0.10 0.07 0.04

7.39 10.82 21.01

b2a b2b b2c

0.81 0.84 0.89

0.03 0.02 0.02

28.67 40.25 52.83

b3a b3b b3c

0.80 0.88 0.84

0.03 0.02 0.03

24.09 35.64 25.24

b4a b4b b4c

0.84 0.90 0.88

0.02 0.02 0.03

36.37 45.81 33.50

c1a c1b c1c

0.83 0.88 0.81

0.19 0.25 0.24

4.29 3.49 3.42

c2a c2b c2c

0.83 0.89 0.86

0.04 0.03 0.04

22.51 29.20 21.19

d1a d1b d1c d1d

0.87 0.89 0.83 0.74

0.02 0.02 0.02 0.04

46.65 59.45 44.62 20.87

d2a d2b d2c d2d

0.89 0.87 0.75 0.85

0.01 0.02 0.04 0.03

76.72 52.89 17.22 30.48

Composite Reliability (ρ) =

p

(∑

i =1

p

(∑

i=1

λ

Variance Extracted (σ) =

λ

i

p

i

) 2 + (∑

i =1

p

∑ λ

)2

Var ( ε

p

i

))

Note: λ = Factor loading; ε = Measurement error

∑ λ i=1

2 i

+

i=1 p

2 i

∑ Var i=1



i

)

Trust in Internet Shopping 35

Table 4: Squared correlations between constructs (diagonal elements are average variance extracted) PTT CE EXP PSC PPC PI PC TPR LF Trust PR

PTT 0.73 0.15 0.00 0.05 0.04 0.00 0.03 0.06 0.01 0.01 0.00

CE

EXP

PSC

PPC

PI

0.66 0.03 0.02 0.01 0.00 0.00 0.03 0.02 0.00 0.00

0.80 0.03 0.04 0.04 0.07 0.01 0.02 0.00 0.00

0.65 0.41 0.25 0.19 0.10 0.13 0.10 0.10

0.72 0.21 0.19 0.14 0.17 0.21 0.13

0.71 0.30 0.04 0.10 0.21 0.09

Keys: PSC = Perceived Security Control PPC = Perceived Privacy Control PI = Perceived Integrity PC = Perceived Competence PTT = Propensity to Trust

PI PC

0.76 0.18 0.03 0.10 0.04

TPR

LF

Trust

PR

0.71 0.12 0.02 0.05

0.74 0.09 0.10

0.70 0.59

0.71

CE = Cultural Environment LF = Legal Framework TPR = Third Party Recognition PR = Perceived Risk EXP = Experience Trust = Trust

in a high trust society” of the construct “cultural environment.” Second, exploratory factor analysis indicates that the item “There are negative outcomes on Internet shopping” of the construct “perceived risk” loaded higher to the construct “trust.” Despite the problems identified, the two items will still be retained for further analysis. According to Chin (1998), factor loading of 0.50 or 0.60 may still be acceptable if there exist additional indicators in the block for comparison basis. “Cultural Environment” is a threeitem construct and the factor loading of the problematic item is 0.68, which is acceptable and should be retained. In contrast, item “perceived risk,” identified as problematic in the exploratory factor analysis, fits well in the structural equation modeling analysis. In particular, it exhibits a relatively high factor loading to its own construct (0.75), and the average variance extracted of its construct is (0.71) higher than all the squared correlations of other constructs. As suggested by Segars and Grover (1993), one of the shortcomings of exploratory factor analysis is that it tends to search for factors in an exploratory way, and the factor solution obtained is just one of an infinite number of possible solutions. Thus, all the 36 items validated through structural equation modeling techniques should be retained in the instrument.

CONCLUSIONS Drawing from findings of recent electronic commerce research and integrating trust theories from the fields of marketing, psychology, and sociology, this study proposes a conceptual model for the investigation of trust, its antecedents, and its consequences in the context of Internet shopping. Also, a 36-item measurement instrument with high reliability and validity for the trust model was developed systematically and validated using both classical and contemporary approaches. Consistent results from both approaches provide users with additional confidence in the desired psychometric properties of the instrument.

36 Cheung & Lee

This proposed research model improves our understanding of trust and electronic commerce and adds to the existing literature. Now that the research model is developed and the measurement instrument has been validated, the stage is set for the empirical testing of the theoretical model, which is an obvious area of future work. The results of such testing will help to clarify and enrich the relevant theories and extend their boundaries. In addition, the results can inform the management of Internet shops how they can manipulate trust antecedents to increase consumers’ trust and hence improve the chances of consumer purchasing from their Internet shops. Finally, the validated research instrument adds to the repository of rigorous research instruments for IS survey researchers to use, thus helping to develop a cumulated tradition for research in the IS discipline.

ENDNOTES 1 http://www.nua.ie/surveys/how_many_online/ 2 Neilsen//NetRatings e-Commerce ratio, Q4,2000, http://www.eratings.com 3 GVU 10th WWW User Survey Report, May 14, 1999: http://www.gvu.gatech.edu/ user_surveys/survey-1998-10/graphs/privacy/q59.htm 4 In practice, the objective of all methods of rotation is to simplify the rows and columns of the factor matrix to facilitate interpretation. The EQUIMAX approach is a compromise between two frequently used methods, QUARTIMAX and VARIMAX. Rather than concentrating either on simplification of the rows or on simplification of the columns, it tries to accomplish some of each.

ACKNOWLEDGMENT The work described in this paper was partially supported by grants from the Research Grants Council of the Hong Kong Special Administrative Region, China [Project Nos. CityU 1191/98H and CityU 1204/97H].

Trust in Internet Shopping 37

APPENDIX 1 Construct

Item

Sources

Percei ved S ecurity Control a1a Internet vendors implement security measures to protect Internet shoppers.

New item

a1b a1c

New item New item

Internet vendors have the ability to verify Internet shoppers' identity for security purposes. Internet vendors usually ensure that transactional information is protected fro m being accidentally altered or destroyed during transmission on the Internet.

a1d I feel secure about the electronic payment system of Internet vendors. Percei ved Privac y Control a2a Internet vendors will sell my personal information to the third parties without my permission.

New item

a2b

New item

Internet vendors are concerned about consumers’ privacy.

New item

a2c

Internet vendors will not divulge consumers’ personal data to other parties.

New item

a2d

I feel sa fe about the privacy control of Internet vendors.

New item

Percei ved Integrity a3a Internet vendors will not charge Internet shoppers more for Internet shopping. a3b

New item

Internet vendors are honest to their consumers.

Moorman et al., 1993

a3c

Internet vendors act sincerely in dealing with customers.

Moorman et al., 1993

a3d

I will not be overcharged by Internet vendors during sales transactions.

New item

Percei ved Competence a4a Internet vendors have the ability to handle sales transactions on the Internet.

New item

a4b

New item

Internet vendors have sufficient expertise and resources to do business on the Internet.

a4c

Internet vendors have adequate knowledge to manage their business on the Internet.

a4d

Most Internet vendors have a good reputation.

New item Doney & Cannon, 1997

Propensity to Trust b1a It is easy for me to trust a person/thing.

New item

b1b

My tendency to trust a person/thing is high.

New item

b1c b1d

I tend to trust a person/thing, even though I have little knowledge of it. Trusting someone or something is not difficult.

New item New item

Cultural Environment b2a A high degree of trust exists in my fa mily.

New item

b2b

People of my co mmunity trust each other.

New item

b2c

I a m living in a high-trust society.

New item

b2d

My friends are generally trustworthy.

New item

Experienc e b3a Using the Internet has been a good experience to me personally.

New item

b3b

I have positive experiences when using the Internet.

New item

b3c

I have good experiences when using the Internet.

New item

Third Part y Recognition c1a There ar e many reputable third party certification bodies available for assuring the trustworthiness of Internet vendors. c1b I think third party recognition bodies are doing a good job. c1c

Existing third party recognition bodies are adequate for the protection of Internet shoppers’ interest.

New item New item New item

Legal Framework c2a

The e xisting business code of conduct is sufficient for the protection of Internet shoppers’ interest.

New item

c2b

The e xisting law is adequate for the protection of Internet shoppers’ interest.

New item

c2c

The e xisting legal fra mework is good enough to protect Internet shoppers.

New item

Trust in Internet Shopping d1a Internet shopping is unreliable.

New item

d1b

Internet shopping cannot be trusted; there are just too many uncertainties.

d1c

In general, I cannot rely on Internet vendors to keep the promises that they make.

Chow & Holden, 1997

d1d

Anyone trusting Internet shopping is asking for trouble.

Chow & Holden, 1997

Percei ved Risk d2a Internet shopping is risky.

New item

New item

d2b

Shopping on the Internet entails uncertainty or vulnerability.

New item

d2c

There ar e negative outcomes fro m Internet shopping.

New item

d2d

I find it dangerous to shop on the Internet.

New item

38 Cheung & Lee

APPENDIX 2

Interviewer

Degree of Agreement Kappa

Interviewer

1 2 0.97 1 3 0.95 1 4 0.97 2 3 0.97 2 4 0.95 3 4 0.95 Table of degree of agreement–Kappa coefficient

ACTUAL CATEGORIES TARGET PSC PPC CATEGORY PSC 16 PPC 16 PI 2 PC PTT CE LF 2 TPR PR EX T Total Placement: 164

PI

PC

PTT

CE

13

LF

TPR

PR

EX

T

TOTAL

Overall Hit Ratio:

16 16 16 16 16 16 12 12 16 12 16 95.73%

1 16 16 16

2

8 12 16 12 16 Hits:

Table of item placement ratio

Keys: PSC = Perceived Security Control PPC = Perceived Privacy Control PI = Perceived Integrity PC = Perceived Competence PTT = Propensity to Trust CE = Cultural Environment LF = Legal Framework TPR = Third Party Recognition PR = Perceived Risk EX = Experience T = Trust

157

TOTAL % 100.00 100.00 81.25 100.00 100.00 100.00 66.67 100.00 100.00 100.00 100.00

.831 .827 .822 .809 .144 .198 .117 2.914E-02 6.256E-02 3.163E-03 3.815E-02 1.378E-02 .149 7.851E-02 3.039E-02 .133 -7.56E-03 4.556E-02 7.703E-02 6.529E-02 8.445E-02 6.088E-02 9.701E-02 1.940E-02 .123 .128 5.703E-02 1.220E-02 .110 7.726E-02 5.060E-02 2.448E-02 -8.57E-02 -9.78E-03 -.108 -9.87E-02

1

.122 .100 3.339E-02 -3.20E-02 -.120 -4.98E-02 .355 -5.00E-04 1.653E-02 -7.30E-03 5.747E-02 2.753E-02 -.115 .111 .136 1.847E-02 .127 2.937E-02 .137 .129 5.019E-02 1.137E-02 5.907E-02 .110 -8.83E-02 .128 -8.36E-03 -1.99E-02 .680 .637 .599 .701 -.425 -.410 -.526 -.365

2

3 9.882E-02 -1.11E-02 .129 -5.65E-03 4.401E-02 .145 -.308 -1.98E-02 -8.03E-03 9.607E-03 -1.81E-02 1.328E-02 .388 -2.38E-02 -1.24E-02 .269 6.384E-02 .193 6.363E-02 -2.37E-02 7.344E-02 5.956E-02 -2.34E-02 7.573E-02 .148 -3.95E-02 .104 7.552E-02 .401 .479 .430 .227 -.709 -.727 -.412 -.664

Extraction Method: Principal Component Analysis. Rotation Method: Equamax with Kaiser Normalization. a. Rotation converged in 12 iterations.

a1a A1B A1C A1D A2A A2B A2C A3A A3B A3C B1A B1B B1C B2A B2B B2C B3A B3B B3C B4A B4B B4C C1A C1B C1C C2A C2B C2C D1A D1B D1C D1D D2A D2B D2C D2D

4 -3.94E-03 2.713E-02 2.421E-03 .118 5.652E-02 2.419E-02 .177 .856 .919 .891 .119 8.866E-02 -3.00E-02 .140 -1.98E-02 -3.35E-02 4.135E-02 2.552E-02 .105 .105 5.577E-02 8.206E-02 5.266E-02 2.379E-02 -7.58E-03 2.879E-03 -8.37E-02 -2.45E-02 -5.38E-02 -3.41E-02 1.606E-03 6.883E-02 3.021E-02 3.045E-02 -5.50E-02 -2.60E-02 .116 7.889E-02 5.303E-02 1.445E-02 4.248E-02 4.984E-02 3.747E-02 -1.31E-02 -3.69E-02 -4.57E-02 4.600E-02 .135 .101 .145 .166 .268 8.737E-02 .161 .134 3.029E-02 .114 8.641E-02 .154 .153 .211 .628 .881 .873 .122 .165 .116 -8.26E-02 -.133 -.151 4.752E-02 -3.19E-02

5

Component 6 5.117E-02 9.908E-02 7.023E-02 6.023E-02 4.978E-02 -8.56E-02 .185 7.749E-02 1.712E-02 8.194E-02 .153 .123 8.840E-02 .186 .159 8.921E-02 .262 .191 .184 .628 .823 .860 9.582E-02 .191 .214 .135 2.354E-02 9.061E-02 3.894E-03 7.306E-02 4.344E-02 .187 -7.16E-02 -2.60E-02 -.145 -.122

Rotated Component Matrix 7 5.372E-02 6.082E-02 .155 3.872E-02 2.781E-03 6.279E-02 7.060E-02 -1.18E-03 3.167E-02 2.637E-02 8.721E-02 .218 .409 .716 .779 .697 .169 .322 8.016E-02 .150 .185 8.681E-02 7.519E-02 .129 .164 .133 .140 .177 .184 .124 .109 7.068E-02 -8.96E-02 -7.65E-02 -8.40E-02 -.140

a

8 7.308E-02 .121 2.046E-02 .101 3.080E-02 5.243E-02 8.984E-02 1.967E-03 4.432E-02 8.863E-03 .118 5.998E-02 8.803E-02 9.165E-02 .152 .176 -6.24E-03 .156 .157 .278 .129 .165 .702 .810 .750 .322 .122 .105 -4.81E-02 2.618E-02 -2.40E-02 .147 -.103 -8.20E-02 -.269 -.184

9 9.618E-02 4.567E-02 3.877E-02 1.058E-02 3.769E-02 9.067E-02 2.794E-02 2.230E-02 6.927E-02 4.233E-02 9.068E-02 4.609E-02 .136 .155 .164 .239 .732 .733 .751 .352 .208 .147 .200 8.012E-02 1.803E-02 .235 9.793E-02 4.408E-02 .157 .155 .169 .197 -.139 -.136 -4.30E-02 -.154

10 7.412E-02 1.358E-02 4.405E-02 8.637E-02 5.936E-02 .102 -5.64E-02 8.579E-02 5.309E-02 6.459E-02 .847 .828 .491 .277 .118 .201 1.427E-02 9.630E-02 .193 .132 .115 .188 .343 2.180E-02 5.158E-02 7.741E-02 .103 9.886E-02 .137 .115 .156 -3.99E-02 -5.33E-02 -6.64E-02 9.851E-02 -6.05E-02

.155 .207 .138 .148 .889 .846 .591 .107 5.628E-02 1.581E-02 7.353E-02 5.824E-02 2.102E-02 3.967E-02 .117 -3.85E-03 9.843E-02 2.800E-02 9.496E-02 -4.07E-03 .120 -2.55E-02 .137 1.889E-03 .101 -3.79E-02 7.032E-02 9.713E-02 3.823E-02 9.312E-03 -7.37E-02 -6.07E-04 -6.11E-02 -7.70E-02 -2.08E-02 -2.27E-02

11

APPENDIX 3: FACTOR LOADING OF THE MEASUREMENT INSTRUMENT

Trust in Internet Shopping 39

40 Cheung & Lee

REFERENCES Bagozzi, R. P., Yi, Y. and Phillips, L. W. (1991). Assessing construct validity in organizational research. Administrative Science Quarterly, 36(3), 421-458. Bailey, J. E. and Pearson, S. W. (1983). Developing a tool for measuring and analyzing computer satisfaction. Communications of the ACM, 26(10), 785-793. Benassi, P. (1999). TRUSTe: An online privacy seal program. Communications of the ACM, 42(2), 56-59. Chin, W. W. (1998). The Partial Least Squares Approach for Structural Equation Modeling. Modern Methods for Business Research, 295-336. Marcoulides G.A. Hillsdale, NJ: Lawrence Erlbaum Associates. Chow, S. and Holden, R. (1997). Toward an understanding of loyalty: The moderating role of trust. Journal of Managerial Issues, 15(3), 275-298. Churchill, G. A. (1979). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, (16), 64-73. Cohen, J. A. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, (20), 37-46. Compeau, D. R. and Higgins, C. A. (1995). Application of social cognitive theory to training for computer skills. Information Systems Research, (6), 118-143. Das, T. K. (1998). Between trust and control: Developing confidence in partner cooperation in alliances. The Academy of Management Review, 23(3), 491-513. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319-340. Dion, P., Easterling, D. and Miller, S. J. (1995). What is really necessary in successful buyerseller relationships? Industrial Marketing Management, 24(1), 1-9. Doll, W. J. and Torkzadeh, G. (1988). The measurement of end-user computing satisfaction. MIS Quarterly, 12(2), 259-274. Doney, P. M. and Cannon, J. P. (1997). An examination of the nature of trust in buyer-seller relationships. Journal of Marketing, (61), 35-51. Dwyer, R. F., Schurr, P. H. and Oh, S. (1987). Output sector munificence effects on the Internal political economy of marketing channels. Journal of Marketing Research, (24), 347358. Fornell, C. and Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, (18), 39-50. Hair, J. F., Anderson, R. E., Tatham, R. L. and Black, W. C. (1998). Multivariate Data Analysis (5th edition). Englewood Cliffs, NJ: Prentice Hall. Hardin, R. (1992). The street-level epistemology of trust. Politics and Society, (21), 505-529. Hoffman, D. L., Novak, T. P. and Peralta M. (1999). Building consumer trust online. Communications of the ACM, 42(4), 80-85. Hofstede, G. (1980). Motivation, leadership, and organization: Do American theories apply abroad? Organizational Dynamics, 9(1), 42-63. Hulland, J., Chow, Y. H. and Lam, S. (1996). Use of causal models in marketing research: A review. International Journal of Research in Marketing, (13), 181-197. Ives, B., Olson, M. H. and Baroudi, J. J. (1983). The measurement of user information satisfaction. Communications of the ACM, 26(10), 785-793. Jarvenpaa, S. L. (1989). The effect of task demands and graphical format on information processing strategies. Management Science, 35(3), 285-303. Jarvenpaa, S. L., Tractinsky, N. and Vitale, M. (1999). Consumer trust in an Internet store. Information Technology and Management, 1(1-2), 45-72.

Trust in Internet Shopping 41

Kini, A. and Choobineh, J. (1998). Trust in electronic commerce: Definition and theoretical considerations. Proceedings of the 31st Hawaii International Conference on System Sciences (HICSS), 51-61. Lewicki, R. J. and Bunker, B. B. (1995). Trust in relationships: A model of trust development and decline. Conflict, Cooperation, and Justice. San Francisco, CA: Jossey-Bass. Lewicki, R. J. and Bunker, B. B. (1996). Developing and maintaining trust in work relationships. Trust in Organizations: Frontiers of Theory and Research. Sage Publications. Limerick, D. and Cunnington, B. (1993). Managing the New Organization. San Francisco, CA: Jossey-Bass. Mayer, R. C. and Davis, S. F. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709-734. Mitchell, V. W. (1999). Consumer perceived risk: Conceptualizations and models. European Journal of Marketing, 33(1-2), 163-195. Moore, G. C. and Benbasat, I. (1991). Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192- 222. Moore, C., Deshpande, R. and Zaltman, G. (1993). Factors affecting trust in market research relationships. Journal of Marketing, (57), 81-101. Morgan, R. M. and Hunt, S.D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, (58), 20-38. Peter, J. L. and Ryan, M. J. (1976). An investigation of perceived risk at the brand level. Journal of Marketing Research, (13), 184-188. Schurr, P. H. and Ozanne, J. L. (1995). Influences on exchange processes: Buyers’ preconceptions of a seller’s trustworthiness and bargaining toughness. Journal of Consumer Research, (11), 939-953. Segars, A. H. and Grover, V. (1993). Re-examining perceived ease of use and usefulness: A confirmatory factor analysis. MIS Quarterly, 17(4), 517-525. Selnes, F. (1998). Antecedents and consequences of trust and satisfaction in buyer-seller relationships. European Journal of Marketing, 32(3-4), 305-322. Steenkamp, J. E. M. and van Trijp H. C. M. (1991). The use of LISREL in validating marketing constructs. International Journal of Research in Marketing, (8), 183-199. Subramanian, A. and Nilakanta, S. (1994). Measurement: A blueprint for theory building in MIS. Information and Management, 26(1), 13-20. Todd, P. A. and Benbasat, I. (1989). An experimental investigation of the impact of computerbased decision aids on the process of preferential choice. Working Paper, School of Business, Queen’s University. Vandenbosch, B. and Higgins, C.(1996). Information acquisition and mental models: Investigation into the relationship between behavior and learning. Information Systems Research, 7(2), 198-214. Vessey, I. (1984). An investigation of the psychological processes underlying the debugging of computer programs. Unpublished Ph.D. Dissertation, Department of Commerce, University of Queensland. Wang, H. Q., Lee M. K. O. and Wang, C. (1998). Consumer privacy concerns about Internet marketing. Communications of the ACM, 41(3), 63-70.

42 Katz

Chapter IV

The Integral Role of Information Technology in Achieving Business Strategy Success: Managing the Information Resources of Global Competitors Jeffrey D. Katz Kansas State University, USA

This chapter examines the impact of culture on the organizational and information technology infrastructures of global competitors while considering the role business strategy plays on both infrastructures. Information technology is increasingly becoming a focal point of the firm’s business strategy for internally derived competitive advantage, as well as a means for linking to the firm’s supply chain partners. Prior research has suggested that when managers consider how culture impacts the information flow and structure of the organization and its supply chain partners, a more effective “fit” between business strategies and performance results. A conceptual framework is proposed comparing national cultures and two important organizational factors—components of organizational structures and components of information systems. Specifically, the framework helps explain how organizational structures and information technology will vary based on national culture within the confines of the firm’s chosen competitive strategy. Examples from three cultures are used to demonstrate how international managers and researchers may better understand the impact of cultural differences on the operations of firms choosing to compete in international markets. Finally, propositions are offered for future empirical testing by information management researchers. Copyright © 2002, Idea Group Publishing.

The Integral Role of Information Technology 43

INTRODUCTION Information technology has fundamentally altered the way firms compete. It has added electronic methods to the firm’s business strategy, forced information technology to the center of the firm’s strategic thinking, and has clearly become a central focus in business integration through the streamlining of the supply chain (Betz, 2001). Strategic information systems now serve to “network” the supply chain by integrating business strategy and information technology (Pun & Lee, 2000). According to Cross (2000), “Information and communication technologies are forcing managers to rethink and reshape their business strategies, their use of technology, and their relations with suppliers and customers. The convergence of new technologies, hypercompetitive markets, and ‘heat-seeking’ financial and human capital that quickly flow to new and untested business models now threatens a number of traditional business models and processes.” (p. 36). However, when organizations expand across national borders, managers are faced with a host of conditions differing from those affecting their domestic operations. These include differences in national cultures, competitive strategies, information sharing processes and worker expectations (Lynn, 1999; Shore & Venkatachalam, 1996). In addition, in international expansions managers are under increased pressure to integrate and coordinate multicountry operations resulting in the need to use human and information resources as effectively as possible (Shore, 1996). Ensuring that workers in every location share knowledge accurately, and in a timely manner, is part of the human resource and information system managers’ responsibilities. This can be a daunting task. For example, global competition in the oil industry led British Petroleum to initiate the International Management Project to implement sweeping changes in the oil company’s global management structure, information sharing processes and corporate culture. The goal of the Project was to achieve management consensus on organizational roles and responsibilities, implement effective information sharing , and develop an organizational structure that would encourage flexible networks rather than formal, face-to-face meetings (Siddall, Willey & Tavares, 1992). Thus, in the push to expand operations to the global marketplace, organizations’ information systems are playing an increasingly key role. In contrast to their operations in the United States, managers are discovering that processes for sharing information are not easily transportable across borders (Ein-Dor, Segev & Orgad, 1993). In order to achieve timely information system implementation, the systems must be developed with an in-depth understanding of local factors (Barker, 1993). Recent research in the area of information technology and information sharing has proposed contingency models explaining how information systems should be based on the need for organizational “fit” by taking into account key external factors (Henderson & Venkatraman, 1991), as well as differences between headquarters and subsidiary operations (Cheung & Burn, 1994). According to contingency theory, competitive strategies and organizational infrastructures should “fit” the external and internal factors influencing the firm in order to achieve superior competitive performance (see for example, Andrews, 1987; Chandler, 1962; Child, 1972; Porter, 1985). Recent research regarding how firms are structured suggests that values underpinning the formal organizational structure convey important information for workers to execute the mission of the firm within the confines of management’s chosen business strategies (Gibson, 1994; Hinings, Thibault, Slack & Kikulis, 1996). For example, Leidner, Carlsson, Elam and Corrales (1999) document the differences in how managers perceive the benefits of information sharing processes as they

44 Katz

relate to decision support and structure in Mexico and Sweden. The authors conclude that cultural differences must be fully understood before information technology can be applied across borders by a firm. Conveying timely and accurate information to workers inside the firm, as well as to members of the firm’s supply chain partners, in a familiar format based on their cultural expectations, is especially important when firms expand into new markets. Managers are responsible for designing effective organizational and information infrastructures to guide workers in these new competitive environments (Harrigan & Dalmia, 1991). Ideally, organizational and information system infrastructures should: 1. help implement the firm’s business strategy, 2. guide how work is performed, 3. guide how information is shared, and 4. determine the effectiveness of managerial decision making within the parameters of the firm’s chosen goals and strategies (Harris & Harris, 1988). For example, a five-year, multimillion dollar study identified the theory of strategic alignment which proposes that the alignment between the firm’s strategy, organizational structure and information technology, as the key to successful global competition in the future (Pankratz, 1991). More recently, Katz, Zarzeski and Hall (1997) empirically examined the relationship between culture, the competitive environment affecting the firm, organizational strategies and firm performance. They found that strategies under the control of functional managers, which are consistent with the goals of the organization and properly “fit” the culture and industry environments in which the firm operates, result in superior performance. In a more directly related article by Shore and Venkatachalam (1995), the authors describe how national culture affects the analysis and design of information systems, which includes strategic information systems planning, system design and implementation. In this chapter, a framework is presented that focuses on the relationships between the culture in which an international firm chooses to compete, how to properly organize the firm based on the host country culture, and how to structure information systems to most effectively use information resources within the firm’s chosen competitive strategy. The discussion begins by defining cultural differences, business strategy, organizational structure and information technology, including the components of each factor that may vary in different cultural situations. Figure 1 graphically displays the relationships between these factors (culture, organizational infrastructure and information systems infrastructure), while also considering the impact of the firm’s business strategies. The model is then extended to two frameworks for understanding how culture affects organizational structure and information sharing (Table 1, Panels A and B). Finally, the frameworks are brought together by applying them to three countries (Table 2) as examples of how managers may apply the frameworks when expanding the operations of their firm to specific cultures. Research propositions and suggestions are provided for testing the framework as a way to encourage future empirical verification.

NATIONAL CULTURE Based on the results of an international empirical study, Hofstede (1980) proposed that the underlying national culture affects the attitudes and information processing of workers. Four factors were suggested which affect worker values: individuality, masculinity/femininity, uncertainty avoidance and power distance. More recently, Hofstede (1991) focused on cultural factors affecting organizational infrastructures cross-nationally by asking two specific questions: 1) what rules or procedures will be followed to attain the desired ends of

The Integral Role of Information Technology 45

Figure 1: Relationships between culture, strategy, structure and information technology

External Environment (Strategic Fit)

Organizational Culture

Task Environment (Functional Integration)

Organizational Infrastructure

National Culture

Business Strategies

Information Technology Infrastructure

Adapted from: Hofstede, 1991 and Henderson & Venkatraman, 1993

the organization, and 2) who has the power to decide? He proposed that cultural norms regarding uncertainty avoidance influence the answer to the first question, while cultural norms regarding power distance influence the answer to the second question. To understand why managers make decisions affecting the design of organizational infrastructures, uncertainty avoidance and power distance have been suggested as important factors of national culture impacting the motivations of workers and behaviors of decision makers (Harvey, 1997; Png, Tan & Wee, 2001; Routamaa, 1985). Briefly, uncertainty avoidance is the extent to which people feel threatened by ambiguous situations and have created infrastructures to avoid these situations. Power distance is the extent to which less powerful members of organizations accept that power is distributed unequally. Power distance creates the motivation for workers to rely on the authoritative role of managers (Hofstede, 1980). Each cultural component is addressed more fully in the following subsections.

Uncertainty Avoidance The extent that people of a culture are threatened by ambiguous situations, and adjust their workplace to avoid these situations, is ambiguity avoidance. Countries with people

46 Katz

who do not like uncertainty tend to have high needs for security and a strong belief in experts. Example countries include Germany, Spain, and Japan (Hofstede, 1980). Countries having a low level of uncertainty avoidance tend to use less formal rules. Conversely, workers in countries having a high level of uncertainty avoidance prefer to have a structure where a system of rules is strongly established (Hofstede, 1991). Coordination of information systems and feedback is favored in countries with high levels of uncertainty avoidance (Jaeger, 1984; Png et al., 2001). Workers in these countries find this is a good way to handle conflict, because the emphasis on standardization of skills is stronger in countries having lower levels of uncertainty avoidance. Thus, face-to-face meetings, or sharing of information electronically, become preferred coordination techniques to formal rules and procedures.

Power Distance The extent less powerful members accept that power is distributed unequally is power distance. In high power distance countries such as France, South Korea and India, organizations tend to have taller, or pyramidal, structures (Hofstede, 1980). In high power distance cultures, task specialization is important where productivity and technical expertise are required. Conversely, in countries having low levels of power distance, the work force tends to be less specialized. In countries having high levels of power distance workers prefer a high level of authority centralization and minimal sharing of information using technology. Conversely, in countries having low levels of power distance workers will prefer decentralization of authority and access to information through technology. Thus, in countries having lower levels of power distance workers will prefer organizational structures that are flatter, and these organizations will have less supervisory personnel. Since the concept of authority delegation is closely linked to decentralization, Hofstede (1991) predicts that countries having high levels of power distance will employ lower levels of delegation. The implications are that in terms of job autonomy, the higher the power distance, the lower the job autonomy expectations of the workers and the less information will be shared. Although the structure of a firm or its information system design in one country may appear to have the same reasons and approaches for information flows as those located in other countries, how workers interpret these structures, and rely on information needs, varies from country to country (Ein-Dor, et al., 1993; Harvey, 1997; Hortum & Muller, 1989; Lewis & Shea, 1996; Png, Tan & Wee, 2001). Thus, organizational structure and information technology infrastructure are the means for conveying information to workers, while designing both systems is about how and why the means are chosen by managers (DeLisi, 1990; Jones, 1995; Shore & Venkatachalam, 1995). For example, a recent study by DiBella (1996) reports the difficulty encountered by one company, which resulted from the different cultural orientations of workers, when managers attempted to change the organizational structure. “The failure to realize intended change is not due to poor managerial communication but from the innate fragmentation of meaning that comes from cultural differences” (DiBella, 1996: 352). In another study by Broderick and Boudreau (1991) it was found that information technology was key to enhancing performance when it was also linked to the strategy, organizational structure and culture of the firm. In a more recent study, Png et al. (2001) examined the corporate adoption of information technology based on uncertainty avoidance and power distance. Their results suggest that firms headquartered in countries with higher levels of uncertainty avoidance

The Integral Role of Information Technology 47

were less likely to adopt newer information sharing technologies. Although their study was broad-based and documented information technology practices in 153 firms in 24 countries, the study did not examine the changes necessary when firms expand across borders based on the expectations of workers and supply chain partners residing in other countries. Finally, Garfield and Watson (1997) report differences in information infrastructures based on national culture.

Summary With the enhanced competitive success of global competitors as the primary goal, a framework is presented for better understanding how culture affects the way firms and information systems are organized within the context of the firm’s chosen competitive strategy. This is an important issue since designing processes to focus the use of resources is a discretionary managerial function that human resource and information technology managers often must make—and one of the most commonly used tools for implementing business strategies (Carillo & Kopelman, 1991). Prior research has suggested that if organizational and information infrastructures are not developed in concert with an understanding of underlying national cultures, the implementation and ultimate success of business strategies will be severely impaired (Calori, Lubatkin & Very, 1994; Ein-Dor et al., 1993; Hill, Loch, Straub & El-Sheshai, 1998; Mascitelli, 1999).

BUSINESS STRATEGY Recent research has identified the importance of the relationship, or “fit”, between the firm’s competitive condition, the strategy chosen to achieve the firm’s goals and the use of information technology in the ultimate success of the firm. According to Shore and Venkatachalam (1996), the competitive environment affects the firm’s choice of competitive strategy, which in turn affects the firm’s organizational and information technology infrastructure choices. For example, Floyd and Wooldridge (1990) examined the competitive strategies of 127 banks and reported that competitive strategies have a significant effect on information technology adoption. Their study examined the competitive behavior of firms within a single industry (retail financial services) but did not consider strategies used by international competitors, which may be impacted by the underlying national culture. According to Porter (1985) there are two basic business strategies adopted by firms in an effort to compete within an industry: Low-cost leadership and differentiation. Applying these generic strategies to the international competitive environment, Porter (1990) suggests firms will implement a particular business strategy either on a country-by-country or global basis. That is, firms choosing to expand into international markets will decide whether using a particular business strategy for one country, or all countries, will be an effective method to achieve its goals. Recently, Katz, Werner and Brouthers (1999) examined the role played by underlying national culture on the competitive strategies of international banks. They reported that national culture affects the firm’s chosen business strategy that in turn impacts the performance of the firm. According to Katz, et al. (1999), Japanese banks competing in the international marketplace will tend to focus on attaining market share at the expense of short-term profits by using a low-cost leadership strategy focusing on the commercial banking sector while U.S. banks tend to target enhanced profitability by implementing a differentiation strategy in the retail segment of the banking industry. Henderson and Venkatraman (1991) proposed the Strategic Alignment Model that considers the use of information technology in the firm’s business strategy execution. The

48 Katz

Model contains two key dimensions: strategic fit and functional integration. Strategic fit recognizes the need to position the firm in an external marketplace where growth can take place while functional integration addresses how to best structure internal systems to execute the business strategy of the firm. Figure 1 depicts the conceptual relationship between these factors. In this conceptual model, organizational structure is proposed to be affected by two externally focused processes, culture and strategy, while information technology infrastructure is proposed to maintain a reciprocal, or “fit”, relationship with organizational structure. In extending the Strategic Alignment Model, Venkatraman, Henderson and Oldach (1993) propose that four basic domains of strategic choice exist for each firm: business strategies, information technology strategies, organizational infrastructure, and information technology infrastructure. As reflected in Figure 1, the first two comprise the strategic fit of the organization to its external environment while the latter two comprise the functional integration of the firm (which is also referred to as the firm’s task environment). Venkatraman et al. (1993) also suggest the approach for achieving strategic “fit” should include governance processes, technology capabilities, human resource capabilities and value management processes which are based on access to information by different groups in the firm. Thus, according to Day and Wendler (1998) the role of top managers of firms choosing to compete in international markets have shifted from master entrepreneurs to architects responsible for, “...designing the organization to achieve the best tradeoff between personal initiatives and enforced cooperation (p.4).” Nelson and Clark (1994) suggest a research agenda for information systems that encourages the inclusion of the firm’s strategic, cultural, organizational, and decision contexts. Their research model indicates that if all relevant contexts are considered, positive firm performance outcomes result. Figure 1 addresses these four factors by identifying two (culture and business strategy) that are affected by strategic fit with the external environment and two factors (organizational infrastructure and information technology infrastructure) affected by functional integration resulting from the firm’s task environment. It is the domain of functional integration in which human resources and information systems managers must operate, and the focus of the framework described in this chapter The impact of information sharing on business strategy success was recently addressed by Lee and Whang (2000) with a special focus on the role of the supply chain. The authors propose that for tight coordination to occur between supply chain partners and the firm’s workers, information sharing must be effective and match the firm’s business strategy. They suggest that critical, shared information may include inventory levels, sales forecasts, order status, and production schedules. Clearly, information technology, particularly information sharing processes, impact the firm’s ability to achieve its business strategy. This occurs through the firm’s ability to effectively share information with its workers and supply chain partners.

ORGANIZATIONAL INFRASTRUCTURE An organization is a goal-directed group of individuals (Aldrich, 1979; Daft, 1983; Hall, 1977). The core of every organization consists of internal relationships, such as how tasks are defined, how information and decision processes are communicated, and how reward systems are designed, which should closely fit the mission and strategy of the firm in order to achieve a reasonable level of individual cooperation and competitive success (Child, 1972; Ouchi, 1978; Gibson, 1994). These relationships define what the organization is and how it

The Integral Role of Information Technology 49

works (Daft & Steers, 1986). Managers have the authority to use organizational and information technology infrastructures to control these relationships (Child, 1972; Galbraith & Nathanson, 1978; Harrigan & Dalmia, 1991; Day & Wendler, 1998). More recently, several authors have suggested that supply chains are beginning to behave as cohesive groups of individuals, much like workers in a single organization, to achieve mutually beneficial goals (Bowersox, Closs & Stank, 2000). For example, according to Lee and Whang (2000), supply chain partners now work in tight coordination due to technological advances that have enhanced the level of information sharing. Bovel and Martha (2000) suggest the traditional supply chains characterized by the “demand-pull” are being replaced by “value-nets” which they describe as, “…a dynamic network of customer/supplier partnerships and information flows…activated by real customer demand and is capable of responding rapidly and reliably to customer preferences.” (p. 24). Whether they are individuals or supply chain partners, five basic organizational characteristics, or factors, comprise the organizational infrastructure of most firms or groups of firms. They are: division of work, degree of specialization, level of decentralization, type of coordination, and degree of formalization (Daft & Steers, 1986; Pugh, Hickson, Hinnings & MacDonald, 1969). Each is discussed in greater detail in the following subsections.

Division of Work The division of work refers to the assignment of tasks among individuals, or supply chain partners, to achieve shared goals. This dimension reflects the predictability and routine-ness of assigned tasks (Dessler, 1976). Division of work is implemented by the creation of departments or through the use of information sharing mechanisms. Departmentalization explicitly defines tasks and reporting relationships and can be accomplished according to organizational purpose (product, customer or location) or by process (dividing work into managerial functions such as planning, organizing and controlling). Integrating processes are used to facilitate communication and coordination among organizational subunits (Lee & Whang, 2000). Such processes include establishing liaison roles, assigning task forces, and using permanent teams or committees to coordinate activities (Galbraith, 1973).

Specialization Specialization occurs when assigning tasks in specific areas results in greater productivity than when several steps of work are accomplished independently. Organizing workers or supply chain members by specialty allows for the further development of expertise and a focus on an area of competitive advantage (Bowersox et al., 2000). According to organization theory, the more work is divided, the more efficient the process should become (Jones, 1995). The degree of specialization is measured by role specificity and task structure. For example, an automotive engineer working in a design department of a large automaker might specialize in the design of impact resistant bumpers, while an engineer working in a smaller automaker might be responsible for the design of the entire automotive frame.

Decentralization How much authority a worker or supply chain partner has is determined by the degree of decentralized decision-making (Udell, 1967; Bovel & Martha, 2000). These dimensions reflect the vertical nature of the organization. Decentralization addresses two important

50 Katz

structural characteristics: the number of hierarchical levels and the span of control. Decentralization is more than delegation of authority. It is also the process of controls through the number of levels and the span of control in the organization (Drucker, 1946; Forehand & Gilmer, 1965; Ouchi, 1978).

Coordination In addition to the horizontal and vertical aspects of organizations, coordination is also important. Coordination is the fitting together of job performance, or the linking of individual efforts, for the accomplishment of major organizational objectives (Dessler, 1976). Coordination is the process of achieving effort among the different subsystems in performing tasks. Coordination of information systems and feedback are particularly important when the firm’s environments involve rapid changes in competitive factors, such as in high technology industries (Emery & Trist, 1965).

Formalization Formalization is the degree that rules are placed in definitive form, thus making them rigid. Rules can be defined objectively and subjectively. For example, structures that emphasize written objectives, standards in decision-making, and bookkeeping standards might be more formalized in one firm than another. Often, formalization is associated with higher levels of bureaucracy and centralized decision-making (Blackburn, 1982).

Typical Organizational Structures Three common organizational structures are: functional structure, unit structure and matrix structure (Daft and Steers, 1986). In the functional structure, workers performing the same kind of task are grouped in the same department or supply chain partner. A functional structure is preferred when efficiency within a highly technical field is a major goal. In the unit structure, functions are grouped around products, lines of products, or services. When the firm’s external environment is unstable, when firms are large and complex, or when technology causes interdependence between functions and success is based on innovation, a self-contained unit structure is often the most commonly used organizational form. Finally, the matrix structure is a combination of the unit and functional structures. Matrix structures are typically employed when firms are medium to large in size and when the firm’s external environment is very uncertain (Aldrich, 1979; Emery & Trist, 1965). Table 1-Panel A depicts the conceptual relationships between national culture and organizational structure. For example, it is proposed that under conditions of high uncertainty avoidance and high power distance, managers will organize in a more structured and bureaucratic organizational form. Conversely, in cultures having lower levels of uncertainty avoidance and lower levels of power distance, managers will organize in a flatter and more decentralized structure. Table 1-A focuses on the relationship between national culture and organizational structure while assuming the business strategy of the firm is already chosen. That is, the table assumes the business strategy choice of the firm is external to the decision making authority of functional managers but that national culture will continue to systematically affect organizational structure decisions. Thus, based on the model contained in Figure 1, organizational structure decisions made by managers will be affected by external factors including the firm’s national culture and previously chosen business strategy. How a firm, or group of firms comprising the supply chain, is organized affect managers’ approach toward designing the information

The Integral Role of Information Technology 51

sharing system. For example, a firm employing a low-cost leadership business strategy in response to a national culture that encourages competition through high worker productivity, such as Japan, will likely result in decisions to implement an organizational infrastructure encouraging production efficiency, such as the functional organizational structure, and information systems that monitor and share the achievement of production costs (Yeh, 1991; Leidner, et al., 1999). The more specific role of culture on information technology is further discussed in the following section.

INFORMATION TECHNOLOGY INFRASTRUCTURE Just as organizational structure plays a role in the implementation of the firm’s business strategy, so does information technology. There are two fundamental sources of information sharing that enhance the application of the firm’s chosen business strategy: 1. the ability to assess the external environment of the firm more rapidly, and 2. the ability to assess organizational resources and capabilities already controlled by the firm. Both business strategy approaches, low- cost leadership and differentiation, indirectly impact the use of information technology. For example, the low-cost leader will likely rely on information systems to both scan the external environment for opportunities to reduce the costs of production and monitor the use of existing organizational resources thereby ensuring the most effective use possible. However, many barriers to the use of information technology across borders involve the expectations and reactions of workers and how they apply information technology based on culture (Lin, Vassar & Clark, 1993; Straub, 1994). It has been argued that taking into consideration the cultural norms relating to information technology will make information sharing more effective (Ein-Dor et al., 1993; Harvey, 1997; Kitchell, 1995; Shore & Venkatachalam, 1995; 1996). Therefore, the use of information technology to extend the decision-making of managers is necessary, particularly when firms adopt an international expansion strategy (Nelson & Clark, 1994; Leidner et al., 1999; Png et al., 2001). The encouragement for managers to broadly share information electronically, however, has not received unanimous acceptance because of concerns about inter-group conflict and information overload to groups not directly responsible for strategy-related decisions. Cook and Eining (1993) propose three elements of an information system that should be considered in the design of an information system: inputs, processors and outputs. The authors suggest the range of access to each information system element (inputs, processors and outputs) will lie along a “sharing” continuum from “need to know” to “complete access.” According to Cook and Eining (1993), one of the assumptions underlying sharing data and processing systems is that improved communication and decision making will result by, “...reducing the uncertainty associated with incomplete information (p.55).” Table 1-Panel B depicts the proposed relationship between national culture and the elements of information sharing infrastructure. It is proposed that high levels of uncertainty avoidance in a culture will result in the need to share information as widely as possible in an effort to reduce decision maker uncertainty. Conversely, in cultures with high levels of power distance access to information systems will not be as widely shared in order to maintain the power base of those in control of the organization. The table assumes the business strategy of the firm and organizational infrastructure decisions are fixed in order to focus on the specific impact of national culture on information technology.

52 Katz

Table 1: Panel A -Relationship between national culture and organizational infrastructure Dimension of Organizational Structure

Level of Uncertainty Avoidance High

Low

Level of Power Distance High

Low

Division of Work More functional

Less functional

More functional

Less functional

Specialization

More

Less

More

Less

Decentralization

Less

More

Less

More

Coordination

Lower

Higher

Lower

Higher

Formalization

Higher

Lower

Higher

Lower

Table 1: Panel B - Relationship between national culture and information system infrastructure Dimension of IT Infrastructure

Level of Uncertainty Avoidance

Level of Power Distance

High

Low

High

Low

Inputs

Shared

Not Shared

Not Shared

Shared

Processors

Shared

Not Shared

Not Shared

Shared

Outputs

Shared

Not Shared

Not Shared

Shared

APPLICATION OF THE FRAMEWORKS IN THREE CULTURES The following subsections provide a description of the differences between American, Japanese, and French organizational infrastructures. The U.S. and Japan were chosen because of the high level of international expansion common to both countries and because the two cultures are often located on opposite ends of the Hofstede (1980) measures of national culture. France was chosen as a representative country of the European Union and is typical of the social democracies found in the EU. The purpose is to explore how Table 1Panels A & B may be applied to distinctly different cultures. Table 2 displays the proposed organizational and information system differences in each country. The goal of Table 2, and the following narrative, is to demonstrate how managers may apply the relationships contained in Table 1 to specific cultures and nations.

Japan The Japanese culture, according to Hofstede (1980), is high in uncertainty avoidance and moderate in power distance. In Japan, the preferred organizational structure is functional (Subhash & Marks, 1993). Lincoln, Hanada and McBride (1986) conducted a study comparing fifty-five American and fifty-one Japanese manufacturing employees.

The Integral Role of Information Technology 53

Table 2: Comparison of organizational and information sharing infrastructures in three countries Dimensions of Organizational Structure

Japan

United States

France

More functional

More self-contained

More functional

Moderate High

Moderate Low

High Moderate

Taller structures Moderate to High Low

Flatter structures Low High

Taller structures High Low

Coordination 1. Task routines 2. Planning 3. Integration

High More formal Higher

Low Less formal Moderate

Moderate to High Less formal Moderate

Formalization

Moderate/High

Moderate/Low

High

IS Inputs

Shared

Not Shared

Not Shared

IS Processors

Shared

Shared

Not Shared

IS Outputs

Shared

Shared

Not Shared

Division of work Departments are: Specialization 1. Horizontal 2. Vertical Decentralization 1. Hierarchy 2. Span of control 3. Autonomy

While three-quarters of the functions common to U.S. and Japanese workers were assigned to specialists in Japanese plants, American firms favored work teams and cross training. Thus, cross-functional expertise seems to be greater in the United States than in Japan. With regard to uncertainty avoidance, in a comparative study of organizational characteristics between Japan and the United States, Yeh and Sagafi-Nejad (1987) contrasted the decision-making freedom of American workers to that found among workers in Japanese firms. The authors reported that American workers made their own decisions and their own rules more easily. This is consistent with other research emphasizing greater job autonomy in United States than in Japan reflecting the differences in the levels of uncertainty avoidance between the two cultures (Tsuda, 1985). By comparison, Japan is closer to France than to the United States in standardization of work processes. However, teamwork is also a very important aspect of the Japanese workplace, thereby moderating the level of standardization (Sai, 1995). In addition, Japanese workers tend to form cohesive teams with a leader who is clearly distinguished (Subhash & Marks, 1993). In contrast, the leader is more an arbitrator than in western teams. In order to reach a proper level of coordination, it is not unusual to see teamwork span departmental boundaries (Goldman, 1994). In Japan, communication is purposely ambiguous to encourage group responsibilities for tasks. Thus, organizational coordination is more implicit in Japan than in the United States where managers are encouraged to delegate specific tasks and hold subordinates accountable (Barsoux & Lawrence, 1990; Hortum & Muller, 1989; Lauenstein, 1985).

54 Katz

There are also differences in the cultural effects of information system design and access. For example, according to Bowonder and Miyake (1993), Japanese firms employ several key ways to ensure the sharing of information such as concurrent engineering, horizonal information flows, technology fusion, and corporate networking for integrating managerial functions and information systems. The authors report how Hitachi, Ltd. and Nippon Steel Corp. use integration processes to combine managerial control systems (organizational structure) and information flow structures (information system infrastructures) (Bowonder & Miyake, 1992). Finally, Palvia (1996) documents the Information Innovation Department at NKK Corp. which is involved in information technology and information systems to automate various processes at the company. At the heart of the automation process is the installation of a networked communication system to ensure everyone from the CEO to the clerical staff has complete access to all company information.

United States The United States scores relatively low on the uncertainty avoidance dimension but higher than Japan on the power distance scale. This means that by comparison to their Japanese counterparts, Americans tend to prefer a more divisionalized form of organizational structure (Mintzberg, 1983). The degree of delegation perceived by American managers is higher in the United States than by managers in Japan (Tsuda, 1985; Yeh & Sagafi-Nejad, 1987; Yeh, 1991). Japanese firms delegate less formal authority than their American counterparts. In the United States, national culture encourages managers to use decentralized organizational structures. American managers are encouraged to push decision making down the line, because they want to achieve lower-level worker involvement in operational decisions. American organizations employ more formal rules than Japanese organizations but tend to be less formal when considering expected informal behaviors (Lincoln, Hanada & McBride, 1986). The main difference between the U.S. and Japan regarding formalization of work is that Japanese organizations tend to use more labor intensive approaches, including administrative work and bookkeeping, than U.S. firms (Yeh & Sagafi-Nejad, 1987; Yeh, 1991). In particular, these two countries seem to have almost the same approach toward written standards and schedules. However, the managerial perception is different; in Japanese firms workers rely more on informal rules and information systems than their American counterparts. With regard to information system design and access, American managers tend to follow the strategic planning approach adopted by many U.S. companies—top down. Thus, according to Pant and Hsu (1999) even though information has emerged as an agent of functional integration, the traditional strategic planning process has limited the access to information inputs and outputs to senior managers. The authors cite a fundamental disconnect between information system planning and development because of the limited access to information flows and processors evident in U.S. firms. This is particularly important because the strategic issues most crucial to U.S. firms involve aligning information technology with business strategy and allowing line managers the authority to select information technology projects most beneficial to their organizational structures (Cortada, 1998; Leidner et al., 1999).

France Consistent with many of the social democracies in the European Union, the national culture of France reflects moderate levels of uncertainty avoidance and high levels of power

The Integral Role of Information Technology 55

distance. In France, a first-line supervisor derives authority by virtue of belonging to the management group. The manager gives orders because he or she has been chosen to do so, a choice that sets her or him apart from subordinate workers. This highlights the fact that in France the bureaucratic organizational structure is particularly popular (Mintzberg, 1983). According to Mintzberg (1983), France, with its preference for pyramidal structure, favors a strong specialization of its workforce. In France, organizations tend to have rigid structures and are highly centralized. Compared to management practices in the U.S., the French are not willing to delegate authority extensively. Aggressive standardization of work fits the French concept of bureaucracy. The type of coordination in France tends to be based on hierarchical authority (Barsoux & Lawrence, 1990; Mintzberg, 1983). Because of the preference for the bureaucratic organizational form, the most formalized country of the three countries chosen is France (Barsoux & Lawrence, 1990; Lauenstein, 1985). With regard to information system design and access, the information revolution has been compared to the French Revolution—very painful for the French people (Kaye, 1994). French managers tend to severely limit access to information system inputs, processors, and outputs. According to Kaye (1994), top French managers need to be more involved in understanding information technology and have the ability to differentiate between being business versus technical managers. The next section presents research propositions reflecting differences in organizational infrastructures based on national culture along with suggestions for future empirical study.

RESEARCH PROPOSITIONS AND IMPLICATIONS The firm’s external environment, size, and level of technological innovation will impact managerial decisions about the firm’s business strategy that in turn will affect the infrastructure of the organization. Despite these factors, recent research suggests that a particular culture will influence managerial decisions particularly with regard to the choice of internal structures (Gibson, 1994; Dologite et al., 1997) and adoption of information technology infrastructures (Png et al., 2001; Leidner et al., 1999). Therefore, it is likely that organizational infrastructures will be more effective in one country than in another depending on whether managers are aware of the relationship between business strategy and their infrastructure design decisions on the expectation of host country workers and supply chain partners. That is, given the business strategy chosen, the international manager needs to know how to design an organization based on the underlying national culture as well as how much to share access to information system elements. In the context of the firm’s chosen business strategy, Table 1-Panel A proposes how the five components of organizational structure will vary according to national culture, while Table 1-Panel B proposes how access to the three elements of information system infrastructure will vary by national culture. Based on the understanding of the firm’s goals and strategies, the tables are designed to guide functional managers in the development of organizational and information technology infrastructures in host countries and suggest research opportunities for future empirical verification. The tables propose that countries having high levels of power distance, controlling for business strategy, will favor functional organizational structures characterized by high

56 Katz

levels of centralization, low levels of delegation, and low levels of job autonomy. In these cultures, task specialization will tend to provide in-depth skill development but will result in limited cross-functional coordination. It is likely that workers in cultures with high levels of power distance will expect to have limited access to most elements of the information sharing systems. In such cultures, managers will favor hierarchical structures using rigid rules and emphasize high levels of formalization. Therefore, the following relationships are proposed. P1a In countries with high levels of power distance, host country workers will expect centralized decision making, low levels of delegation, and low levels of job autonomy. In such cultures, functional structures will be the most commonly used organizational structure. P1b In countries with high levels of power distance, host country workers will expect information system elements that are not shared at all levels of the organization. In such cultures, access to information system elements will be based on rank rather than on function in the organization. In countries having high levels of uncertainty avoidance organizational structures will tend to be more team-oriented, controlling for business strategy. Such structures are often characterized as self-contained unit structures. They will tend to have greater levels of team decision-making and lower levels of job autonomy. Cultures in this category will have written, as well as unwritten, rules that encourage informal communication. In addition, the high level of uncertainty avoidance will encourage sharing of information to ensure all members of the organization and supply chain partners, in some way, share the decisions made. Therefore, the following relationships are proposed. P2a In countries where high levels of uncertainty avoidance exist, host country workers will expect organizations that allow group decision making. In such cultures unit and matrix organizational structures will be most common. P2b In countries with high levels of uncertainty avoidance, host country workers will expect information system elements that are shared at all levels of the organization. In such cultures, access to information system elements will be based on organizational membership rather than rank. According to Knorr (1990), competitive advantage of world-class organizations will come from attention to seven dimensions: strategy and competitive positioning, information technology, organizational structure, performance measurement, cost structures, organizational skill development, and culture. The relationships previously discussed address methods to implement five of these dimensions. Recent research has begun to address issues raised with regard to propositions P1b and P2b relating to adoption of information technology differences based on national culture. Png et al. (2001) recently reported the results of a broad-based study assessing the adoption of telecommunication technology based on the firms’ home country. The authors report that firms located in higher uncertainty avoidance countries were less likely to adopt more innovative information technologies. Specifically, they indicated that a one-point increase in Hofstede’s uncertainty avoidance index would result in three percent lower likelihood of adopting more innovative information technology. However, the authors did not find a significant relationship between the level of power distance and information technology adoption. In addition, Leidner et al. (1999) surveyed managers using Executive Information Systems (EIS) in numerous organizations in three distinctly different cultures. They found significant differences in the impact of EIS on managerial decisions based on national culture. The authors concluded that information technology is used to reinforce decision

The Integral Role of Information Technology 57

making processes within a culture, underscoring the notion that, “cultural differences need to be understood before information technology developed for organizations in one country can be effectively implemented in organizations in another country” (p. 633). To encourage the empirical verification of the propositions, the following discussion for information management researchers and practitioners is offered. The underlying assumption of this discussion is that greater variation in organizational infrastructures exists between diverse national cultures than within a particular national culture because the differences in expectations of workers in their home country will be more similar than those of workers in a host country, controlling for the effects of business strategy.

Implications for Information Management Researchers There are several opportunities for management researchers to empirically test the propositions offered. First, the recent work by Nelson and Clark (1994) provides a framework for developing empirical assessment of our research propositions. They suggest the use of diads comprised of managers from two different cultures in which questionnaires can be administered in the managers’ native language. The questionnaires would assess differences in decision contexts and their impact on organizational and information technology infrastructure choices. An excellent example of this approach was used by Straub (1994) to assess the effect of national culture on information technology diffusion with a focus on the use of e-mail and FAX. In the case of our propositions, a beneficial empirical study would be designed as a decision making simulation measuring managerial preferences for organizing (functional, unit, or matrix structure) given different business strategies (low-cost versus differentiation) and level of information sharing for each IS element (inputs, processor and outputs). The simulation, a three-by-two-by-three design, would be administered to managers in the U.S., France, and Japan to measure their preferences for matching organizational decisions to the business strategy and national culture. The research methodology adopted by Harvey (1997) could be used as a model for this ethnographic approach. Another empirical test would involve using cross-national strategic alliances as a research site. This approach would provide an opportunity to assess how organizational infrastructures vary by national culture within the same organization and potentially in the same business strategy context. Comparing the organizational infrastructure characteristics based on venture ownership would provide an interesting test of the effect that cultural factors have on managers’ decisions to design the infrastructure of business units in different countries. That is, investigating the dominance of power distance over uncertainty avoidance in strategic alliance partners would provide useful information for international managers about how to structure organizations and information systems in different countries. This approach could be tested in a manner similar to that employed by Hill et al. (1998), in which a qualitative assessment of information technology transfer based on culture was accomplished using focus groups coupled with a field study. Under this approach, it would be possible to identify several strategic alliance partners operating in different countries, then interview focus groups of managers and workers in each location as to the fit between the national culture, business strategy, and infrastructures based on their location. The field study could be used to develop case studies regarding the organizational and information technology infrastructures in each cultural context. Finally, empirical verification of the research propositions would be possible using the approach employed by Straub (1994). In this approach, paired comparisons of workers in two cultures are measured and tested for statistical significance. In this approach, it would be advisable to experimentally control for business strategy and organizational structure,

58 Katz

while allowing the level of information sharing to be Likert scaled (1-7, low-high). The objective of the experiment would be to assess the validity of Table 1-Panel B, based on the national culture of the respondents. Each approach contains advantages and limitations. However, additional empirical examination of the issues discussed in this chapter relative to supply chain behavior is believed to be a rich topic worthy of further investigation.

Implications for Global Managers Managers need to keep in mind that organizational infrastructures form the heart of business strategy implementation (Andrews, 1987). The competitive environment, organizational mission, and business strategies form the foundation for the design of the firm. Since managers influence the organizational design choices, their organizational design decisions play a significant role in the performance of the organization. The framework presented in this paper argues that managers need to maintain flexibility when designing organizational systems, especially in a host country. After formulating strategies, managers should adjust the structure to be congruent with strategy implementation and the expectations of host country nationals. Thus, the needs of workers and managers must be compatible (Hinings et al., 1996). The link between strategy formulation and strategy implementation is organizational infrastructure. Managers should clearly define tasks that reflect worker skills, information sharing, hierarchical authorities, and anticipated structural changes. Managers should design an appropriate structure that takes into account feedback of performance, compensation systems, incentives, and systems of control. Managers should provide a structure that gives an optimal chance for the task to be performed by implementing clear plans of action and by putting into place efficient systems of information and coordination consistent with cultural differences guiding worker expectations. For example, in the U.S. a self-contained unit structure is likely to be the most effective kind of structure based on worker expectations. By comparison, Japan and France display stronger levels of uncertainty avoidance and power distance than the United States. Thus, we believe it will be easier for Japanese and French managers to integrate businesses since both countries have similar preferences for organizational design factors but more difficult to integrate information systems because of their different approaches in sharing access to information system elements. On the other hand, the United States manager will likely have greater difficulties working with Japanese or French managers because the nature of the cultural differences will have an impact on how information is shared and how workers are organized.

A Few Final Thoughts This chapter focuses on the organization and to some extent members of the supply chain linked by common goals. However, it is important to acknowledge that external and internal factors may independently impact the design and success of organizational infrastructures. These might include the environment, the size of the firm, the technology available, and the goals of the firm’s dominant coalition. These factors need further consideration. In addition, much of this chapter assumes that business strategies of the firm are an externality to managerial decisions. That is, most managers will not be able to substantially change the business strategy of the firm but will have the authority to make infrastructure decisions. In this regard, Hofstede (1991) suggests six questions that researchers should take into account. First, is the goal of the prospective SBU to be more process-oriented or results-oriented? Second, should the structures be employee-oriented

The Integral Role of Information Technology 59

or job-oriented? Third, do employees derive their identity from the organization or from the type of jobs? Fourth, is the system for organizing closed or open to interaction with the environment? Fifth, is organizational control tight or relaxed? Finally, should the units of the structure be market-driven or more dependent on rules (normative units)? (Hofstede, 1991). These additional questions should help us better understand whether the choice of infrastructures prior to expanding globally is a systematic determinant of international success.

REFERENCES Aldrich, H. (1979). Organizations and Environments. Englewood Cliffs, NJ: Prentice Hall. Andrews, K. (1987). The Concept of Corporate Strategy. Homewood, IL: Irwin. Barker, R. (1993). Information system development in a global environment. Business Forum, 18(1-2), 57-59. Barsoux, J. and Lawrence, P. (1990). Management in France. London: Cassell. Betz, F. (2001). Executive Strategy: Strategic Management and Information Technology. New York: John Wiley & Sons. Blackburn, R. (1982). Dimensions of structure: A review and reappraisal. Academy of Management Review, 7, 59-66. Bowersox, D., Closs, D. and Stank, T. (2000). Ten mega-trends that will revolutionize supply chain logistics. Journal of Business Logistics, 21(2), 1-16. Bowonder, B. and Miyake, T. (1993). Japanese innovations in advanced technologies: An analysis of functional integration. International Journal of Technology Management, 8(1-2), 135-156. Bovel, D. and Martha, J. (2000). From supply chain to value net. Journal of Business Strategy, 21(4), 24-28. Broderick, R. and Boudreau, J. (1991). The evolution of computer use in human resource management. Human Resource Management, 30(4), 485-508. Calori, R., Lubatkin, M. and Very, P. (1994). Control mechanisms in cross-border acquisitions: An international comparison. Organization Studies, 15(3), 361-379. Carillo, P. and Kopelman, R. (1991). Organization structure and productivity. Group and Organization Management, 16(1), 44-60. Chandler, A. (1962). Strategy and Structure: Chapters in the History of the American Industrial Enterprise. Cambridge: MIT Press. Cheung, H. and Burn, J. (1994). Distributing global information systems resources in multinational companies–A contingency model. Journal of Global Information Management, 2(3), 14-27. Child, J. (1972). Organization structure, environment, and performance: The role of strategic choice. Sociology, 6(1), 1-22. Cook, L. and Eining, M. (1993). Will cross functional information systems work? Management Accounting, 74(8), 53-57. Cortada, J. (1998). Focus on the essentials. Journal of Business Strategy, 19(6), 27-31. Cross, G. (2000). How e-business is transforming supply chain management. Journal of Business Strategy, 21(2), 36-39. Daft, R. (1983). Organization Theory and Design. St. Paul, MN: West Publishing. Daft, R. and Steers, R. (1986). Organizations: A Micro/Macro Approach. Glenview, IL: Scott, Foresman.

60 Katz

Day, J. and Wendler, J. (1998). The new economics of organization. McKinsey Quarterly, (1), 4-18. DeLisi, P. (1990). Lessons from the steel axe: Culture, technology, and organizational change. Sloan Management Review, 32(1), 83-93. Dessler, G. (1976). Organization and Management: A Contingency Approach. Englewood Cliffs, NJ: Prentice-Hall. DiBella, A. (1996). Culture and planned change in an international organization: A multilevel predicament. International Journal of Organizational Analysis, 4(4), 352-372. Dologite, D., Fang, M., Chen, Y., Mockler, R. and Chao, C. (1997). Information systems in Chinese state-owned enterprises: An evolving strategic perspective. Journal of Global Information Management, 5(4), 10-21. Drucker, P. (1946). Concept of the Corporation. New York: The John Day Co. Ein-Dor, P., Segev, E. and Orgad, M. (1993). The effect of national culture on IS: Implications for international information systems. Journal of Global Information Management, 1(1), 33-44. Emery, F. and Trist, E. (1965). The causal texture of organizational environments. Human Relations, 18, 21-36. Floyd, S. and Wooldridge, B. (1990). Path analysis of the relationship between competitive strategy, information technology, and financial performance. Journal of Management Information Systems, 7(1), 47-64. Forehand, G. A. and Gilmer, B. (1965). Environmental variations and studies of organizational behavior. Psychological Bulletin, 62, 361-82. Galbraith, J. (1973). Designing Complex Organizations. Reading, MA: Addison-Wesley. Galbraith, J. and Nathanson, D. (1978). Strategy Implementation: The Role of Structure and Process. St. Paul, MN: West Publishing. Garfield, M. and Watson, R. (1997). Differences in national information infrastructure: The reflection of national cultures. Journal of Strategic Information Systems, 6, 313-337. Gibson, C. (1994). The implications of national culture for organization structure: An investigation of three perspectives. Advances in International Comparative Management, 9, 3-38. Goldman, A. (1994). Doing Business with the Japanese. Albany, NY: State University of New York Press. Hall, R. (1977). Organizations: Structure and Process. Englewood Cliffs, NJ: Prentice Hall. Harrigan, K. and Dalmia, G. (1991). Knowledge workers: The last bastion of competitive advantage. Planning Review, 19(6), 4-9. Harris, P. and Harris, D. (1988). Decision making for new work culture. Management Decision, 26(5), 508. Harvey, F. (1997). National cultural differences in theory and practice: Evaluating Hofstede’s national cultural framework. Information Technology & People, 10(2), 132-146. Henderson, J. and Venkatraman, N. (1991). Understanding strategic alignment. Business Quarterly, 55(3), 72-78. Hill, C., Loch, K., Straub, D. and El-Sheshai, K. (1998). A qualitative assessment of Arab culture and information technology transfer. Journal of Global Information Management, 6(3), 29-38. Hinings, C., Thibault, L., Slack, T. and Kikulis, L. (1996). Values and organizational structure. Human Relations, 49(7), 885-916. Hofstede, G. (1980). Culture’s Consequences: International Differences in Work-Related Values. Beverly Hills, CA: Sage.

The Integral Role of Information Technology 61

Hofstede, G. (1991). Cultures and Organizations: Software of the Mind. New York, NY: McGraw-Hill. Hortum, M. and Muller, L. (1989). Management and culture. Supervision, 50(1), 14-17. Jaeger, A. (1984). The appropriateness of organization development outside North America. International Studies of Management and Organization, 14(1), 23-35. Jones, G. (1995). Organizational Theory. Reading, MA: Addison-Wesley. Katz, J., Werner, S. and Brouthers, L. (1999) Does winning mean the same thing around the world? National ideology and the performance of global competitors. Journal of Business Research, 44(2), 117-126. Katz, J., Zarzeski, M. and Hall, J. (1997). The importance of strategy to global competitors: The strategy-performance relationship across cultures. Journal of Business Strategies, 14(2), 126-140. Kay, D. (1994). Managing the information revolution: Lessons for business from the French Revolution. Futures, 26(4), 416-429. Kitchell, S. (1995). Corporate culture, environmental adaptation and innovation adoption: A qualitative/quantitative approach. Journal of the Academy of Marketing Science, 23(3), 195-205. Knorr, R. (1990). Managing resources for world-class performance. Journal of Business Strategy, 11(1), 48-50. Lauenstein, M. C. (1985). Strategic planning in Japan. Journal of Business Strategy, 6(2), 78-84. Lee, H. and Whang, S. (2000). Information sharing in the supply chain. International Journal of Manufacturing Technology & Management, 1(1), 79-93. Leidner, D., Carlsson, S., Elam, J. and Corrales, M. (1999). Mexican and Swedish managers’ perceptions of the impact of EIS on organizational intelligence, decision making, and structure. Decision Sciences, 30(3), 633-658. Lewis, D. and Shea, T. (1996). The impact of culture on management information needs. International Journal of Management, 13(2), 241-248. Lin, B., Vassar, J. and Clark, L. (1993). Information technology strategies for small businesses. Journal of Applied Business Research, 9(2), 25-29. Lincoln, J. R., Hanada, M. and McBride, K. (1986). Organizational structures in Japanese and U.S. manufacturing. Administrative Science Quarterly, 31(3), 338-364. Lynn, B. (1999). Culture and intellectual capital management: A key factor in successful ICM implementation. International Journal of Technology Management, 18(5-8), 590603. Mascitelli, R. (1999). The Growth Warriors: Creating Sustainable Global Advantage for America’s Technology Industries. Northridge, CA: Technology Partners. Mintzberg, H. (1983). Structure in Five: Designing Effective Organizations. Englewood Cliffs, NJ: Prentice-Hall. Nelson, K. and Clark, T. (1994). Cross-cultural issues in information systems research: A research program. The Journal of Global Information Management, 2(4), 19-29. Ouchi, W. (1978). The relationship between organizational structure and organizational control. Administrative Science Quarterly, 22, 95-113. Palvia, P. (1996). The expert’s opinion. Journal of Global Information Management, 4(1), 32-33. Pankratz, H. (1991). Strategic alignment: Managing for synergy. Business Quarterly, 55(3), 66-71.

62 Katz

Pant, S. and Hsu, C. (1999). An integrated framework for strategic information systems planning and development. Information Resources Management Journal, 12(1), 15-25. Png, I., Tan, B. and Wee, K. (2001). Dimensions of national culture and corporate adoption of IT infrastructure. IEEE Transactions on Engineering Management, 46(1), 36-45. Porter, M. (1990). The Competitive Advantage of Nations. New York: Free Press. Porter, M. (1985). Competitive Advantage. New York: Free Press. Pugh, D., Hickson, D., Hinnings, C. and MacDonald, K. (1969). An empirical taxonomy of structure of work organizations. Administrative Science Quarterly, 14, 115-126. Pun, K. and Lee, M. (2000). A proposed management model for the development of strategic information systems. International Journal of Technology Management, 20(3-4), 304-325. Routamaa, V. (1985). Organizational structuring: An empirical analysis of the relationships and dimensions of structures in certain Finnish companies. Journal of Management Studies, 22, 498-522. Sai, Y. (1995). The Eight Core Values of the Japanese Businessman. New York: International Business Press. Shore, B. (1996). Using information technology to coordinate transnational service operations: A case study in the European Union. Journal of Global Information Management, 4(2), 5-14. Shore, B. and Venkatachalam, A. (1996). Role of national culture in the transfer of information technology. Journal of Strategic Information Systems, 5(1), 19-35. Shore, B. and Venkatachalam, A. (1995). The role of national culture in systems analysis and design. Journal of Global Information Management, 3(3), 5-14. Siddall, P., Willey, K. and Tavares, J. (1992). Building a transnational organization for BP Oil. Long Range Planning, 25(1), 37-45. Straub, D. (1994). The effect of culture on IT diffusion: E-mail and FAX in Japan and the U.S. Information Systems Research, 5(1), 23-47. Subhash, D. and Marks, N. E. (1993). Japanese Business: Cultural Perspectives. Albany, NY: State University of New York Press. Tsuda, M. (1985). The future of the organization and the individual to Japanese management. International Studies of Management and Organization, 15(3), 89-125. Udell, J. (1967). An empirical test of hypotheses relating to span of control. Administrative Science Quarterly, 12, 420-39. Venkatraman, N., Henderson, J. and Oldach, S. (1993). Continuous strategic alignment: Exploiting information technology capabilities for competitive success. European Management Journal, 11(2), 139-149. Yeh, R. (1991). Mangement practices of Taiwanese firms: As compared to those of American and Japanese subsidiaries in Taiwan. Asia Pacific Journal of Management, 8(1), 1-14. Yeh, R. and Sagafi-Nejad, T. (1987). Organizational characteristics of American and Japanese firms in Taiwan. Best Papers Proceedings of the Academy of Management, 111-115.

The New Realities of Online Trading in the New Global Marketplace 63

Chapter V

The New Realities of Online Trading in the New Global Marketplace Bridget Allgood University College Northampton, UK Stock brokering services are available via the Internet, and investors in many countries can currently experience the benefits of Web-based share trading. Internet-based share trading is far more than completing the same transaction over the Internet. The Internet contains a wealth of information, and many sites provide sophisticated facilities that allow the investor to research and monitor investments. This paper examines how Internet based share trading differs from more traditional forms of share trading. It then explores how Internet-based share trading is changing investment practices of private individuals and goes on to discuss how (with reference to Michael Porter’s work on competition) the global nature of the Internet-based share trading will impact upon the players in the stock brokering competitive environment.

INTERNET-BASED SHARE DEALING The Internet has brought about a new opportunity for business to trade with customers electronically and fundamentally changes the way business interacts with clients (Nath et al., 1998). The stock brokering sector is particularly suited to Web-based trading, as physical goods do not need to delivered, thus eliminating the logistics problems which many ecommerce retailers experience. Internet based share-dealing services enable investors to check real time share prices and execute trades immediately (online). Payment for the shares is completed through direct debit transfer from an individual’s bank account, or a debit is made from a holding account which has previously been set up with the broker. Wap technology has emerged in recent years to allow users of Wap-enabled mobile phones to conduct share transactions. Although these systems are not as rich or sophisticated as the Internet-based services, they offer a convenient means of trading for many investors. The trend is for existing Internet-based brokers to provide additional Wap services for their clients. Copyright © 2002, Idea Group Publishing.

64 Allgood

Internet share dealing is often referred to as “online trading.” However, to avoid confusion with other types of online share trading systems it is referred to as Internet-based shared dealing or Web-based share dealing within this paper.

Investor Benefits Internet-based share dealing provides an improved service for investors over more traditional dealing methods. The investor is able to deal at the price viewed immediately, whereas using more traditional dealing services, an investor will often have to wait in a telephone queuing system to get through to the dealing desk and when trading may have to wait for the price of a trade to be confirmed. The Internet investor also benefits from 24-hour, 7-day-a-week access to placing orders (although outside of trading hours deals will need to be placed as “limit orders” where a limit price is given). Intense competition for clients has brought prices down, and Web-based share dealing services are typically cheaper than other forms of “execution only” dealing service. Lower prices, combined with the provision of company information services and portfolio services, make Internet share dealing an attractive option for investors who wish to use an “execution only” service in an informationrich environment. The Internet also gives investors access to deal in new products, such as equities traded on foreign exchanges, and futures and options.

Broker Benefits From the stockbroker’s viewpoint, Internet share dealing is beneficial as it allows the broker to reach a wider customer base and potentially increase market share. Webbased share trading on the Internet may well attract people to dealing who, have never dealt before, thus allowing brokers to reach new groups of customers who by virtue of using the Internet, represent a higher than average income group (Ng et al., 1998). Reduced operating costs are also seen as a benefit of Web-based share trading (Carr, 1999). The level of share trading at any time is influenced by market conditions; there are days and times when trading is particularly heavy and traditional telephone-based services have difficulties in responding to these demand peaks, resulting in investors being unable to establish contact with their brokers. Internet-based share dealing offers a cost-effective solution to the demand peak problem (although steps need to be taken to ensure good availability of services). Customer transaction information is input by the customer, then automatically passed straight through to the back office systems, then through to the retail service provider; this “straight through” processing means less staff are needed to handle the transactions. Web-based trading also has the effect of increasing the number of trades completed by Internet account holders, which generates more commission for brokers. Use of the Internet allows stock-brokering companies to build up a wealth of customer information, which can be used to focus marketing material on distinct customer profiles.

Internet-Based Share Brokers The market in the United Kingdom (UK) is growing rapidly. (In April 1999 there were just six UK based organizations providing Internet-based brokering services, now, two years later, there are over twenty-five organizations competing in this marketplace.) There are three main categories of players in this market; firstly, the traditional banks and stockbrokers who have ventured into Web-based trading; secondly newcomers who have not previously operated in this marketplace but have seen opportunities in this area; and thirdly, interna-

The New Realities of Online Trading in the New Global Marketplace 65

tional players who have run successful Web-based trading operations in other countries and have decided to enter the UK marketplace. The United States (U.S.) market is well established, and there are more than 100 Internetbased brokering operations; many of the organizations only offer Internet dealing. It is estimated that online trading accounts for 40% of all share trades in the U.S. (Temple, 1998). It is expected that, as more people gain access to the Internet and more stockbrokers offer Internet-based share trading facilities, this style of share dealing will dramatically increase in the near future. Outside of the U.S. and UK, there are organizations operating Internet share dealing services; for example tradex.com (http://www.tradex.com) is a South African-based stockbroker specializing in Internet share dealing.

RESEARCHING AND MONITORING INVESTMENTS Financial information is ideally suited to delivery via the Internet as it can deliver rapidly changing information such as merger news and stock prices. Stockbrokers and financial institutions have traditionally had access to a wide range of information supplied through their own systems. In the past the small private investor has been less fortunate than the institutional investor and has needed to rely upon newspapers and investment advice from stockbrokers. The wealth of information, which is now available on the Internet, means that the private investor can access a range of up-to-date financial information which was previously only available to financial institutions. This information relating to news items, global markets and share prices, and trading statistics can be used to inform the private investor’s investment decisions. Information on shares, futures, options, investment trusts, unit trusts, bonds, and gilts is also available on the Internet.

Share Prices and News The investor can easily obtain information on share prices, company news, and good quality research information from a wide range of sites. Many financial sites operate at two levels, providing both free information and additional subscription-based information. The Interactive Investor site (http://www.iii.co) gives equity prices for stocks traded on the London Stock Exchange as well as other major international exchanges. The site also provides company information, historic stock price information, investment advice, and much more. More traditional sources of information, such as newspapers and business magazines, are also represented on the Internet and often incorporate powerful news searching facilities. For example the UK-based, Financial Times site (http://www.ft.com) provides the investor with the facility to search current and back issues of the newspaper for articles on a specified topic or company. A number of sites provide chat/message boards, where participants can post financial queries and comments and also respond to messages, thus forming virtual communities.

Personalized Information The Internet provides the facility to view personalized information. Thus, an investor can visit a Web site and see stock prices and company news that relate to their investment interests. Portfolio services also exist on the Internet; investors can set up their investment portfolios and then return to the Web site to view their portfolio valuations with current stock

66 Allgood

prices at any time. Web technology makes it possible for information to be automatically delivered and appear on a user’s PC; thus, if a share price falls below a specified limit or say, news on a particular company is reported upon, the information will be delivered to the desktop.

Personal Finance Programs There is a wide range of Personal Finance Programs designed to run on users’ PCs, which can automatically access share prices and news on the Internet. In the future, it is envisaged that personal finance programs running on investors’ personal computers will use information fed from the Internet to continually monitor and automatically take action to maximize the return on a portfolio or maintain a balanced, diversified portfolio (Eckett, 1997). Personal data such as risk preferences, preferred investment strategy, life expectancy, income, and expenditure plans might be used to personalize the program so that decisions reflect an individual’s needs. As stock markets are global in nature and respond to each other on a daily basis, with price changes in one market affecting other markets, it is also likely that these future investment programs will make and carry out investment decisions based upon global market events.

Information Services Provided by Internet-Based Stockbrokers To attract and retain investors, Internet-based brokers need to develop value added information services for their customers. Some brokers have chosen to collaborate with business partners to deliver these information services. For example, Charles Schwab (http:/ /www.schwab.co.uk), a leading U.S. and UK-Web-based stockbroker offers portfolio valuations and also access to the Reuters Company news and information service.

CHANGING INVESTMENT METHODS Access to improved market information and international information, combined with the potential for Internet trading, may well change patterns of investment, encouraging investors to explore the possibilities of global investing, day trading, and derivative products.

Investing Directly in Global Markets The global nature of the Internet means that international markets that in the past could not be monitored and accessed by the private investor are now reachable. An individual may well wish to diversify his/her portfolio by investing directly into international markets, rather than resorting to more traditional investment methods through intermediaries such as unit trust managers. Thus, an investor, say, based in the UK and wanting to buy U.S. stocks can execute the deal with a Web-based broker based in the U.S., rather than paying the traditionally high commission charges. To deal internationally, an investor will need to register with the new broker and set up an account in the appropriate currency before trading can take place. However, as global markets develop, stockbrokers acquire an international presence, and the systems technology develops, these transactions may someday be dealt with automatically. Global investing is becoming a reality as it becomes easier and cheaper; in future years it will become a natural extension of the current facilities that the Web-based investor utilizes.

The New Realities of Online Trading in the New Global Marketplace 67

Time Proximity to Markets International time differences may also encourage investors to deal at a global level. For example, investors who are in full-time employment within the UK will find the American stock exchanges are just halfway through the trading day when they return from work, whereas the London stock exchange has closed. Thus an investor based in the UK, although geographically closer to the London stock exchange, may well be better situated to deal with stocks that trade on the American exchanges.

Day Trading Internet-based dealing may change the time span in which investors deal. As it becomes physically easier to trade shares, “day trading” may prevail. This is where investors buy and sell shares within a short time frame which are being driven up by the momentum of the market (perhaps only holding the equities for a matter of minutes). This method of trading is particularly suited to electronic markets, where investors have instant access to price information and news (Investors Chronicle, 2001).

Derivative Products Improved market information, Web-based educational material, and easier access to trading may also create demand for more sophisticated traded products, such as futures and options. A limited number of brokers already give access to these products.

CHANGES IN THE STOCK BROKERING COMPETITIVE ENVIRONMENT Web-based trading will affect the competitive environment in which stock brokering companies operate. Porter’s (1979) five forces model provides a useful framework for examining the potential impact of Internet-based trading within the competitive stock brokering marketplace. Porter describes how the nature and degree of competition within an industry depends on five forces: the threat of new entrants, threat of substitute services, bargaining power of customers, bargaining power of suppliers, and the jockeying for position among current competitors (see Figure 1).

New Entrants Web-based share trading has attracted many organizations: traditional stockbrokers wishing to exploit this new channel, foreign stockbrokers wishing to expand into global operations, organizations who already have a presence in the financial services market, such as building societies, and new start-up companies with an Internet-only presence. Internet-based financial information providers who attract customers regularly to their Websites are in an ideal position to enhance the service they offer by also offering share trading facilities. This may be through collaboration with existing stockbrokers or by setting up the venture themselves. The Web offers a cost-effective means of reaching customers. The influx of new players into this marketplace has intensified competition, as firms battle to attract and retain customers. There are a number of barriers which new entrants face. For example, investment firms need authorization and vetting by the appropriate regulatory body to conduct business

68 Allgood

Figure 1: The potential impact of Internet-based trading on the competitive place (adapted from Porter, 1979) Threat of New Entrants

Organizations providing Internet-based dealing services Power of Suppliers

Power of Buyers Increased bargaining power of investors

Rivalry amongst existing stockbrokers

Market makers who quote prices at which they will buy and sell shares, thus creating a share market.

Threat of Internet-based dealing providing a value added, low-cost service

Threat of Substitute Products

within a country (i.e., the Financial Services Authority to conduct investment business in the UK). Also new entrants may find it difficult to gain new customers in a marketplace where customers are loyal to “their” brokers.

Substitute Service Internet-based share trading offers an attractive alternative to current dealing arrangements for many investors. They are likely to be price sensitive to commission charges due to the fact that they are buying a standard product (e.g., 100 British Telecom shares). Also, as investors are seeking to make money from share dealing, they are by nature price sensitive and will be attracted to the lower-cost substitute services. This is coupled with the fact that Web-based share dealing services offer an improved service in an information-rich environment over other types of execution-only services (such as telephone dealing), which means that Internet dealing will be attractive to investors. However, there are some switching considerations that may deter customers from moving their business to an Internet-based broker. The investor will need to engage in some administration to set up a new account, and if the investor wishes to move his/her existing portfolio, it may be costly to move share holdings which are held electronically from one broker to another. However, a recent survey showed that UK online investors have an average of 2.7 accounts each (Gomez, 2001), showing that investors can be motivated to try new services.

The New Realities of Online Trading in the New Global Marketplace 69

Bargaining Power of Suppliers Within the share-trading environment there are a number of retail service providers (RSPs) and market makers that supply and buy shares at slightly differing prices to the brokers, thus creating a share market. These supplier organizations may pose a threat to existing brokers if they integrate into the brokering business. With the advent of online trading, it becomes easy to perceive a system where investors connect directly with the suppliers, cutting out the stockbroker middleman and thus challenging the traditional stockbroker-customer relationship.

Bargaining Power of Investors Cost comparison between online brokers can be difficult because of the wide variety of charging structures in existence. Some brokers charge flat fees, others percentages, and there may also be sliding scales and administration charges. The Internet is a powerful tool which investors can use to search for brokers offering the best price structures. It is quick and easy to search for substitute products using the Web. Intermediaries on the Internet have emerged who act as trading posts and actively search for “best buys” in terms of commission paid to stockbrokers for a given share transaction. Investors may also demand access to multiple exchanges and multiple products (such as derivatives). Price transparency and demand for new products will lead to greater competition within the stockbroking industry, leading to the need for stockbroking organizations to develop ways in which to add value to the customer experience in order to differentiate themselves.

Rivalry between Competitors Internet-based share dealing services will deliver increased capacity to the market place; this new capacity in the market will increase competition. Web based share dealing will also attract new entrants to the marketplace and open up local markets to global competition. These factors are likely to increase rivalry between competitors and have a profound impact on the stockbroking industry.

GAINING AND SUSTAINING COMPETITIVE ADVANTAGE In the changing global marketplace, stockbroking organizations need to consider when and how to move into Internet-based trading to gain and maintain a competitive advantage. Initially the fact that a company is providing Internet-based share trading is enough to differentiate it from its competitors who do not offer this service and give it a competitive advantage. As rivals respond by also delivering Internet-based trading services, first movers will need to carefully consider how to preserve their competitive edge. To be effective, a company must be able to both gain and sustain competitive advantage. Keeping one step ahead of rivals is important, and an organization needs to continually improve itself to keep ahead of the game (Porter, 1990). Information technology will make it feasible for stockbrokers to look beyond national boundaries, compete internationally in the global marketplace, and offer new products. If a company can successfully adopt a “global strategy” and compete in the new international

70 Allgood

marketplace, it can gain significant competitive advantage (Porter, 1986). To compete internationally, stockbrokers may need to form alliances with new business partners. Information technology will make it possible for organizations to develop sophisticated value-added information and portfolio management services, which will serve to differentiate the organization from the competition. Increasing globalization will intensify competition, as all suppliers can, by utilizing the Internet, and gaining access to global markets. Geographic proximity will no longer be a barrier to entry. The challenge of survival within the global marketplace will mean that brokers will need to apply information technology in innovative ways to differentiate themselves to the customer and to maintain a market share. As technology is changing rapidly, the technology itself will provide the opportunity for firms to innovate rapidly and so maintain a competitive edge (Porter and Millar, 1985). Global competition will mean that companies who do not choose to expand their operations beyond national boundaries will have to compete with those who have expanded in this way. This is already evident in the United Kingdom where a number of United Statesbased stockbrokers have established a presence in the UK market. Porter (1996) discusses the problems facing managers when developing a strategy in a business undergoing revolutionary technological change. Managers face uncertainty about customer attitudes, customer needs, services most desired by customers, and how to deliver the services with the new technology. However, early innovators gain from obtaining an earlier and clearer view of what the emerging buyer needs.

BARRIERS TO ELECTRONIC TRADING There are a number of issues that concern both investors and organizations when embracing the new technology and will influence the development and take-up of new initiatives.

Security and Privacy Issues Investors may be unwilling to participate in global Internet-based share dealing because of security concerns. In the new digital era of global paperless transactions, with large sums of money flowing across international boundaries, security in terms of the confidentiality and integrity of the information flowing across the Internet and within the receiving systems which handle the transactions is of paramount importance. Encryption techniques can effectively be used to encode the data and make it secure, but lack of an international standard for encryption poses an obstacle to electronic trade across international boundaries (Edouard and White, 1999). Security fears about electronic payment systems are frequently cited as one of the major reasons the Internet has been slow to expand commercially (Loh and Ong, 1998).

Trust and Relationships In the new Internet world, investors will be faced with unfamiliar companies without a physical presence offering share dealing facilities. Internet dealing is also characterized by paperless trading methods. This means that online trading investors will no longer hold share certificates, as shares ownership records will be stored electronically. Investors therefore need to have confidence that the “virtual brokers” will execute and hold the shares for them.

The New Realities of Online Trading in the New Global Marketplace 71

It is important that the investor can verify independently certain information about the company with which they are dealing, inorder to check the company registration, location, and regulatory framework in which it operates. Regulatory bodies (such as the Securities and Futures Commission in Hong Kong and the Financial Services Agency in the UK) can be used to check out companies who provide Internet-based share dealing.

Availability of Service The share trading market can be very volatile: share prices may fall sharply over a short time frame. When shares are held in a nominee account with an online broker, the investor has to deal through that broker. Therefore, brokers need to ensure that their systems are reliable and can deal with large peaks. From the customer’s point of view, the provision of an alternative telephone-based service to access their account is desirable if net access has failed. However, many Web-based stockbrokers are able to cut staff and transaction costs (which are then reflected in a lower commission) by electronically dealing with the transaction. This subsequently means they do not have the staffing to provide an alternative telephonebased service in times of difficulty.

Tax and Regulations The international nature of transactions means that the tax laws, consumer laws, and regulations which an investor is familiar with in his/her own country will not apply in other countries. These international legal, tax, and regulatory problems have yet to be dealt with by governments to ensure the customer is both protected and understands the laws covering Internet transactions. Effective regulation and enforcement of this regulation is difficult in an environment which is both global and rapidly changing. The OECD is instrumental in providing a forum for countries to get together to exchange views and develop frameworks and policies which will support this new way of trading (OECD, 1997). Also, the International Organization of Securities Commissions, through its 135-member agencies tries, to ensure that the national regulatory bodies work together to improve international regulation of the financial markets. Regulations and tax issues will influence both an investor’s willingness to participate in global trading and also the willingness of firms to provide access into new markets. Due to the regulations, and tax and legal issues, the way in which shares are bought and sold may differ significantly across country boundaries. For example, share purchases of UK equities incur a 0.5% stamp duty. The Internet gives rise to the possibility of a new type of “virtual” organization, which is established in cyberspace and therefore outside of existing legislative, tax and regulatory frameworks.

CONCLUSION Looking to the future, the Internet provides a route for individuals to expand their investment horizons and for brokers to reach out to new global markets. Internet technology demolishes the geographical boundaries of local markets, and geographical proximity to customers or markets will no longer restrict trading. Internet-based share trading is more than simply automating today’s telephone-based execution services. Investors are looking to the

72 Allgood

Internet to enable them to have easy access to multiple exchanges and multiple instruments transacting through one broker using one currency (Flenner, 2001). In addition to this, investors expect added value in the form of company information related to their personal portfolios and access to the same tools as the professional traders. Internet-based share dealing has the potential to change the dynamics of the stockbroking industry, due to the ease in which global markets can now be reached. Initially brokers will gain competitive advantage simply through providing Web-based dealing services; however as more Internet-based dealing services emerge, brokers will have to carefully assess how to maintain their competitive position. Companies which anticipate the power of information technology will be in control of events. Companies that do not respond will be forced to accept changes that others initiate and will find themselves at a competitive disadvantage. (Porter and Millar, 1985, p. 160). Global Internet-based share trading raises a number of issues that are of concern to both investors and stockbroking organizations. Investors will need to trust Web-based share dealing and have confidence in the new “virtual brokers” operating in the Web-based world. Existing legislation, taxation, and regulations were not designed to cope with the new challenges Internetbased share trading brings and have yet to be addressed by governments.

REFERENCES Carr, R. (1999). Ready to make your fortune as an E-trader ? Investors Chronicle, (128)1626, 14-15. Eckett, S. (1997). Investing Online. Pitman. Edouard, N. and White, W. (1999). The Development of the Internet and the Growth of ECommerce. London: Management Consultancies Association. Flenner, I. (2001). Survey online investing–Access all areas using new technology. Investors Chronicle, 136(1737), 86-88. Available from: http://globalarchive.ft.com. Accessed August 8, 2001. Gomez. (2001). The State of Online Investing in the UK, Gomez Consulting. Available from: http://www.uk.gomez.com. Accessed August 8, 2001. Investors Chronicle. (2001). Survey: Day trading–The difference a day makes, 136(1731), 8486. Available from: http://globalarchive.ft.com. Accessed August 8, 2001. Loh and Ong. (1998). The adoption of internet based stock trading: a conceptual framework and empirical results. Journal of Information Technology, 13, 81-94. Nath, R., Akmanligil, M., Hjelm, K., Sakaguchi, T. and Schultz, M. (1998). Electronic commerce and the Internet: Issues, problems and perspectives. International Journal of Information Systems, 18(2), 91-101. Ng, H., Pan, Y. and Wilson, T. (1998). Business use of the World Wide Web: A report on further investigations. International Journal of Information Management, 18(5), 291-314. OECD. (1997). OECD policy brief. Electronic Commerce, (1). Porter, M. (1979). How competitive forces shape strategy. Harvard Business Review, 137-145. Porter, M. and Millar, V. (1985). How information gives you competitive advantage. Harvard Business Review, 149-160. Porter, M. (1986). Changing patterns of international competition. California Management Review, 28(2), 9-40.

The New Realities of Online Trading in the New Global Marketplace 73

Porter, M. (1990). The competitive advantage of nations. Harvard Business Review, 73-93. Porter, M. (1996) What is strategy? Harvard Business Review, November-December, 61-78. Temple, P. (1998). Instant execution. Investors Chronicle, 124(1579), 80.

74 Moores & Gregory

Chapter VI

The Impact of Culture on the Development of Information Systems: A Case Study Trevor T. Moores University of Nevada Las Vegas, USA Frank H. Gregory Independent Consultant, Thailand

The development of an information system typically requires debate between interested parties. In particular, debate between users is meant to define a common set of functional requirements. The applicability of this approach depends, however, on the willingness of participants to enter into such an open discussion. However, while holding differing, perhaps conflicting views, is seen as acceptable in Western cultures, in Eastern cultures more importance is placed on social consensus and aligning one’s views with that of the group. This paper reports on a case study that highlights the problems of using debate as an analysis tool in an Eastern culture, namely, Hong Kong. The case study involves the use of Soft Systems Methodology to guide a feasibility study for a marketing system within Hongkong Telecom. Three main problems were identified: 1) Group discussions were avoided; 2) Interviews were conducted in multiple languages; and, 3) High staff turnover made it difficult to develop and maintain mature stakeholder views. The study suggests that culturally sensitive development methods are needed to ensure culturally appropriate ways of developing an information system.

INTRODUCTION Culture is often defined as a pattern of basic beliefs, assumptions, and values held by the people concerned (Deal & Kennedy, 1982; Hall, 1973; Hofstede, 1980, 1997; Parsons, 1951). In an organizational context, culture is also taken to include the control and exchange mechanisms inherent in the organization (Jones, 1983; Wilkins & Ouchi, 1983), that is, the “way things are done around here.” Following Hofstede (1980, 1997), it is generally Copyright © 2002, Idea Group Publishing.

The Impact of Culture on the Development of Information Systems: A Case Study 75

accepted that Western cultures are more individualistic, tolerant of different behavior and opinions, and have a lower regard for status differences than Eastern cultures. Eastern cultures, on the other hand, tend to be more collectivist, have higher uncertainty avoidance, and have an acute regard for status levels. These cultural differences are now being seen to affect the way people participate in the design, development, and use of information systems. At one level, software developers have been described as having their own intrinsic culture, where community is an important aspect of a successful team and debate is an integral part of the development process (Dube & Robey, 1999; Mackey, 1999; Sharp et al., 2000). Since information systems are developed with expectations of how end-users are meant to interact and work with the system, there are also potential conflicts when the values embodied within the system conflict with the actual culture of the end-users (Nissenbaum, 2001). The furore caused by Intel’s Pentium III chip with its embedded personal serial number is one example. At the national level, cultural differences in perceptions of information technology by end-users have led to a need to be culturally sensitive in the design of user interfaces (Evers & Day, 1997; Gando & Nielsen, 1996). The page layout, navigation, and information content of Websites have also been found to have a strong cultural dimension (Becker & Mottay, 2001). Culture has also been related to the relative effectiveness of Group Support Systems (Watson et al., 1994), and in the adoption of new technologies (Png et al., 2001). Hong Kong is perhaps a unique place to investigate these cultural issues, since it had been Western-governed for more than 150 years but has remained an essentially Chinese culture. Although Chinese management is more group oriented (Lockett, 1988), where collective ideals are emphasized, Hong Kong comes relatively low in terms of uncertainty avoidance in Hofstede’s model (Hofstede, 1980, 1997; Leung & Bond, 1989), suggesting they are willing to take risks. This suggests that Hong Kong, while having a basically Chinese group-oriented culture, has also adopted an entrepreneurial spirit, where business is as dynamic and high risk as in any other Western country. The question, therefore, is whether the culture defined for Hong Kong by other studies also includes the willingness to carry out an open debate, as required by most system development methodologies. Furthermore, would there be any other features of Hong Kong culture that might create problems when attempting to develop an information system. A case study will be presented based on a Master’s thesis (Lau, 1996) supervised by one of the authors. The project attempted to apply Soft Systems Methodology (SSM) in order to assess the feasibility of developing an Executive Information System (EIS) for a marketing unit in Hongkong Telecom. SSM is a general problem structuring method devised by Checkland (1981) that requires the building of rich pictures, root definitions, and conceptual models of the human activity under study. With the importance placed on identifying stakeholder viewpoints, Soft Systems Methodology (SSM) is a general problem structuring method that is capable of recognizing the cultural values that may be central to an organization. These cultural values would include the beliefs, assumptions, and values shared by members of the organization. The stakeholders are meant to build the root definitions and conceptual models themselves in an iterative debate organized by a facilitator. Therefore, it would seem that SSM would be an ideal tool to use for IS development since it can take account of the complexity of any given situation. However, SSM also indicates that negotiation and debate are essential ingredients when discussing the feasibility of developing an information system and for the capture of information system requirements. The need for debate, and the expected conflict between

76 Moores & Gregory

different stakeholder views, relies on the fact that the participants are willing to enter into such an open discussion, putting forward their personal views of whether the system is needed, and if so, what the system should do. But this, in itself, rests on the Western cultural assumption that debate is constructive and the most effective means of eliciting the information required. The problems of applying any such method in Hong Kong quickly became apparent, however. In particular, three main problems were observed. Firstly, these interviews were typically conducted in Cantonese but written in English. A feature of Hong Kong is that three languages are in common use: Cantonese, Putongua, and English. As such, interviews could be conducted in different cultural modes, depending on the language used. Secondly, no group debate took place between the stakeholders. The student, being Chinese, avoided the confrontation inherent in this approach and conducted interviews on a one-to-one basis. Thirdly, another feature of Hong Kong is the relatively high rate of staff movement among executives and professionals. Before the study had been completed all of the original stakeholders had been transferred. This clearly has a drastic effect on the ability of mature, experienced stakeholder views to be developed or captured. The next section will outline the use of SSM in the development of information systems and the Logico-Linguistic Modeling (LLM) enhancement used to determine the information sources required. Section 3 will then describe the case study, while Section 4 will identify the key issues that prevented a “normal” implementation of the debate and exploration process expected.

SSM AND ITS LLM EXTENSION Soft Systems Methodology (SSM) is often used as an information system design method. Developing the SSM models requires an analyst to determine the principal components of the activity or system under study. These principal components are guided by CATWOE analysis. CATWOE involves the identification of • the Customer, who is the victim or beneficiary of the system; • the Actors who carry out the transformation; • a Transformation that underpins the system; • the Worldview (Weltanschauung) that makes the root definition meaningful; • the Owner, who can stop the system; and • the Environmental constraints that subsume the activity. For instance, analyzing a university may identify the transformation as the process of transforming ungraded students to graded students. In this case, the customers might be the future employers, the actors might be the lecturers; the worldview may be that grades are necessary to identify the best employable people; the owner might be the fee-paying students or the funding government; and the environment might be the job market. Wilson (1990) extended the method by showing how it could be used to support the detailed restructuring of an organization (organizational mapping), and how the use of Information Categories and the Maltese Cross could support the conversion of a SSM model into a detailed information system design. Information categories show the required information inputs and outputs from the activities in an expanded conceptual model, while the Maltese Cross is a matrix of these categories that shows where the new information processing procedures are required. The systems devised are meant to be culturally feasible and systematically feasible.

The Impact of Culture on the Development of Information Systems: A Case Study 77

Shortcomings of the method include the fact that Wilson’s method is restricted to the design of transaction processing systems and provides no facility for building other types of information system, such as decision support or knowledge-based system. The method also assumes that the people in the organization know what information is needed to support their activities. This is particularly striking, since the method often calls for new activities to be included in the model. These shortcomings are also found in other methods linking SSM with information systems design. Logico-Linguistic Modeling (LLM)(Gregory, 1993, 1995) attempts to add a strong logical foundation to the SSM method by making a distinction between information that is empirically true and information that is true by definition. When this is done, the logically enhanced LLM-SSM models can be developed into a knowledge-based system that has learning capability.

THE CASE STUDY The study described here centers on the application of LLM-SSM to derive the requirements for an Executive Information System (EIS). The analyst involved in the project, Lau Siu Pong, was undertaking a part-time Master’s degree in the Information Systems Department at the City University of Hong Kong. He was a market analyst within Hongkong Telecom, which had been the sole provider of telecommunications services in Hong Kong, until the franchise was opened up with the granting of two other licenses in July, 1995. The market for communication technology and services is particularly lucrative in Hong Kong, with mobile phones and pagers being carried by most sectors of the community, from university students to sales and company executives. In the face of increased competition, Lau was interested in developing an EIS to support the activities of the Business Market Business Unit (BMBU). The BMBU is charged with planning the marketing of new business products. It was not envisaged that the project would immediately result in a fully computerized EIS. Rather, it was intended to undertake an analysis to determine if an EIS would be suitable, and if so, to broadly define its configuration. The project proceeded in a manner very close to the conventional seven-stage model of SSM (Checkland & Scholes, 1990). Stages 1 and 2 (situation entry and problem expression) normally require an outside consultant to act as an analyst/facilitator. This is meant to bring an independent, and possibly fresh, perspective to the problem situation. In this case, however, Lau acted as both the analyst and as one of the stakeholders of the system. While this ensured access to the people concerned, Lau’s own views about what should be done had the potential to cloud his judgement about the nature of the system required. For this reason, the role of facilitator was taken by one of the authors, Frank Gregory. The debate between the analyst (Lau) and the facilitator (Gregory) often required the analyst’s assumptions and his evaluation of the system to be constantly challenged and criticized. Stage 3 involves the formulation of root definitions of relevant systems of purposeful activity. Stage 4 involves building conceptual models of the systems named in the root definitions. The situation was slightly unusual in that the analyst was himself one of the “actors” defined in the CATWOE analysis. The root definitions and conceptual models produced by the analyst were discussed with Gregory, colleagues, and managers within Hongkong Telecom.

78 Moores & Gregory

Stage 5 involves comparing the model to the real world in order to determine the extent to which actions suggested by the model actually occur. Based on differences between the model and the real world, Stages 6 and 7 involve identifying changes that are needed that are also acceptable to the stakeholders. In this case, a number of actions were proposed, including setting up a news clippings database, establishing feedback channels from sales staff, and increasing the flow of marketing information between departments. It was during Stages 3 and 4, where discussion and debate were required between stakeholders, that the cultural features inherent in applying the method in Hong Kong became apparent.

THE CULTURAL ISSUES While an IS project can often encounter problems simply because of the technical nature of the development project, there were three issues that were clearly rooted in the way things are done in Hong Kong.

Avoidance of Group Debates In a context that involves Western and Chinese elements, a “debate” can involve a head-on clash of values. For a Westerner, a debate can be seen as constructive and beneficial. When people have adversarial positions, a debate may help the truth to emerge and so be considered as beneficial to all parties, even for those that “lost” the debate. For Chinese, however, the display of adversarial positions depends very much on the social standing of the two adversaries. An argument between peers is acceptable, while a junior arguing with a senior is seen as unacceptable. In SSM, the building of root definitions and conceptual models proceeds by means of an iterative debate involving the stakeholders and a facilitator. It is assumed that much of this work will take place in a series of group meetings in which the stakeholders will express their individual viewpoints. In these circumstances, it can be expected that the viewpoints of two or more stakeholders will be antagonistic. The role of the facilitator, therefore, is to mediate and produce a model that accommodates as many viewpoints as possible. In this case, however, it was clear to both analyst and facilitator that the confrontational debate approach needed to be avoided if colleagues and managers were to be persuaded to participate in the project. The result was that the analyst avoided the group debate approach and conducted one-on-one interviews in order to build the SSM models. While this diverged from the principle of stakeholder debate required by SSM, the one-on-one interviews were seen as a more culturally-sensitive means of eliciting opposing views. The “debate,” in other words, was carried out on paper where status levels could be more easily respected. The implication here for other IS development methods is that any technique used to elicit user requirements that involves debate between non-peers would be doomed to failure in an Eastern culture. This would suggest that techniques such as Joint Application Design (JAD) and the Nominal Group Technique (NGT) could not be successfully implemented in Hong Kong, since both approaches require debate. The JAD approach is specifically designed to use group meetings between users, stakeholders, and IS professionals, in order to debate the form and function of a proposed information system. It is exactly this phase that was avoided during the SSM project. The NGT approach requires participants in a group meeting to write down ideas and suggestions anonymously, before a facilitator summarizes and presents the ideas for debate. The purpose

The Impact of Culture on the Development of Information Systems: A Case Study 79

is to prevent the meeting from being hijacked by one or more dominant individuals, and to allow all views and opinions to be heard. However, it is entirely possible that sub-ordinates would resist criticizing an idea in the subsequent debate just in case the idea originated from one of their superiors. In short, while it is not possible to say whether the paper-based debate strategy used was entirely successful, it is clear that the uncertainty avoidance and power distance dimensions of Chinese culture make formal debates an unacceptable approach to eliciting system requirements.

The Use of Multiple Languages The analyst reported that the SSM models, written in English, would be discussed with his colleagues in Cantonese, although the discussion itself would be written down in English. It is not uncommon for two Cantonese speakers to intersperse English words into their conversation if the speaker is familiar with the fact that the English word requires less effort than the Cantonese alternative. The ‘effort’ here is the number of syllables required to pronounce the word or phrase. This sprinkling of English is only made, however, when the speaker believes the listener is sufficiently familiar with English words that the basic meaning will still be communicated. Because of the recent history of Hong Kong, such linguistic acrobatics are quite common. In particular, while the local Chinese dialect is Cantonese, university teaching is conducted in English, although the recent transition in sovereignty has meant Putongua (i.e., Mandarin) is now being taught in Cantonese secondary schools. If philosophers of language and thought are correct, however, the language one uses will determine the breadth and depth of ideas that can be communicated. For instance, Whorf (1956) suggests that what one can think of is constrained by the language that one has to express these thoughts. The often quoted myth attributed to Whorf is the suggestion that North American Eskimos have a richer picture of frozen or freezing water because they have over 70 words for concepts such as snow, ice, slush, etc. The intimate connection between language and thought is also central to the philosophy of Wittgenstein (1953). On this basis, Cantonese has a richer linguistic set than Putongua. Cantonese has nine tonal levels, while Putongua has four. A simple mistake in tone can change the Cantonese phrase “cigarette lighter” to the more sinister “kill the waiter!” In common with Putonguaspeakers in Taiwan, Cantonese uses the older Chinese script, while the Mainland uses a more modern, simplified character set. In terms of tonal levels and complexity of script, therefore, Cantonese is perhaps unrivalled in its Whorfian ability to express thoughts, with subtle differences in tone or pen stroke communicating radically different ideas. Because of the added linguistic complexity in Hong Kong, however, it is not even clear what language is being used to express the beliefs, assumptions, and values of the stakeholders. Where there is a choice, the language adopted by the discussants (e.g., Cantonese or English) is likely to be a strong indication to both parties which underlying culture is being assumed. When the analyst (Lau) discussed the SSM diagrams with Cantonese colleagues the Chinese culture could be expected to dominate. Since the analyst and stakeholder share a common language, the view of the stakeholder is likely to have been communicated effectively. Given that something is always lost in translation, however, the fact that the models were then developed in English suggests the stakeholders’ views must then be transformed into an English concept before it could be represented in the SSM diagrams. Even if the analyst understood the viewpoint of the stakeholder, it is unlikely there would always be an English word to match the idea expressed in Cantonese.

80 Moores & Gregory

To compound this problem, when the analyst discussed the same diagrams with the English facilitator, one would have to assume the shared culture is Western, since the facilitator (Gregory) does not speak Cantonese. One of the likely consequences of this shift is that the facilitator would bring to bear a more precise understanding of the English words used. In this case, there is the possibility that some of the ideas inadequately expressed in English by the analyst were misinterpreted by the facilitator in subsequent discussions. In reviewing the models, some of the words may have been changed to express an idea closer to the analysts’ English description of what the stakeholder said, although the subsequent rewording shifted the SSM model further from the original viewpoint. In turn, when discussing the revised models with the stakeholders, the analyst would then be faced with translating the new words into Cantonese. Given this continual shift from Cantonese to English (and back again), it is unlikely that the models developed could ever fully represent the (Cantonese) views of the stakeholders. While most IS projects struggle to derive models that faithfully reflect the views of those concerned, it would seem to be an extra twist of the knife to require a further transformation from one language to another. Formal methods, such as Z, that use mathematical notation rather than free text, would not be a solution here, since there is still a transformation from free text to the Z notation. Furthermore, it would not be acceptable to suggest that an information system should only be developed in one linguistic context, since the development of the World Wide Web and electronic commerce has opened up the boundaries of business. As information systems stretch across the globe, it would seem sensible to at the very least include aspects of Hofstede’s work in the curriculum of IS management courses. Information system projects of the next millennium would seem to demand more training in issues of culture shock, ethnocentrism, stereotyping, and the differences in language and humor. This suggests that IS development projects that tap more than culture should include a cultural feasibility study to determine the extent to which any cross-cultural issues could jeopardize the project.

Rapid Turnover of Stakeholders While the problems outlined above can be overcome to some extent, there was one problem that neither the analyst nor the facilitator could do much about. By the time the feasibility study had been completed, Hongkong Telecom had transferred all of the stakeholders involved in the model building project. Indeed, Hong Kong’s high growth and relatively low unemployment has helped to produce an environment in which staff movements between companies is relatively high. Recent government figures indicate that staff movements among executives and professionals is nearly 5% per year (CSD, 1997), and it is generally accepted that junior staff will change jobs every 18-24 months. In this case, the Head of the BMBU was transferred a few weeks before the project was finished, and within three months the analyst and all other members of the unit had been transferred. While the project was successful in determining the scope of the EIS required, it has to be admitted that the SSM project itself was a failure. The key issue here is that SSM is intended to work because the stakeholders produce and “own” the solution. With their viewpoints inherent in the models, the stakeholders are meant to recognize the solution as being of benefit to them, and thus to support the implementation of the solution. If the original stakeholders have moved on, then the new staff will not be the owners of the solution, and thus, nobody can be expected to implement the solution. The SSM project would have to be started again.

The Impact of Culture on the Development of Information Systems: A Case Study 81

The time scales involved in staff movements suggests that, perhaps, no SSM project will be successful in Hong Kong. Although there are no published figures on the duration of SSM projects, six months elapsed time is an appropriate benchmark. SSM evolved through a learning cycle using Master’s students at the University of Lancaster. Most of the Masters students completed an action learning project lasting approximately six months. Much of the development of SSM was based on what was learned during these projects. Of the six SSM projects conducted at City University (all of six months duration), four were badly affected by the movement of major stakeholders within the organizations concerned. It would appear, therefore, that for SSM to be applied in Hong Kong, the project time would need to be significantly reduced to cope with staff movements, or significantly increased, to include a second iteration that includes the new stakeholders.

CONCLUSIONS The paper has presented the results of a case study that applied LLM-SSM in the design of an Executive Information System (EIS) for the newly demonopolized Hongkong Telecom. The EIS was to support the Business Market Unit, a unit that plans the marketing of new products. The project aimed at determining the feasibility of the development of an EIS, and if so, determine its configuration. The cultural problems inherent in the project were focused on using a debate approach to elicit views of the problem area. In particular, Eastern cultures are more strongly group-oriented rather than individualistic. The debate between users expected in the building of root definitions and conceptual models are not easily applied in such cultures. It was found that group discussions were avoided, with interviews being conducted in a more discreet one-on-one situation. Although the project itself was a success in terms of identifying the basic system requirements, further problems were identified; in particular, by the time the system requirements were determined, the unit that was to have used the system was effectively disbanded. This case study has highlighted some of the features and problems inherent in using SSM to develop information systems in Hong Kong. The uncertainty avoidance and power distance dimensions of Chinese culture make stakeholder debates an unacceptable approach to building the required conceptual models, while the multi-language and high staff movement that is a feature of Hong Kong presents further difficulties. While SSM and its LLM extension showed promise in allowing system requirements to be identified, the method still has a number of practical problems that require further analysis. The cultural problems of implementing a given IS development methodology is clearly an area that needs further research.

REFEFENCES Checkland, P. B. (1981). Systems Thinking, Systems Practice. New York: John Wiley. Checkland, P. B. and Scholes, J. (1990). Soft Systems Methodology in Action. New York: John Wiley. Deal, T. and Kennedy, A. (1982). Corporate Cultures. Reading, MA: Addison-Wesley. Dube, L. and Robey, D. (1999) Software stories: Three cultural perspectives on the organizational context of software development practices. Accounting, Management, and Information Technologies, 9(4), 223-259.

82 Moores & Gregory

Evers, V. and Day, D. (1997) The role of culture in interface acceptance. In Proceedings of the 6th IFIP International Conference on Human-Computer Interaction, 260-267. Gando, E. M. and Nielsen, J. (1996) International User Interfaces. New York: John Wiley. Gregory, F. H. (1993). Cause, effect, efficiency and soft system models. Journal of the Operational Research Society, 44(4), 333-344. Gregory, F. H. (1995). Soft system models for knowledge elicitation and representation, Journal of the Operational Research Society, 46(5), 562-578. Hall, E. T. (1973) The Silent Language. Garden City: Anchor Press/Doubleday. Hofstede, G. (1980). Culture’s Consequences: International Differences in Work-Related Values. Beverly Hills, CA: Sage. Hofstede, G. (1997). Cultures and Organizations: Software of the Mind. New York: McGraw-Hill. Jones, G. R. (1983). Transaction costs, property rights and organizational culture: An exchange perspective. Administrative Science Quarterly, 28, 454-467. Lau, S. P. (1996). Logical soft system models for information requirement analysis/ elicitation. MA Thesis, Department of Information Systems: City University of Hong Kong. Leung, K. and Bond, M. H. (1989). On the empirical identification of dimensions for crosscultural comparisons. Journal of Cross-Cultural Psychology, 20(2), 133-151. Lockett, M. (1988). Culture and the problems of Chinese management. Organization Studies, 9(4), 475-496. Mackey, K. (1999). Conscious conflict. IEEE Software, 16(5), 112-113. Nissenbaum, H. (2001). How computer systems embody values. IEEE Computer, 34(3), 118-120. Parsons, T. (1951). The Social System. New York: Free Press. Png, I. P. L., Tan, B. C. Y. and Wee, K. L. (2001). Dimensions of national culture and corporate adoption of IT infrastructure. IEEE Transactions on Engineering Management, 48(1), 36-45. Sharp, H., Robinson, H. and Woodman, M. (2000). Software engineering: Community and culture. IEEE Software, 17(1), 40-47. Watson, R. T., Ho, T. H. and Raman, K. S. (1994). Culture: A fourth dimension of group support systems. Communications of the ACM, 37(10), 44-55. Whorf, B. L. (1956). Language, Thought and Reality. London & New York. Wilkins, A. L. and Ouchi, W. G. (1983). Efficiency cultures: Exploring the relationship between culture and organizational performance. Administrative Science Quarterly, 28, 468-481. Wilson, B. (1990). Systems: Concepts, Methodologies and Applications. New York: John Wiley & Sons. Wittgenstein, L. (1953). Philosophical Investigations. Oxford: Blackwell.

Dysfunctional Development Pathways 83

Chapter VII

Dysfunctional Development Pathways of Information and Communication Technology: Cultural Conflicts G. Roland Kaye and Stephen Little Open University Business School, UK

This paper argues that there are three factors that counter the inevitable movement towards globalization. Firstly the incremental force of technology as illustrated by the growth stage model of the development of Information and Communication Technology (ICT) is flawed. This model implies a linearity of development and an inevitability of stage following stage. While this stage model may provide historic explanation for the development in the developed world and amongst the mature users, the model fails when used predictively for the developing nations or for the late adopters. Secondly the imperialism of technology overcoming all barriers fails to reconcile the cultural dimensions of both the developing context or the application domain. Technology is not culturally neutral but is developed in a cultural context and in the case of information rich applications carries that cultural context within its design. Applications of culturally developed systems, such as office and management systems, assume the user’s compliance with the design culture, but this inevitably leads to cultural clashes as we apply outside the design context. Thirdly, the assumption of universality of economic access and development is incompatible with both the reality and development paths in both developed and developing countries. This inevitably will lead to a divided society split between the internationally mobile, technology-supported communities and those communities disadvantaged economically and technologically but culturally rich. The failure to bridge this gap may leave society as a whole weakened through lack of access to “variety.” The paper discusses these perspectives and illustrates the case with evidence from NE Asia and the United Kingdom. In particular, it focuses on software development and information-rich contexts. Appeared in Journal of Global Information Management, vol. 8, no. 1, 2000. Reprinted by permission.

84 Kaye & Little

TECHNOLOGICAL INEVITABILITY–GROWTH STAGE MODEL Gibson and Nolan (1974) provide a growth stage model (GSM) of the development of Information and Communication Technology (ICT). The initial model had four stages: Initiation, Expansion, Formalization, and Maturity. Nolan (1979) extended this model to a six-stage model reflecting recognition that there were more sub-stages reflecting the growth of knowledge and technology from the implementations. The model reflected not only a shift in the objectives being sought (explanation of budgetary expenditure on ICT) but the involvement of management and the control and orientation of the evaluation. The revised model (Figure 1) demonstrates a balance between slack permitting growth and innovation and phases of control ensuring cost effectiveness and integration. The rate of growth inferred in the graph below is quite gentle, but for many the actual rate is explosive, causing problems of control. The stages reflected distinct categories of evolution from batch processing through time-share data processing to PCs and networks of communicating processors. With hindsight, that may have been the experience of the larger U.S. companies who had been involved with computers from the early days. This could not be described as the experience of the later entrants often small and medium enterprises (SMEs), whose first foray into the arena came with turnkey and proprietary software, or the more recent entrants with commodity based PCs and software. Freidman (1994) suggests that GSM is not purely a description but rather reflects a dynamic process embracing all of the stages and dictating their order of appearance. It is the “likely crises in the organization’s experience of computers” and management reactions to large scale and poorly understood technical events and organizational change. Freidman suggests that the underlying emotional characteristics are • Caution when dealing with unfamiliar subsidiary issues, • Optimism that follows success, • Pessimism that follows disappointment, and • Balance that follows experience of variations. GSM inevitably suggests a product life cycle S-curve, but perhaps the difficulty encountered with the model rests with the lack of separation of the life cycles of individual component technologies from the long-run product life cycle. If we select any phase within the model, we find that this incorporates several technologies; some of which are in maturity, others that are only emergent, and some which are expanding. This overlapping of life cycles confuses the long-term trend, as some technologies superseded stepwise changes, and other technologies were additional (Figure 2). In the early phases, certainly many hardware Figure 1: Nolan’s six-stage model

$

Time Initiation

Contagion Control

Integration Data Administration

Maturity

Dysfunctional Development Pathways 85

Figure 2: Product life cycle S-curves

Volume

Time

Figure 3 Freidman’s four-phase model

Phase 1

Phase 2

Phase 3

Phase 4

developments led to straightforward replacement, but later complex networks developed, involving several generations of hardware and likewise software. One area where the proliferation of technologies may be found inter-working is the office in which office automation has not replaced but rather led to increasing varieties of technologies used to perform overlapping tasks. Only recently have we seen attempts at reducing the redundancy and integration of the technologies of printers, faxes, copiers, and scanners. This represents the micro-level of the problem represented by the “information archipelago” of MacFarlan and McKenney (1982) in their extension of Nolan’s model to office automation in general. The GSM model purports to provide identifiable stages and identifiable categories. However, the empirical evidence to this position is lacking, and the underlying time dimension, as expressed in age of IS function, does not correlate. Despite the evidence from Drury (1979, 1980, 1983) and Benbaset et al. (1984), the model continues to be used either explicitly or similarly: Galliers and Sunderland (1991), Jayasuriya (1993), and Wastell and Sewards (1995) to project and suggest development pathways for ICT. Freidman suggests the continued usage reflects 1) The only explicit model of time pattern of IS function development, 2) Clear and testable hypothesis, 3) Prescriptive content, and 4) The model does summarize some experiences of organizations.

ALTERNATE MODELS Earl (1996) accepts the prescriptive nature of the GSM model and argues for its use to judge the balance between user focus and specialist focus as a measure of technological maturity. Mature use should by now be common, yet the “productivity paradox” arising from

86 Kaye & Little

the lack of correlation between performance measures and IT expenditure suggests that the mature phase itself is either elusive or extremely varied in outcome. Brynjolfsson and Hitt (1993) and Brynjolfsson et al. (1994) offer evidence that productivity gains are achievable ultimately, reflecting a tendency towards smaller IT-driven organizations. Freidman with Cornford (1989) divided the history of Information Systems (IS) into a series of phases reflecting the domination of a particular problem that was seen as a constraint on computerization. They separated the phases into hardware capacity constraints– until mid 1960s; software productivity constraints–mid 1960s to early 1980s; user relations–early 1980s to 1990s; and, organization environment constraints–1990s and on. Freidman (1994) suggests these phases may be seen as overlapping but with increasing stretch, resulting in the shift in the location of IS control. He argued that at the boundary between the phases there was mediation between the respective environments. This shift in control demonstrates how IS reaches out to seek to control and influence its boundaries, as each problem or constraint is overcome. All these models assume a linearity of development that is incompatible with the opportunities faced by later entrants into the technology. Not only does a late entrant have access to the development path of others, but also to their experiences. The lessons from experience are available to both the new entrant and old. Thus the new entrant may gain access to the current technologies and applications without recourse to working through the development path. Okot-Uma (1988) presents a matrix to represent the varying levels of capability in information technology that exists between developing countries within the British Commonwealth. Such a model demonstrates the need for suppliers to address the capabilities of users. Later entrants are able to “leap-frog” the early development stages and gain access to the more mature technology, but they still must assimilate the learning from the early stages. This learning may be facilitated with experience of neither the associated organizational structure and task changes nor the primitive technology. The possibility of “leap-frogging” (i.e., missing out the early stages and going straight to the latest technologies) suggests that access to technology as it develops should increase. However, evidence from Northeast Asia suggests that late entrants to the technology have not assimilated; instead, dysfunctional divisions are appearing between the “have’s and have nots.” Okot-Uma also introduces the concept of “back-frogging” where technologies are prematurely deployed and either negative experience occurs or there are problems in assimilation. In those circumstances, not only can the technology fail but barriers to other new technologies may be created. A choice of “back-frogging,” (i.e., backward steps in development) related to concepts of “appropriate technology” is presented as an equally necessary strategy to leapfrogging the competition. Perrin (1982) shows in an intriguing study that the Japanese were familiar with gunpowder and guns and had in fact developed them in the 16th century. However, when the American traders forced them to open their harbors in the 19th century, they had none. The Samurai had banned firearms as “un-chivalrous,” since they prevented the personal confrontation implied in warriors engaged in battle. This example further reinforces the view that linear technology development is not a forgone conclusion. The concept of inevitable technological development along a linear growth path may explain some developments, but is not universal. Opportunities to enter later stages or to retrace earlier stages are possible.

DIFFUSION AND LOCALIZATION Institutional and technical aspects of diffusion of innovations in technology have been subsumed in the S-Curve representation discussed by Rogers (1983). Strang and Meyer

Dysfunctional Development Pathways 87

(1994) distinguish between diffusion, “flows among formally autonomous units” and implementation, “flows among hierarchically placed units.” Organizations seeking to diffuse centrally developed systems must either accommodate intra-organizational cultural differences or demand that the end user groups adapt to the technology. This raises interesting questions for the movement of technology developed in one cultural context and transferred to another cultural context where the hierarchical controls may not exist to force the acceptance of the new technology. Software systems may be carriers of implicit values and styles that may be alien or even subversive to other cultures. “Localization” may be a response to this perception, but this fails to recognize that the software assumes a cultural context at the design stage, and this deep-seated cultural value cannot be overcome by cosmetic redesign of interfaces. Keniston (1997) discusses the software localization process whereby software is localized through a process of initially retro-fitting appropriate user interfaces to more current practice of joint modular development of software with appropriate user interfaces. For Keniston the translation of software to allow for different character sets and right, left, or vertical scrolling reflects the industries’ valid response, given the importance of software to the U.S. economy. However, cultural dimensions reach far beyond the character sets of the user interface. It is also important to separate software into the socially influencing and the technically dominant. The former are those systems and applications for which the assumptions of organization structure and process, as well as individual user interfaces, are culturally sensitive (e.g., accounting and information systems, groupware, etc.). These contrast with the technical systems, such as CAD-CAM, where the cultural aspects of the interface are cosmetic, and the underlying structure and processes are technically dependent and culturally neutral. Further, communications differ reflecting the cultural attributes of the respective languages. Hall (1976) provides a clear split between High Context Cultures Much of the information being transmitted is in the physical and social context of the conversation; relatively little information is in the explicit message. Communication is indirect; there is room for ambiguity and interpretation. People expect others to know what they mean—associated with collectivist cultures. Low Context Cultures Most of the information is conveyed explicitly, directness is valued and there is little ambiguity—clear and to the point information—associated with individualistic cultures. Kaye and Little (1996) illustrate the problems of diffusion through two case studies and demonstrate how diffusion, from the developer and adoption in context, leads to adaptation and redevelopment, leading to version drift and eventual incompatibility. Figure 4 presents two dynamics of diffusion. An initial adopter, Organization A, undertakes a technical development on the basis of the needs of the organization at time Tl. This technology is deployed in the organization at time T2, at which point the organization itself has developed institutionally through interaction with its environment. In order for the technology, based on a snapshot of the organization at T1, to be usefully employed, the organization must enter into a period of adjustment between the delivered system and the changed needs. Organization A reaches the beginning of its payback period on the innovation at T3. Organization B, having observed A’s process of deployment, initiates its own development

88 Kaye & Little

of the new technology at T3. In order to adapt the technology to its own needs it pursues a process of redevelopment followed by its own process of adjustment, so that payback for Organization B begins at T5. Organization C may follow the same development path as B but in this case it would trail technologically. Alternatively, it may seek to mimic and develop at the same time as B. If this latter strategy is adopted, then it will not be able to draw on B’s experiences and must instead attempt to reverse engineer, which may achieve the earlier payback but entails a higher risk. This example demonstrates that linearity of diffusion may not always occur and localization can be problematic. Localization has problems of cultural barriers of which the most obvious is language that restricts access. Making the software culturally accessible places an additional translation cost in the system that may further raise barriers to acquisition of the system. These problems of localization place a limit on the globalization and universality of ICTs.

PROBLEMS OF UNIVERSALITY Sachs et al. (1997) suggests there are limits to convergence. Transportation costs associated with distance and physical barriers create economic limits and inequalities of income. Physical geography, government policy and demographic changes have been shown to influence economic growth in the period 1965-90. While specific policies, such as openness of markets combined with regulation, may overcome some deficiencies, the longterm aspects of physical geography may limit growth. Under free trade, each country should specialize in those products in which it has relative advantage; however, Matsuyama (1992) suggests that those endowed with good arable lands and natural resources might encourage agricultural growth at the expense of industrialization. Evidence from the growth of Latin American countries suggests that economic growth can be achieved from both agriculture and industry by ensuring that not only does productivity rise in the factory, but also on the farm. Such increases in productivity can be achieved by increasing the value added by, for example, moving from production of fruit to production of wine. However, “Engels’ law” suggests that as incomes increase the proportion spent on food will decrease, setting a limit on agricultural growth. This limit has a consequence for the distribution of incomes and may ultimately limit industrial growth if redistribution is not facilitated to avoid economic migration. Rodrik (1997) suggests that those favoring globalization underestimate the effect of free trade and capital flows, which allow organizations to relocate production to low-wage economies. The very threat of this causes insecurity and weakens established capital-labor relationships. Further, the quality of working conditions are more than a protectionist cry but reflect genuine concern for minimum standards of decency and reflect the lack of social insurance and the increasing burden on the state. Rapid growth in favored areas creates regional imbalances, while localized growth may cause instabilities. Ohmae (1990) argues for regional synergies to maximize development regardless of pre-existing national boundaries. However, the sustainability of such development is dependent on a balanced economy, and any unevenness may result in internal, international, or regional migration on such a massive scale as to create instability. In the UK imbalances may be found between the forecast GDP for 1997 of 4% for the whole UK with regional variations of 1% in Wales and 5% in Greater London (Economist, 1997a).

Dysfunctional Development Pathways 89

Figure 4: Diffusion under the S-Curve from Kaye, R.. & Little, S. E. (1996). Global business and cross-cultural information systems: Technical and institutional issues of diffusion. Information Technology & People, 9(3), 30-54

Reverse engineering Adjustment Org C

Org C

Payback Org C

Instututional Dev Institutional (mimetic) Diffusion Redevelopment

Org B

Org B

Org B

Adjustment

Payback Org B

Instututional Dev

Technical Diffusion

Redevelopment

Technical Dev Adjustment Org A

Org A

Payback Org A

Instututional Dev T1

Org A

Org A

T4

T5

Re-adjustment

Instututional Dev T2

T3

EMPIRICAL EVIDENCE FROM NE ASIA AND THE UK Japanese factories have achieved worldwide acclaim for their productivity and quality, their automation and technology, and for innovation and market awareness. The high technology factories are in stark contrast to the office and administration systems. The factories use the latest technology of CAD/CAM, FMS, etc. to achieve worldclass manufacturing of the technologies that the West then utilizes in their offices and administrative systems. Meanwhile, the eastern offices and administrative systems are reminiscent of an earlier generation of Western practices with large generally open plan offices, long rows of uniform desks and a proliferation of calculators, telephones and faxes but few workstations or personal computers. Computers are shared and as such cannot easily be used for networking with e-mail and groupware products, which now form the backbone of many Western systems. Graven (1994) describes attempts by Shiseido, Japan’s oldest and largest cosmetic company, to achieve office automation. In 1957 it introduced its first IBM computer to track orders, plan production, and handle logistics. In the 1970’s it computerised its distribution system. However, it still faces the challenge of office automation as it moves from this centralized information and data processing systems to distributed end-user computing where the individual user must interact through the technology. Facsimile transfer, developed in Europe in 1890’s, led to the modern-day electronic fax, which was jointly developed by Xerox and Magnavox in 1965. The adoption of this technology was most rapid in Japan (by 1985 an installed base of 850,000). Where the telegraphic systems of communication had required either the adoption of English language or the development of a separate Japanese code to represent Kanji characters, the fax allowed

90 Kaye & Little

the direct transmission of traditional Japanese characters thus overcoming the language barriers within the technology. The acceptance of fax technology in the U.S. and Europe was much slower, and the comparable figures for 1985 are 550,000 and 120,000, respectively. While many fax machines sold carried Western brand names, the bulk were manufactured in Japan. During the same period, 1981 to 1984, the sales of Word processing packages in the U.S. increased more than tenfold to approximately 2 million; and by 1987, two of the more popular versions were selling in excess of 600,000 copies per annum. In a similar period the sales of faxes rose from 50,000 in 1983 to 295,000 by 1987. Word processing was not only compatible with Western culture, but also with Western business practices. The ability to send text through teleprinters had been available from 1897; and, while sending digital messages across networks was common practice, there was some delay before wide-scale usage of electronic mail emerged. In the meantime, the fax could be used. This apparently retrograde step was a surprise to many technology forecasters, who had expected the next step to be e-mail. However, while the technology was available, there were few wide-area networks with the necessary capability to allow e-mail. The large scale development of the Internet provided that facility internationally but was in use predominantly within a closed technical community for a considerable time. A recent report (Economist, 1997b) suggests that corporate Japan is now commencing the wiring of offices, which the U.S. undertook a decade earlier. Sales of PCs have risen in the last five years, from 2.2m units in 1992 to 8.3m in 1996, of which 5.6m were corporate purchases. Suddenly Japanese office workers have access to computing with 20% of desks sporting a PC. Most of these PCs have been used for Japanese word processing, reflecting the long-awaited access to kanji and kana character sets. In recent months, sales of fileservers to support network systems and UNIX machines have increased by 89% in Japan (IDC Japan reported in Economist, 1997b), which implies about one third of all PCs are networked. However, the bulk of these appear to be for e-mail and some access to the Internet (replacing the fax and copier). The dearth of groupware products, such as Lotus Notes or other application-sharing and workflow systems available in Japanese, are not the only delay. The hierarchical corporate culture of Japan discourages the flatter organizational style assumed in these technologies. A further limit is the lack of information managers and technologists whose specialist development is not possible, given the job rotation approach of Japanese management. Nonaka, Ray and Umemoto (1998) analyze the contrast between knowledge creation and management in Japanese and Anglo-American environments, setting these in the context of Japan’s “three sacred treasures” of lifetime employment, age-based promotion, and company unions. These support a milieu in which the explicit mechanisms promoted in the West are unnecessary. Davenport’s observations of the rise and fall of business process re-engineering as a fad indicate a growing Western appreciation of the role of implicit knowledge and undocumented procedures in delivering performance (Davenport & Short, 1990; Davenport, 1993; Davenport, 1996). Whether Japanese and Anglo-American office environments are converging, or perhaps exchanging, their traditional characteristics remains to be seen. Until recently, the most common language on the Web after English was German. However, a recent survey by the Internet Society (1996) of a global sample of the 30m computers with Internet addresses found that 82% were English; 4% German; 1.6% Japanese; 1.5% French; 1% Spanish. Given the population distribution of the world and the languages and character sets, the data indicates a very uneven distribution between the alphabet-based cultures and those

Dysfunctional Development Pathways 91

using ideograms. However, the rapid growth of networked PCs in Japan is rapidly moving Japan into second place. This reflects a move by the Japanese Ministry of Posts and Telecommunications to counter the American domination of this new technology. Switching between Japanese and English is problematic, as English requires only one byte to define the character, while Japanese needs two bytes (Chinese uses a subset, which can be accommodated within the two-byte character set). While the solution has been to run two machines with different operating systems, new product launches of Twinbridge, KanjiKit, and Japanese Language Kit for Macintoshes enable some coworking (see endnote). The Motorola Report (1996) draws on a survey undertaken on its behalf by MORI of households and leading industrialists in the UK. The report characterizes the nation as “IT haves and have-nots.” The division exists on the basis of sex, age, employment, and social class. Not only were the younger (16-44) administrative and managerial groups more likely to have access to computers in their workplace but these same social classes were likely to provide access to computers in their homes, facilitating educational access for the next generation. In contrast the lower social classes (C2DE) were more likely to be employed in roles where ITC had less impact on their daily activity, and as they were part-time working or unemployment further placed barriers in their development of IT skills, “caught in an IT void.” This survey concludes that barriers continue to the adoption of ITC across the whole of society. An IMF working paper by Slaughter and Swagel (1997) indicates a widening gap between the high skilled (graduate and above) and the low skilled (secondary educated and below) in the USA and Britain, but less marked in other Western economies. This gap has been explained in various ways. First, “low-wage economies” drives down the unskilled rates. Secondly, technological development had boosted the productivity and wages of skilled workers. At the same time, these economies have experienced recession with high levels of unemployment and increased part-time work, leaving workers less able to negotiate a fairer allocation of funds between capital and labor providers. While the explanation for the gap may be disputed, all agree that technology can only increase the gap.

GLOBALIZATION AND CITY-STATES The forces of globalization and universality lead to two divergent models. In the first case, globalization encourages the city-state to emerge. This produces stronger cultural links to the international standards of exchange and trade with the reference group being similar trading groups elsewhere. Hence, the universal language of business is English and the software classically American. In contrast, the national identity is at odds with international trading, as evidenced by the survival of distinctive national cinemas and their stalwart defence by national governments in the face of the World Trade Organization. This disharmony reflects the cultural proximity (or distance) of the emergent global standards. The indigenous culture has strong roots and has evolved to complement the structure and processes of society. In contrast, the exogenous culture may be distant both at the surface or deeper. In the case of Western organizations, delayering and technological supports have enabled flatter organizations to emerge with short chains of command. In contrast, China and Japan have long chains of command and strong hierarchies which permeate deep into society and behavior. This places these cultures as distant, which when applied to the context of, say, China, leads to significant gaps between Shanghai’s development and the average province of China–thus, tension between the rural supportive infrastructure and the urban

92 Kaye & Little

internationally referent city-state. This divergence was not present in the first industrial revolution as developments took place subsequent to an agricultural revolution within a consistent cultural tradition and context. Today, we have two divergent reference groups. Robertson (1992) uses the concept of relativization to represent the increasing challenge to particular perspectives on, and individual participation in, the overall globalization process. His model draws attention to the increasingly interrelated thematization of societies, individual selves, international relations, and humankind. He goes on to identify a temporalhistorical path to the current circumstances: Germinal phase nationalism; Incipient phase homogeneous unitary state, formalized international relations; The take-off phase - emergence of a single international society, increasing global communication; Hegemony phase - international wars and emergence of United Nations and recognition of national interdependence; Uncertainty phase - concept of global consciousness further amplified by space travel, technological acceleration, and multicultural societies. Hampden-Turner and Trompenaars (1993) surveyed 1500 “upper-middle” managers who had some international responsibilities. They found that while managers from Britain, Holland, Sweden, and the U.S. concentrated on individual self-interest, which, when congruent with the organizational goals, meant they would automatically serve their customers and society better. In contrast, Japanese, French and German managers concentrated on serving customers and society, relying on the organization to reward them for pursuing the organizational goals. The relationships of the individual self interest, the organizations and customers or societies they saw as interdependent, but fundamentally the value systems of the cultures were significantly different. However, this oversimplification fails to portray how the cultures vary subtly and therefore cannot be easily classified. The Hampden-Turner and Trompenaars’ study led them to identify seven cultural values which underpin the alternative wealth creation systems (p. 10). They characterize these values in the form of dilemmas where choices have to be made between the extremes: Universalism vs. Particularism Analyzing vs. Integrating Individualism vs. Communitarianism Inner-directed vs. Outer-Directed Time As Sequence vs. Time as Synchronization Achieved Status vs. Ascribed Status Equality vs. Hierarchy To quote, “To say that the cultures of various nations differ on the relative importance of those values necessary to wealth creation is an understatement. Typically, these issues are loaded with ideological fervor” (p. 11). For all its limitations, Hofstede’s pioneering study of the employees of a single global corporation confirms the durability of national cultural differences in the face of the hegemonic socialization employed by the company (Hofstede, 1980).

Dysfunctional Development Pathways 93

CONCLUSION This paper has argued that the assumptions of universality and common linear pathways of development are flawed. It has argued that the opportunity to enter the development are multiple and lessons from prior users can be drawn up on. Further development paths may be nonlinear as accelerated learning and technology innovation allows “leap-frogging,” or deferment. Given the global development of organizations and the dependence such developments have on ICT, it is important that we have both descriptive and predictive models of the development of ICT. Nolan’s GSM has been a useful historical description of developments in large Western organizations. It has failed to be proven, and other models have provided explanations of the complex development paths available. Other models have contributed further explanations: Freidman’s phase model recognizes the constraints that limit growth and development, while Okot-Uma provides the possibility of both forward and backward stepwise developments reflecting both the opportunities faced by late adopters and the potential for both positive and negative experiences with innovation. The diffusion model also provides an explanation for adoption at varying rates and potential of discontinuous development of versions. These descriptive models collectively suggest a complex Web of development which is unlikely to be linear or functional. Globalization encourages the belief or aspiration in software developers to universality; and while certainly some evidence of “Macdonaldization” is perceivable at the superficial level, the underlying characteristic is of localization. However, this localization is frequently no more than the user interface and ignores the cultural assumptions embodied in the technical design. Kirlidog (1996) reveals the extent to which the assumptions embedded in a management information system were in conflict with the established working patterns of the overseas subsidiaries of the transnational corporation seeking to deploy it. Not only does superficial “localization” and underlying universality cause dysfunctional divisions between the host culture and the imperialistic, but also this dysfunctional division can go deep into society and the economical framework of the society. The assumption of the neutrality of technology is flawed as it fails to recognize that technology develops in a cultural setting and consequently embodies that culture within its design. The developments in ICT are based on a Western alphabet-oriented culture that contrasts with the ideogram of the North Eastern Asian cultures, (Shepard, 1993; Haywood, 1995). The development of the Qwerty keyboard and the encoding of the characters of the West (predominantly English) in the design of 8-bit and ASCII characters within 101 keyboard do not easily support the Chinese characters nor their embodiment of the ideas and culture of that society. This dysfunctional division suggests a limitation on growth and a nonlinear development path. Such a restriction suggests serious economical as well as cultural consequences. Current predictions of growth based on ICT suggest no limits, but the reality is that society is already dividing and the migration pathway between the two communities is increasingly difficult due to access barriers and cultural differences. The access barriers are primarily economic, reflecting the divide in economic benefit of development rather than the technical access issues as it is the same territories that have become the primary home of manufacturers of the technology. The juxtaposition of production of ICT resources in factories for which the office is still primarily manual with limited access to ICT emphasizes the divide (Graven, 1994). The East Asian reality of the urban technology worker and the peasant economy of

94 Kaye & Little

much of this region heightens the divide. In the former, the access to technology may be exploited as the community adopts the Western culture, hence joining the global development pathway. In contrast, the rural community may increasingly be disadvantaged through lack of access and increasing cultural and economical barriers to entry. These conditions may be reminiscent of the industrial revolution, but in the west agrarian developments were more harmonious with industrial. Today the rate of technological development in ICT is such that a dramatic gap may exist well within a generation. Also contributing to the nonlinear development path are the uneven economic access and cultural barriers which advantage some groups and disadvantage others. Consequently, globalization and universality are not achievable; rather a divided society is likely to emerge with the “haves and have-nots.” This may further exacerbate the problems of development.

ENDNOTES 1 The following Websites contain details of these applications: www.twinbridge.com/ www.pspinc.com/lsg www.macos.apple.com/multilingual

REFERENCES Attewell, P. and Rule, J. (1984). Computing and organizations: What we know and what we don’t know. Communications of the ACM, 27(12), 1184-1191. Benbaset, I., Dexter, A. S., Drury, D. H. and Goldstein, R. C. (1984). A critique of the stage hypothesis: theory and empirical evidence. Communications of the ACM, 27, 476-85. Brynjolfsson, E. and Hitt, L. (1993). New Evidence on the Returns to Information Systems. Cambridge, MA: MIT Press. Brynjolfsson, E., Malone, T. W., Gurbaxani, V. and Kambil, A. (1994). Does information technology lead to smaller firms? Management Science, 40(12), 1628-1644. Davenport, T. H. and Short, J. E. (1990). The new industrial engineering: information technology and business process redesign. Sloan Management Review, 31, 11-27. Davenport, T. H. (1993). Process Innovation. Boston, MA: Harvard Business Press. Davenport, T. H. (1996). Why reengineering failed: The fad that forgot people. Fast Company, Premier Issue, 70-74. Drury, D. H. and Bates, I. E. (1979). Data Processing Charge-Back Systems: Theory and Practice. Hamilton, ON: Society of Management Accountants of Canada. Drury, D. H. (1980). A survey of Data Processing charge-back practices. INFOR, 18(4), 342-53. Drury, D. H. (1983). An empirical assessment of the stages of DP growth. MIS Quarterly, 7(3), 59-70. Earl, M. J. (Ed.). (1996). Information Management: The Organizational Dimension. Oxford: Oxford University Press. Economist (1997a). A divided nation (again). The Economist, October, 80-83 Economist (1997b). Wiring corporate Japan, The Economist, April, 80-83 Freidman, A. L. (1994). The stages model and the phases of the IS field. Journal of Information Technology, 9, 137-148.

Dysfunctional Development Pathways 95

Freidman, A. L. and Cornford, D. S. (1989). Computer Systems Development: History, Organization and Implementation. London: John Wiley. Galliers, R. D. (1991). Strategic information systems planning: Myths, Reality and Guidelines for Successful Implementation. European Journal of Information Systems, 1, 55-64. Galliers, R. D. and Sunderland, A. R. (1991). Information systems management and strategy formulation: the ‘stages of growth’ model revisited. Journal of Information Systems, 1, 89-114. Gibson, C. F. and Nolan, R. (1974). Managing the four stages of the EDP growth. Harvard Business Review, 52, 76-88. Granet, M. (1934). Le Pensee Chinoise. Paris as cited in Goody, J. and Watt, I. (1975). The consequences of literacy. In Goody, J. (Ed.), Literacy in Traditional Societies. Cambridge: Cambridge University Press. Graven, K., (1994). The Eastern front, CIO, October 1. Reported in Gupta, U. G., Management Information Systems: A Managerial Perspective. Minneapolis St Paul: West Publishing Co. Hall, E. T. (1976). Beyond Culture. New York: Anchor Books/Doubleday. Hall, R.K., (1949). Education for a New Japan. Newhaven: Yale University Press. Hampden-Turner, C. and Trompenaars, F. (1993). The Seven Cultures of Capitalism. NewYork: Doubleday. Haywood, T. (1995). Info-Rich Info-Poor: Access and Exchange in the Global Information Society. London: Bowker-Saur. Hofstede. G. (1980). Culture’s Consequences: International Differences in Work-Related Values. London: Sage. Jayasriya, R. (1993). Stage of growth in end-user computing: applications in the health sector of developing countries in Asia Pacific. Journal of Information Technology, 8, 151-159. Kaye, G. R. and Little S. (1996). Setting standards: Strategies for building global business systems. Information Technology & People, 9(3), 30-54. Keniston, K. (1997). Software localization: Technology transfer and or cultural imperialism. IRMA 97, Vancouver, Canada. Kirlidog. M. (1996). Information technology transfer to a developing country: Executive information systems in Turkey. Information Technology & People, 9(3), 55-84. MacFarlan, F. W. and McKenney, J. L. (1982). The information archipelago: Maps and bridges. Harvard Business Review, 60, 413-25. Matsuyama, K. (1992). Agricultural productivity, comparative advantage and economic growth, Journal of Economic Theory, 58. Motorola. (1996). Making Use of Technology: A Motorola Report. http://www.mot.com/ General/Reports/British-Tech/. Nolan, R. (1979). Managing the crisis in data processing. Harvard Business Review, 115-126. Nonaka, I., Ray, T. and Umemoto, K., (1998). Japanese organizational knowledge creation in Anglo-American environments. Prometheus, 16(4), 421-438. Ohmae, K. (1990). The Borderless World. London: Collins. Okot-Uma, R. W. O. (1988). A synthesis perspective of information technology capability of commonwealth countries. Science & Technology News - Special Issue on I.T. & The Commonwealth. Perrin, N. (1982). Keine Feuerwaffen mehr–Japans Ruckkehr zum Schwert 1543-1879. Frankfurt. Roach, S. (1985). The New Technology Cycle. New York: Morgan Stanley Economic Perspectives.

96 Kaye & Little

Roach, S. (1992). Technology Imperatives. New York: Morgan Stanley. Robertson, R. (1992). Globalization: Social Theory and Global Culture. London: Sage. Rodrik, D., (1997). Has Globalisation Gone Too Far? Washington DC: Institute for International Economics. Rogers, E. M. (1983). Diffusion of Innovations, (3rd edition). New York: Free Press. Sachs, J., Bloom, D., Panayotou, T., Radlet, S. and Williamson, J. (1997). Emerging Asia. Asian development bank report as reported in Economist, 14 June, 21-24. Shepard, J. (1993). Islands in the (data) stream: Language, character codes, and electronic isolation in Japan. In Harasim, L.M. Global Networks: Computers and International Communication. Cambridge MA: MIT Press. Slaughter, M. and Swagel. P. (1997). The effect of globalization on wages in the advanced economies. IMF working paper, April. Strang, D. and Meyer, J. W. (1994). Institutional conditions for diffusion. In Scott, W. R. and Meyer, J. W. (Eds.), Institutional Environments and Organizations: Structural Complexity and Individualism. Thousand Oaks, CA: Sage. Unger, J. M. (1996). Literacy and Script Reform in Occupied Japan. Oxford: Oxford University Press. Wastell, D. G. and Sewards, A. (1995). An information systems profile of the UK manufacturing sector. Journal of Information Technology, 10, 179-189.

Ten Lessons That Internet Auction Markets Can Learn 97

Chapter VIII

Ten Lessons That Internet Auction Markets Can Learn from Securities Market Automation J. Christopher Westland Hong Kong University of Science and Technology, Hong Kong

Internet auction markets offer customers a compelling new model for price discovery. This model places much more power in the hands of the consumer than a retail model that assumes price taking, while giving consumers choice of vendor and product. Models of auction market automation has been evolving for some time. Securities markets in most countries over the past decade have invested significantly in automating various components with database and communications technologies. This paper explores the automation of three emerging market exchanges (The Commercial Exchange of Santiago, The Moscow Central Stock Exchange, and Shanghai’s Stock Exchange) with the intention of drawing parallels between new Internet models of retailing and the older proprietary networked markets for financial securities.

INTRODUCTION Internet auction markets, such as those offered by Amazon, eBay, Priceline, OnSale, and CNET’s Shopper.com, are earning increased business and investment by offering customers a new model for price discovery. This model places much more power in the hands of the consumer than a retail model that assumes price taking, while given consumers choice of vendor and product. The monetary impact on retailing is currently small (Internet E-commerce in the U.S. in 1998 totaled $7.8 billion. Compare this to Wal-Mart’s retail sales in 1998 of $130 billion, or total U.S. retail sales in 1998 of $1.7 trillion. But Internet sales are growing faster than traditional retailing, reflecting the appearance of auction-based price discovery that has been made possible by Internet automation of many retailing outlets. Appeared in Journal of Global Information Management, vol. 8, no. 1, 2000. Reprinted by permission.

98

Westland

Models of auction market automation have been evolving for some time. For example, securities markets in most countries over the past decade have invested significantly in automating various components with database and communications technologies. Various other technologies have a mechanized securities markets for over a century; for example, stock tickers have provided automated real-time reporting of securities prices for nearly a century. Without automation, markets are constrained to operate at the speed of their human facilitators—frequently too slow and localized for complex or high volume market services. This paper explores the automation of three emerging market exchanges (The Commercial Exchange of Santiago, The Moscow Central Stock Exchange, and Shanghai’s Stock Exchange) with the intention of drawing parallels between new Internet models of retailing and the established proprietary networked markets for financial securities. Emerging market innovations can elucidate more clearly specific issues arising in automation, because the projects are not hampered by tradition, volumes may be smaller, trading more localized, and offerings more homogeneous. It may thus be easier to discern the rationale behind a particular technology choice. Each of these three exchanges has, over the past decade, experimented with information technology appropriate for its market. Each discovered unique issues and pitfalls in automating its particular exchange operations. This article summarizes what was learned from market automation. The implications for automation of retail markets in general (what has come to be called electronic commerce, or e-commerce are drawn from the lessons learned in automating these securities markets.

CASE STUDY IN THE AUTOMATION OF SHANGHAI’S STOCK EXCHANGE Deng Xiaoping initiated the “responsibility system” in 1979–a capitalist innovation which abolished central quotas and allowed farmers and some township enterprises to sell their goods on the open market. Many became rich in the ensuing decade. Savings grew with the growing wealth of the populace, endowing China with one of the highest savings rates in the world—between 35% and 40% of GDP over the past decade. With the introduction of economic reforms in the 1980’s, average annual growth was pushed to nearly 10%. This modernization required substantial amounts of investment capital, which was the main role of the Shanghai Stock Exchange, China’s main securities market. Most trading takes place in what are called “A” shares, which may be traded on the floor, but more often, trading bypasses the floor completely and is handled through automated systems. The Shanghai Stock Exchange issues a security card that identifies an individual as being authorized to trade in listed securities. The security card allows traders to bypass brokerage firms (and reduce brokers’ commissions). This provides a more efficient use of the floor, by channeling routine trades directly to computer matching. Trades are captured at a brokerage room when traders post an order with the clerk. The order is posted in the Exchange’s system through computer-to-computer communication. Matching takes place automatically. Security card holders still have to go through a brokerage firm to buy or sell securities and to settle and clear the transaction. Security cards make the job of brokerage firm easier by identifying the client to the brokerage and allowing automated management of the client’s account balance. In

Ten Lessons That Internet Auction Markets Can Learn 99

Table 1: Market Capitalisation of Selected Markets in 1997 Exchange New York Tokyo London Hong Kong China (Shanghai-Shenzhen) Russia 1 2

Capitalisation Billions of U.S.$1 7600 2685 1500 440 140 100

Stockmarket Turnover % of Market Capitalisation2 84 47 42 46 225 10

Asiaweek, 28 Feb. 1997, p. 62 Economist, 14 June 1997, p. 124

contrast, “B” shares (which are designed for trading by overseas investors) are only traded through brokers and on the floor. Liquidity is quite good in “A” shares, where there is widespread participation by China’s populace through approximately 3000 brokerage rooms around China. Turnover is around five times per year. “B” shares are less liquid, and there has been considerable interest in engaging more overseas investors in their purchase. Much of the current trading in “B” shares is done by Chinese within China, though they must pay for these shares with dollars in accounts outside China. “A” shares can be traded by any trader with a Shanghai Stock Exchange issued security card. These traders can trade directly with the exchange’s computer system. Thus orders may come directly from trading counters—where there is a transaction-by-transaction cash settlement and exchange of securities—or through member brokers—where the exchange provides net settlement of brokers accounts. Settlement is completed in the same day as the trade (T+0). Exchange commissions are approximately 0.65% of trades. As in many emerging markets, price volatility can be more strongly influenced by the money coming into the market than by the business fundamentals of the traded firm. Volatility in the Shanghai market is very high by the standards of most developed economies, a situation that poses difficulties to orderly trading. This tends to compound the exchange’s challenges in the face of problems inherent in securities laws that are still evolving and in the difficulties faced by the China Securities Regulatory Commission in policing insider trading, misuse of funds, and false disclosure. Volatility is tightly monitored—important in a market where there is a rapidly expanding group of investors of varying levels of sophistication. The exchange will not allow prices for a given security to vary more than 10% from the prior day’s closing price. When the 10% mark is reached, trading is not halted. Rather the settlement prices simply are not allowed to exceed 10% of the prior day’s price. This limit does not apply to initial public offerings (IPOs) on the first day; it only takes effect on the second day after issue. The Shanghai Stock Exchange (and its sister exchange in Shenzhen) use a continuous double auction, order-driven trading system, assisted by a computer network to transfer order information from brokers to the floor and back again. During opening hours, information on trade prices and volumes is continuously disseminated to traders, and buy and sell orders

100 Westland

are continuously received. Both have supporting clearing houses for transaction completion and settlement. The exchange floor opens Monday through Friday from 9:30am until 11:30am (at which time prices are frozen) and re-opens after lunch from 1:00pm to 3:00pm. Opening prices are generally set to clear the maximum number of outstanding bid and ask orders on hand at 9:15am. They may be set differently if there are substantial imbalances between buying and selling, and it is felt that a different price is appropriate to clear the market. The opening prices for thinly traded stocks will be set at their closing prices from the last business day. The trading system runs on a Hewlett-Packard HP9000-T500 computer, with peak processing of 5000 transactions per second. This is sufficient to handle peak exchange volume of two million transactions per day, with peak periods at the beginning and end of the trading day. The HP9000 is, with the addition of processing boards, scalable to 20 million transactions per day. Cisco routers regulate traffic on the local and wide area networks controlling traffic to and from brokerage rooms throughout China. These rely on a fault tolerant combination of direct digital network, satellite and analog transmission. The exchange has been clever in applying a cost conscious set of components for exchange automation that make effective use of both microcomputer and state-of-the-art computer and communications technology. Trading on the Shanghai Stock Exchange is dominated by small investors–99.4% of the Shanghai Stock Exchange’s clients are individuals; another 0.6% are institutions with only a vestige of direct participation by securities houses. The Shanghai Exchange has implemented a number of innovations to bring online trading to the small investor. China’s limited telecommunications capacity requires several levels of support to reach its desired customer base. Brokerage rooms provide the equivalent of Internet cafés for those who do not have a telephone, modem, and computer at home. In addition, real-time stock transaction quotations are provided online, free to subscribers of Shanghai’s cable television service. A computer card can provide the same cable TV quotation service, but on a computer screen; over 50,000 were sold in their first year. Traders with a modem and computer can open a trading account with one of the local banks. The exchange then provides them free connectivity software and free real-time quotations, allowing trading in and out of that bank account. The StockStar electronic trading Internet site provides services similar to E*Trade in the U.S., though the level of support provided to traders directly from the exchange has provided it little latitude for profit. Information dissemination on exchange transactions is a significant problem because of China’s size. China is a continent-sized country which needs sophisticated communications systems to allow access to its exchanges. Shanghai’s Stock Exchange has invested in two networks: STAQ, a national-wide system for broadcasting transaction prices; and NET, an automatic security trading system for stocks as well as government bonds. Sixteen networked securities trading centers with broadcast and screen facilities have been established around China. In 1993, these received new fiber optics and satellite communication systems which replaced dedicated telephone lines. Communications are now supported by an optical fiber network throughout the city of Shanghai. A combination of local fiber loops, dedicated satellite communication systems (using the AsiaOne satellite for communications linkage), and telephone lines offers two-way communication with 3000 trading counters in over 300 cities around China. There are roughly five million investors around China, and around half of the monetary trading volume originates outside Shanghai.

Ten Lessons That Internet Auction Markets Can Learn 101

The exchange floor has gone through substantial changes in its first decade. It has supported as many as 6000 seats (i.e., booths with microcomputers and telephones) spread over eight trading halls, representing the 500+ authorized financial institutions who are members of the exchange. Firms must have registered capital of more than RMB 5 million to be members. The new 27-floor exchange building in the PuDong (literally Shanghai East) area of Shanghai will dedicate all space from the ground up to the ninth floor to exchange operations. The exchange floor in PuDong holds 1700 traders on a floor double the size of Tokyo’s and triple the size of Hong Kong’s floor. Most business, though, is conducted off the floor, through terminals and automated matching and floor brokers spend a disproportionate time reading their newspapers. The floor provides no formal trading or market-making functions. It provides a focal point for market management and supervision and for market sentiment. Li Quian, an exchange official, explains that “It is also a symbol of Shanghai’s success.” To facilitate transparency of market activities, the communications network automatically disseminates trading information: 1. to the exchange’s trading room display (a large digital screen on the trading floor), 2. to a telephone inquiry network for people not on the trading floor, 3. to over 20 news organizations, TV and radio stations, 4. to Reuters, Telerate, and other global financial services, and 5. to the public by publishing a newspaper, Shanghai Security, with a circulation of several hundred thousand. There are 1000 minor satellite downlink stations and one major uplink station in PuDong. In brokerage rooms in Shanghai and other major cities in China, satellite systems typically provide backup order placement functionality in case telephone lines are not operable. In remote parts of China, such satellite systems are the sole communication channel allowing investors to participate in the trading of securities listed on the Shanghai Stock Exchange. Since 1993, the state has encouraged individuals and institutions outside Shanghai to participate in securities listed on the Shanghai Stock Exchange, which it would like to truly be a national exchange. There are currently more transactions posted by traders outside than inside Shanghai. The exchange actually engages two systems vendors to provide software and services for the collection and dissemination of trading information. They provide basically the same service. Since customers can choose either one, this provides considerable incentive for both vendors to maintain their quality of service and sophistication of software. China’s Securities and Exchange Commission worked to insure reliability of disseminated information. It has installed stringent controls over the information reported by companies and has prohibited the accounting profession from certifying forecasts of corporate performance. Information on all transactions is also transferred to the exchange’s market monitoring group which attempts to control insider trading, rumors, and collective efforts to control prices. First priority in automation of exchange functions has gone to the “A” shares. “B” shares are purchased in dollars, though denominated in renminbi (RMB), and information on those shares is broadcast in both English and Putonghau. Thus, additional systems development is required in supporting “B” shares. Customer participation in any auction market requires that customers have confidence that orders will be priced fairly and not subject to manipulations which would put them at a disadvantage. Chinese securities laws impose some order on the off-floor brokerage function.

102 Westland

Brokers are required to comply with customers’ explicit terms for transactions conducted on their behalf, including: (1) type of security, (2) volume of trade, (3) bidding conditions and margin level, and (4) time of authorization. To prevent “insider” trading, employees of securities authorities, managers of the securities exchanges, employees of the broker handling the transaction, and employees of the governmental agency regulating or controlling the issuing company are prohibited from trading in its shares (Zhao and Li, 1992). Compliance with these guidelines is enforced by a Monitoring Department and a Compliance Department (i.e., investigation) in the exchange organization. Seats in the Shanghai Stock Exchange are classified as “real” and “virtual.” Virtual seats are either remote terminals or handsets that are connected to the matching software through DDN, satellite uplink, or analog telephone link. There exist 1000 low bandwidth satellite uplinks for posting orders remotely, and one broadband satellite uplink (in Pudong at the Shanghai Stock Exchange) to disseminate bid, ask and settled trade prices, and volumes. Order placement security is enabled by either a touch tone or a magnetic card verification system. The touch tone system requires the cashier/broker to enter trader identification, security code, password, limit price, and quantity of the buy or sell order. The magnetic card system (i.e., a standard-sized credit card for stock transactions) enters trader identification and security code automatically requiring keying only of password, limit price, and quantity of the buy or sell order. Magnetic card system is provided by the Shanghai Stock Exchange and provides an additional level of security over trading. Individuals wishing to trade in the Shanghai Stock Exchange need to procure a magnetic security card from the exchange. This card uniquely identifies the trader’s account with the exchange. Since the market is a cash market, every share is owned by someone, and this is recorded in the exchange’s databases. The exchange handles its own transfer/clearing accounting. Through this account identification, traders can obtain information about their account position from the exchange. Banks and brokerage houses issue separate debit cards which allow traders to buy and sell without exchanging cash. These reflect account balances with the bank. Traders who do not have a security card can trade, with the broker acting on their behalf. Some banks incorporate the security card number and information into the debit card. Not all brokers have a direct connection to the exchange. Particularly in remote areas of China, brokers need to dial into their seat on the floor (or through satellite uplink) to place orders. Even in Shanghai, some brokerage firms insist on their own staff dialing for traders, even if the traders can directly place orders with the exchange. This provides an additional layer of security and control over trading. Clearing and settlement are the responsibilities of a wholly-owned subsidiary of the Shanghai Stock Exchange—the Shanghai Securities Central Clearing and Registration Corporation. Central depository, trades, and clearing are all paperless. There is no need to print a physical copy of the security, as a database is maintained of ownership of all shares. A particularly exciting innovation in Chinese securities trading in recent years has been the introduction of neighborhood brokerage rooms to encourage the investment of funds by China’s people, who have one of the highest savings rates in the world. The immense popularity of trading counters and brokerage rooms in China has done much to abate liquidity shortages that dogged the market until recently. Before the widespread installation of trading rooms, China’s securities markets suffered from a lack of capacity to absorb large buying or selling pressure without causing severe adverse price movements.

Ten Lessons That Internet Auction Markets Can Learn 103

China’s investment banks rely on a clever mix of appropriate technologies, cost effectively delivered, to bring their services to the people. Brokerage rooms provide just the right amount of technology for a country with straggling telecommunications infrastructure and few PCs. Electronic investment technology is brought close to every home, but not into it. Investors meet in a pleasant and convivial neighborhood atmosphere. In China, as elsewhere, more is not always better. Appropriate technology is intelligently embraced which fits the customer and the tasks at hand. Small investors are served in a large hall crowned with an electronic bulletin board broadcasting the latest securities prices. In the front of the hall are windows through which clerks can process, buy, or sell orders for listed securities. The bank also provides software for technical analysis of securities prices on computers at the front counter. Over 50 algorithms for technical analysis may be called up in the software; moving averages and other cycle analysis can be customized by the user. Major investors–those with over RMB 500,000 in the market–are treated to their own desks and computers (complete with technical analysis software) in one of several rooms cloistered within the maze of halls leading from the main brokerage room. Despite the greater stakes, the unique and convivial atmosphere of investing still permeates these rooms. Though orders could be posted directly to the Stock Exchange’s electronic matching system directly by these computers, the China Commercial Bank chooses to act as an intermediary. A secretary is provided for each of the VIP rooms and actually places buy or sell orders at the request of the investors. Besides providing a professional touch for the customerinvestor, this also provides the bank with an additional modicum of control.

CASE STUDY IN THE AUTOMATION OF MOSCOW’S CENTRAL STOCK EXCHANGE Russian president Boris Yeltsin’s government has rapidly moved to open markets since its economic reform program was launched in January 1992—by freeing prices, cutting defense spending, eliminating the old centralized distribution system, completing an ambitious voucher privatization program in 1994, establishing private financial institutions, and decentralizing foreign trade. The Soviet economy was founded on heavy industry–on large-scale factories, smelters, refineries, extraction and processing of raw materials and natural resources. Much of Soviet budgetary allocation through the 1950’s and 1960’s was directed toward infrastructure development. The country formed a national electricity grid (now a joint stock company known as “Unified Energy Systems”), a gas exploration, drilling, and refining behemoth (now joint stock company “Gazprom”), a nationwide telecommunications system (“Svyazinvest” and “Rostelecom”), oil exploration and distribution corporations (“Lukoil” and “Surgutneftegaz,” inter alia), car and truck manufacturers (“Logovaz,” “Zil,” and “Kamaz”), diamond and gold processing organizations. These industries provided the securities marketed on the Moscow Central Stock Exchange. The establishment of modern financial markets and the modernization of banking helped the non-state sector to contribute approximately 75% of official GDP by 1997, up from 62% in 1994. This contribution may understate the true contributions from privatization. By some estimates, the official GDP figure of around US$700 billion (at purchasing power parity, with a bit over 10% contributed by exports) is only 50% of the actual GDP. The underreporting

104 Westland

of economic activity to avoid an inequitable tax system obscures actual income. Russians save 32% of their income (slightly short of the Chinese savings rate) while the Russian government consumes only 15% of the official GDP (compared to 30% of GDP in the U.S.). These amounts are comparable to state sector consumption in China, where official figures understate GDP and faulty tax collection is again to blame. To privatize these firms, the Russian government issued vouchers, worth ten thousand rubles each, to every Russian citizen (children included). These could be picked up at local offices of the state bank for a nominal transaction fee; eventually some 144 million out of 147 million Russians received their vouchers. They could be exchanged for shares in companies through the mechanism of auctions (essentially they were currency but could only be used to bid for shares. Markets grew for buying and selling vouchers; they were even sold from farmers kiosks just like carrots or cabbages. Voucher funds’ ability to influence enterprise performance depended on a number of factors, including management compliance with new policy, management disclosure practice, and support by local and regional government officials. With the absence of access to enterprises’ financial information, however, voucher funds were severely restricted in their evaluative capacity–a significant liability for an instrument which was initially motivated by a need to value enterprise assets. Given voucher funds’ relatively small ownership stakes, the funds also often required the cooperation of other shareholders in order to put forward new policies. The Moscow Central Stock Exchange (MSE) opened in 1992 as the first over-thecounter marketplace in Russia. It became the focus of the fledgling equities trading market at the first stage of privatization. Russia’s first standardised futures contract was designed by Moscow Stock Exchange’s Clearing House in March 1994 to facilitate trading on USD/ RUR exchange rate. In November 1995 a government bond (GKO) market evolved, and the MSE offered futures contracts on GKOs (index-linked, yield curve, when issued series, physically deliverable, etc.) with American-style options on futures as underlying. Later, contracts on Federal floating rate Notes (OFZs) were added to the list, making the Russian Bond futures market as an MSE franchise. Since then, the GKO and OFZ futures contracts have become increasingly important additions to the domestic treasury bond cash market. In June 1997, the exchange’s daily traded volume was between 30-40% of GKO and OFZ overall spot market turnover. These figures were considered proof of the growing need of large institutions to counterbalance their volatile GKO/OFZ holdings (borne out in the collapse of the debt market in August 1998). The exchange operates a screen-based multifunctional trading system. Trading is order driven, in a continuous double-price auction, where orders may be placed by interactive remote access. The exchange has implemented online clearing which has added liquidity and traded volume. The exchange has engaged IBM to migrate from existing PC-based network to IBM RS/6000 servers. The IBM Global Network (IGN), utilized by the exchange, makes viable the full-scale remote access and order routing for overseas brokers. During the early days of equity trading, Russia’s market consisted of brokers and dealers with stock shares, vouchers, and other privatization certificates. Price search was conducted over the telephone and fax (reminiscent of the over-the-counter market that preceded the US’s National Association of Securities Dealers Automated Quotations system) and by traveling to distant regions by plane, train, and sometimes even on foot. In 1994 and 1995, a broker’s trip to a distant region where shares of Lukoil or Rostelecom were being sold at below-market prices by local residents (and, occasionally, by company officials) could net a 100% return on purchased shares merely from transporting them to

Ten Lessons That Internet Auction Markets Can Learn 105

Moscow. This may mark one of history’s extreme cases in a liquidity-driven market. Improved market efficiency from automation dropped that geographic-liquidity-induced return to around 20% by 1997. As equity trading became more regulated, and information spread to the regions through paper and Internet information bulletins from 1995 on, brokers and traders moved to computerized trading systems such as the Russian Trading System and AK&M (similar to the NASDAQ). Despite the large size of Russia’s economy, total capitalisation of shares listed on the Moscow Stock Exchange still amounted to only around US$100 billion in 1997 (up from around $20 million two years earlier). Though Russia’s market capitalisation is roughly that of China’s stock markets, turnover was less than one-twentieth of China’s (Table 1) at the time. However, the August 17th 1998 moratorium on repayment of Russian government debt threw the stockmarket into turmoil. By the end of September 1998, trades are less than $10 million per day (down from 10 times that amount only six month’s earlier), and liquidity is almost nonexistent as reflected in bid-ask spreads running from several 10s to several 100s of percent. By 1997, a host of computerized systems existed in Russia’s investment houses to actively and effectively process trades. While no singular system dominated the Russian market, larger investment houses were focusing on system implementation as trading volumes doubled and tripled. At Renaissance Capital, Russia’s premier investment bank (headed by former CSFB Russia Director, Boris Jordan), for example, ten Sun Microsystems servers running proprietary software processed trades and analyzed positions for the company. Dealer boards—electronic devices that allowed equity traders to hold several telephone conversations at once—were installed in early 1997. Renaissance had a telephone system of 450 lines for its equity traders, with a Western-style trading floor with raised steel floors overlying high-bandwidth communications lines. Most Western brokerage and trading houses import their own systems and procedures from the U.S. or Europe. In contrast, Russian banks prefer to keep their proprietary information technology closed to outsiders. Russia has a well-educated workforce that excels in technology products such as software. By keeping systems and data feeds proprietary, banks can keep outsiders from knowing secrets that they believe provide them with a competitive edge. Financial statements of listed firms typically are stated under both Russian Accounting Standards and International Accounting Standards (similar to U.S. Generally Accepted Accounting Principles). Russian Accounting Standards provide more accurate accounting under inflation, or when asset valuation is ambiguous, as it is in an economy that still settles 40% of its industrial purchases through barter. Both Russian and Western banks use Russian accounting software packages, the two major packages provided by Diasoft and Program Bank. As the market matures, information technology providers should see considerable growth in business volume. In contrast to more developed markets where settlement and clearing are instantaneous processes linked to price search, Russian back-office software “has to deal with a market in which completing a trade [often requires] sending a courier to Surgut to get the trade recorded in the company register.” Furthermore, equity traders require at least one other employee in the back office to register trades, obtain stamps and seals, and comply with mountains of government regulations. Russia’s market regulating authorities fully understand and acknowledge the difficulties trading and brokerage houses face in a market dominated by telephone trade. According to Skate Press and several Moscow brokerages, at least 60% of trades are completed over

106 Westland

the telephone and not registered with regulatory bodies. Furthermore, much of this trade takes place “offshore,” making trade tracking all the more difficult. This creates a major problem for both market makers and investors, as actual trading (market) prices often differ significantly from published (recorded) prices. As an example, for most of early August 1997, recorded trade volume for Gazprom averaged 100,000 shares traded per day. Actual trade volume, according to several Moscow-based brokers, was at least twenty times that amount. Because settlement of trades (a paper-based system) takes place independently of price negotiation and discovery (automated on the computers of the stock exchanges), prices can be manipulated and published on the exchange, without actual trades ever taking place. Unfortunately, because of poor liquidity and policing, broker abuses are common in Russia. Insider trading is rampant. Laws exist but are seldom enforced. And other widespread practices have proved even more damaging. In one example, written up in Russia Review, Johnny Manglani, an Indian tailor and amateur investor, claimed his Russian broker cheated him out of more than $100,000 on the Russian market. According to Manglani, Charles Shearer of the Moscow investment house Rinaco Plus called one night with bad news: Sberbank preferred stock, which Manglani had bought the previous year at $4.80 per share, was down to $1.70 and would shortly fall to $0.30. Manglani wanted to hold on but claims that Shearer, who had not previously taken much interest in his trades, kept calling until he agreed to sell. Soon after, Sberbank preferred jumped to $2.15. This is an example of an investor not getting the best price. Other common abuses include: Price manipulation. Because the Moscow Stock Exchange does not handle settlement, 80% of trade in Russian shares happens outside exchange, and investors are often left at the mercy of their broker to tell them how much a given stock really costs on a given day. Favoring the house. Many of Russia’s biggest houses do a lot of trading on their own account, known as the “proprietary book.” When the traders want to get out of something, they tell their clients that it is a great buy. Front-running. In a market as illiquid as Russia’s, one big traCan send a stock’s price soaring. A broker can first buy a little for the proprietary book, or for himself, then profit when the price rises on the client’s purchase order. Brokers can also front-run on research reports. Research flogging. In an up market, brokerages come out with glowing reports on companies in which they already have stakes. For example, the brokerage house Aton issued a report in September 1997 recommending the previously obscure Kazan Helicopter Factory. Aton analyst Nadeshda Golubeva noted that Aton took a “big stake” in the factory before even beginning research. Soon after Aton’s report came out, the stock plunged, and by March 1998 was removed from the Russian Trading System. Drawing the line between ethical and unethical behavior can be difficult in an emerging market such as Russia’s. In a mature market, when a bank’s research department issues a buy recommendation on a formerly obscure company, clients can immediately purchase the shares through their brokers. In Russia, acquiring shares for such second-tier companies usually requires that someone physically travel to the company and buy the shares in cash from workers and other small shareholders, a process which can take weeks. Procurement can require brokers to drive long distances over dilapidated country roads, carrying millions of rubles along with a bodyguard or two. Moscow brokers need to start buying shares in advance so they will be available when the research report is published. In the U.S. this would be called front-running, but in Moscow, many fund managers are willing to pay a premium for this service.

Ten Lessons That Internet Auction Markets Can Learn 107

On August 17th 1998, the Russian government defaulted on its own debt, ordered private borrowers to default on foreign loans, and abandoned its support for the ruble. The stock market, never liquid in the best of times, literally ground to a halt. From 1994 to 1998 the Russian government raised around $45 billion from selling securities, while the Russian firms issuing these securities ran up debts of another $20 billion. During the same period, an estimated $60 billion left Russia as “flight capital,” finding new homes in offshore accounts in the Grand Caymans and Bermuda. (The Economist, 19 December, 1998) Many of the weaknesses in Moscow’s Central Stock Exchange were symptoms of deeper faults that exposed Russia’s worst inclinations. In the aftermath of the Russian privatization debacle, governments throughout Europe and Asia reconsidered the wisdom of open capital markets. Yet in Russia, it was not the concept of open capital markets that was to blame–(it was their implementation through incomplete and flawed electronic trading systems. Three specific failings destroyed Russia’s financial credibility: 1. De facto limitation of participation to a small circle of traders with inside information. 2. Separation of price discovery and settlement operations, which allowed brokers wide latitude in manipulating posted market prices. 3. Failure of listed firms to provide transparent information about the quality and profitability of their operations. Had Moscow’s Central Stock Exchange designed these problems out of their system prior to operation, the denouement might have been much different.

CASE STUDY IN THE AUTOMATION OF SANTIAGO’S COMMERCIAL STOCK EXCHANGE Chilean privatisation and economic reforms were initiated under the régime of Generallissimo Augusto Pinochet. Over the 1980’s these reforms steadily improved the statecontrolled economy inherited from Salvador Allende’s Marxist régime. By the time that Pinochet resigned in 1990, Chile had the strongest economy in Latin America. Shares of Chilean firms were traditionally traded in the Bolsa Comercio de Santiago (Santiago’s Commercial Stock Exchange), which was Chile’s largest and oldest stock exchange. Trading volume exploded in the 1980’s with the initiation of a national pension plan that allowed workers to choose their own investments. During that period, Chile’s economy grew in excess of 10% annually. An urgently needed update to the automated systems of the exchange took place from 1992 to 1994. The exchange automated in the face of stiff competition from rival Bolsa Electrónica, a completely electronic exchange started in 1989. The Bolsa Electrónica used a computer trading system very similar to the new electronic system developed for the Bolsa Comercio, but without a trading floor or the accompanying expenses of a floor. In just three years it had captured 30% of transactions in securities of listed companies, where virtually all of the companies listed on the Bolsa Electrónica were also listed on the Bolsa Comercio. The Bolsa Electrónica was cannibalizing transaction volume directly from the Bolsa Comercio! Success fomented new demands by brokers. Many brokers resided in the exchange building, which possessed Ethernet network links providing much faster communications than existing 9600 Baud lines. They were willing to pay the Bolsa considerably more for access to Ethernet links. But linking some brokers to fast communications lines, while leaving other brokers to trade on slower 9600 Baud lines, could create a caste system which would rob

108 Westland

otherwise qualified brokers of business simply because of their location. Even more divisive was the demand by brokers for the ability to “program trade”—to trade automatically relying on decisions made by software programs, using information gleaned from the Bolsa’s electronic information feeds. Both problems presented significant challenges to the Bolsa’s fairness policy, by allowing brokers who invested in information technology to place orders faster through precedence in information processing speed. Fairness, in the sense that no trader could systematically gain an advantage over another trader by exploiting idiosyncrasies of the exchange mechanism, was particularly important in Chilean trading. Many of the traders traded on their own account, with family, rather than institutional capital. Family traders would be less inclined to invest in information technology than institutions and brokers, and without the Bolsa’s commitment to fairness, would be at a disadvantage. The Bolsa was legitimately concerned that this, in turn, might drive away their business. So important was fairness to the Bolsa that they insisted on owning the PC terminals used by brokers and tightly controlling the software residing on those PCs, assuring that no program trading was spirited onto the PC platforms. They recently sold their new computer trading system to the Cali exchange in Columbia, which supported trading in both Cali and Bogota. In order to assure fairness, trading in Cali was purposely delayed by a fraction of a second to equalize order posting delay originating from telephone switching delays from Bogota 200 miles away. In 1992 the Bolsa Comercio supported the following three market mechanisms for trading: 1. Open outcry. Open outcry is the traditional method for market trading in a “pit.” This market is opened from 9:00am to 11:45am Monday through Friday. Here, for example, a trader enters the “stock” pit, shouts out a particular price to buy a number of shares, another trader shouts back acceptance of the price, constituting the consummation of a contract. 2. Bulletin board. This auction system is used primarily for thinly traded stocks, where it may take several days for bids and asks to match. This was traditionally a chalk board on which any listed company could post securities. 3. Direct order. This exchange allows person A to request to buy or sell directly from or to person B but broadcasts the offer to see if there are others who would like to try to outbid either party. The stock exchange was concerned about maintaining transparency of market activities. To this end, it supported a series of information dissemination activities to distribute information on a timely and accurate basis. The Bolsa published a series of informative brochures on various aspects of investing in the market and on industries and firms represented in the market, including their Revista Tendencias Bursátiles (Market Trends Magazine). They issued a daily Boletín Bursátil, listing opening, closing prices, transaction volumes and so forth, along with trimester and annual reports. This information was the basis for postings in daily newspapers. As an integral part of their system to assure market transparency, the Bolsa provided a system of computers, microcomputer terminals (in brokers and Bolsa administrative offices), and television monitors which provide up-to-the-second information for investment decisions, drawing on the Bolsa’s central databases. This terminal network provided information on price movements, price indices, income, historical and trend information on individual stocks, financial records, international price indices, interest and exchange rates, futures contracts, news, and other important information.

Ten Lessons That Internet Auction Markets Can Learn 109

Information from the Bolsa’s central databases is communicated throughout the Bolsa and to banks, brokers, and governmental officials through a secure PBX (private branch exchange, a proprietary telephone system). In addition, the Fono Bolsa and Data Bolsa systems make price records from the Bolsa available through commercial telephone lines (through dial-up to a 700 number, which is the equivalent of a U.S. 900 number). Through this service, the Chilean telephone exchanges allowed the general populace access to the Bolsa’s videotex services on televisions at home or the workplace, provided an automated voice query service for specific securities, and provided a “what if” capability through either voice or videotex to compute the value of an investor’s specific portfolio. They also provided their Centro de Informatión Bursátil to handle voice queries on questions which investors could not answer through the other services. Until 1991, electronic trading at the Bolsa Comercio de Santiago was handled on a Wang computer, which was responsible for reporting from and, through intermediaries, posting transactions to the open outcry system. The new system was operational in January 1994, after 18 months of programming and testing by a team of four programmer/ analysts under the direction of Dr. Carlos Lauterbach. Software was tested for six months prior to going online, by running simulations involving two to three days of actual transactions. These simulations compressed or expanded time to assure that the systems processing capacity was adequate to reliably handle worst case scenarios. The Bolsa Electrónica and the Bolsa Comercio provided very similar capabilities to traders. To maintain its image of fairness, the Bolsa Comercio rents out hardware and software services on the electronic exchange, forcing all transactions to be input through keyboards and to be output through computer screens. The exchange mechanism is systematically leveled at the bottom—all clients have the same equipment and software, as well as the same communications line delays. This disallows “program trading” (i.e., the following of stock prices through electronic feeds, with posting of transactions by computer programs, expert systems, or artificial intelligence software which automatically looks for arbitrage or investment opportunities). Communications line delays are also tightly monitored, so that terminals (e.g., in the same building as the Bolsa), are not able to post their transactions more quickly than those further removed. The electronic system captures trading that previously would have been handled through open outcry. By default, the open outcry system is fair—it is essentially selfregulating in this regard (though perhaps not efficient). It is the goal of the electronic exchange to capture that same image of “fairness” through sophisticated sets of trading policy implementations. The goal is to make the exchange systematically fair to all traders and to avoid any features that would make the exchange systematically unfair (i.e., reward some classes of traders at the expense of other classes of traders). For example, in the U.S. issues of fairness have arisen concerning the National Association of Securities Dealers dealerquote driven NASDAQ system, which has traditionally exhibited wider bid-ask spreads than the competing order-driven New York Stock Exchange. The Bolsa maintains a network for broadcast of market transactions and indicators through several services: 1. continuous updates to Reuters news services through dedicated communications lines owned by Reuters; 2. dial-up modem services to personal computers; 3. videotex services to televisions at home or the workplace, provided by the Chilean telephone exchanges;

110 Westland

4. 5.

an automated voice query service for specific securities; and a “what if” capability through either voice or videotex to compute the value of a given portfolio. In both open outcry and electronic trading, the market provides a nexus for all offers and queries brought to the market. This presents the traders with a potential bottleneck. The electronic system relies on extremely fast online transaction processing (OLTP) by a Tandem Cyclone computer. In data servers (i.e., computers dedicated to “serving up” data records for update or reporting) such as the Tandem Cyclone, typically several CPUs (central processing units=the computing part of the computer, of which a desktop PC computer has only one) are required to achieve the desired processing performance. The Bolsa Comercio de Santiago set as a benchmark that 95% of matchable transactions (i.e., where the bid and ask prices cross) could be completed on the Tandem within one second. In order to do this, the Tandem computer needed to be run at 40% of its peak load capabilities (a statistic established by the manufacture Tandem Computers, Inc.). This required a Tandem computer with six CPUs to support 30,000 transactions per day (an average of 10 transactions per second) from 600 dedicated terminals in the offices of agents and brokers. In addition, the system needed to support the broadcast of stockmarket information (essentially a full-screen electronic ticker tape) through dial-up modem connections, and voice announcing provided through the equivalent of U.S. 900 numbers available through the telephone companies. Tandem processing was estimated to be linearly scalable to about 300% of this performance, at a maximum configuration of 16 CPUs. In traditional markets (including the open outcry market at the Bolsa Comercio) the matching bottleneck is handled by breaking out trading by specific assets—i.e., certain stocks are traded only in one pit. This was tried in the electronic exchange. It failed to yield improvements in matching speed due to time clustering of trades in any given security or industry during the day, usually around the time of release of critical information such as financial reports. 1992 transaction growth of 20 to 30% annually threatened to outstrip the Tandem’s processing capacity in around five years. Yet they knew that the New York Stock Exchange, using Tandem computers, was able to maintain similar performance standards, applying a more complex trading policy, for transaction volumes in the 1 million to 10 million transactions per day—30 to 350 times the current volume on the Bolsa Comercio. One of the most critical decisions in systems development for software spanning several machines is the scheme for splitting the workload between computers. The Bolsa had chosen to dedicate matching activities to the Tandem, broadcast activities (to Reuters, the Fono Bolsa, and other news reporting services) to the Sun, and transaction acquisition and real-time reporting to the PCs located in brokers’ offices. Several factors are important in the Bolsa Comercio system: 1. The constrained machine resource is the Tandem Computer, which is the nexus for transaction matching. The Tandem should not be responsible for broadcasting information on trades, volumes, and so forth. Broadcasts are required by the dial-up services offered over the telephone lines, as well as continuous updates to Reuters and other news services. 2. The constrained resource for systems development are the programmers, who are skilled in COBOL programming but know little of event triggering (e.g., posting an offer), or of software development for Intel PC client terminals. 3. The performance and reliability of the Intel PC client terminals may be uncertain, since they are in users’ hands. Thus their responsibilities need to be limited to reporting status of offers and trades and to input of transactions (offers or queries).

Ten Lessons That Internet Auction Markets Can Learn 111

4.

Cost issues were important. The six CPU Tandem cost US$1 million, depreciable over four years (mainly technological obsolescence). The client terminals were i286 or i386 PCs, costing around $1,000 per terminal. These constraints suggest that broadcast of information—such as required by Reuters or the telephone voice and data broadcasts of market information—can be handled by neither the Tandem nor the terminals. For this reason, an intermediate tier of Sun workstations ($5,000 per terminal) were installed to handle two tasks: (1) format updates to the market for broadcast to the telephone companies and to news services such as Reuters, and (2) reliably format transactions coming from the client terminals for forwarding to the Tandem. The Sun workstations were wired to both the Tandem and to the client terminals in pairs to assure that if one Sun went down, another “hot standby” was ready to complete the transaction in less than one second (i.e., in “real time”). Each Sun workstation could handle transactions from around 100 terminals. All of this technology was employed in the achievement of three computational performance targets: 1. 95% of matching transactions (on Tandem) completed in less than 1 second 2. Sun batching and reporting introduced 200 milliseconds (.2 second) delay 3. No PC software module greater than 10Kbytes in size (which translates into less than 3 seconds download to update a PC terminal’s software, done at time of logon) The Tandem is mainly used as a data server, though some of the server programs (coded in COBOL) provide information search and formatting services. Since reliability, accuracy, security and integrity are paramount to this function, the Tandem implements software and data mirroring. Software and data mirroring are services that Tandem is famous for and one of the reasons that they are the primary supplier of computers for stock exchanges and other mission critical applications where any downtime is unacceptable. The majority of software and processing in the client terminals supports the graphical user interface, running under Windows. This is programmed in the ECU language, which is compiled (i.e., translated) into a very compact intermediate language (which can be quickly downloaded across 9600 baud communications lines if needed), and is interpreted (i.e., run) by the kernel. The Sun workstations contain no custom code. They use packaged software for information broadcasting, for managing the client terminals, for performing transactions for posting to the electronic exchange (on the Tandem), and for managing the multiple sessions running simultaneously. By not using custom code on the Sun workstations, and by performing only rudimentary services, well suited to a Tandem server on the Tandem, the job of updates, maintenance, extensions, and revisions to the system are kept mainly on the client terminals. And the client code is loaded to the Tandem, where it is automatically downloaded to each client whenever it logs on. The Sun workstations: 1. display amounts on the bulletin board in the sala de ruedas used for displaying transactions; 2. handle routing of transactions; storing and forwarding of transactions for integrity; 3. randomize the order of transactions in the queue to be forwarded to the Tandem for matching, and randomize the order of transactions in the similar queue to be broadcast to traders. This function is provided by the stock exchange to insure fairness in trading, and to thwart attempts at front-running. The Sun workstations add a 200 millisecond-per-transaction delay (i.e., 20% of total one second target time for processing a transaction) due to storing and forwarding, but this is

112 Westland

considered worthwhile, since during peak loading, the Tandem is freed for search and matching functions.

TEN LESSONS THAT INTERNET AUCTION MARKETS CAN LEARN In all three cases of stock market automation presented in this paper, the move towards automation and linkage into a wide-area order gathering and information dissemination network improved liquidity, extending the reach and participation in the market. Each of the exchanges used information technology to rewrite its trading geography, making off-floor transactions more appealing than on-floor transactions. Yet none of the moves off-floor was without costs and problems. These provide valuable insights for e-Commerce’s automation of retailing. This section summarizes ten lessons from securities exchange automation, with examples from Santiago, Moscow, and Shanghai stock markets.

ELECTRONIC AUCTION MARKET ADVANTAGES (LESSONS 1 THROUGH 5) Lesson 1: Customers are attracted to electronic auction markets because they provide greater liquidity than traditional markets; ceteris paribus this greater liquidity results directly from greater geographical reach provided to commercial transactions by electronic networks. A liquid market is one in which traders can quickly satisfy their demand to buy or sell specific commodities. Traders (i.e., the customers of securities markets) fall into two broad groups: those who participate in order to buy the product (usually called value-motivated traders) and those who hope to resell the product in the same market, taking advantage of supply-demand imbalances (usually called speculators). Speculators can be valuable liquidity providers but can also undermine efficiency and orderly trading by distorting the reflection of underlying value of a product in the market. The importance of a liquid market was not lost on Shanghai Stock Exchange officials when they automated their exchange. In their case, raising capital was of secondary importance, since the Chinese save 40% of their earnings, which was the basis for much of the lending to state enterprises. Rather, they wanted their exchange to attract foreign capital to these state enterprises, while assuring that any savings transferred to investments from the Bank of China (the major lender to the state sector) were still invested in the state sector. They accomplished this by listing State-Owned Enterprises (SOE) on the exchange, and insuring that trading was easy, attractive, and accessible to all Chinese people. This insured liquidity, by attracting a large base of active traders in listed securities by applications of computer and communications technologies appropriate to China’s populace. In contrast, the Moscow Central Stock Exchange kept trading in the brokerage houses as a result of its inability to provide network connections to the people. Further unsatisfactory outcomes may be attributed to this failure to provide widespread “geographical reach.” The problem was even more pronounced in the Russian voucher market, which offered no automation whatsoever.

Ten Lessons That Internet Auction Markets Can Learn 113

Liquidity becomes important in product markets with large trading volumes (e.g., for raw materials and intermediate production factors). These products happen to currently provide the most lucrative electronic commerce markets. The liquidity required in any particular market depends heavily on the lead time required for its use. In financial markets, this can be extremely short. In tangible goods, it may be much longer (e.g., concrete and steel contracted for a highway which will be constructed over the next decade need not be traded more than once every few months for buyers to consider the market sufficiently liquid). It is important to note that the traders who can potentially offer the most liquidity to a given market are not necessarily those traders who are either able to or are willing to be continuously present in a given market. This last premise is particularly important in assessing the impact of automation on markets, since advances in computer and telecommunications technology have obviated the need for physical presence at a trading location. This latter aspect of automation, along with the possibility of maintaining a quasi-continuous presence using the vehicles of automated (intelligent) agent software to continuously monitor and transact in an automated market, is perhaps the greatest source of impact of automated markets. Lesson 2: Electronic auction markets can more efficiently discover the best price at which to trade in a product. Buyers and sellers prefer—and will choose to trade in—a market that insures that the best price is discovered which can maximize buyers’ and sellers’ preferences. One reason to trade in an auction market, rather than simply search for a product in a retail market, is the potential to discard the “price-taking” assumption of the retail market. This forces the seller to give up the producer surplus—the amount received in excess of a particular buyer’s valuation of the good—to the advantage of the buyer. Such a market will attract more sellers, because it can ensure them a better deal. It will attract more buyers, because of the greater liquidity arising from there being more buyers. Best price discovery should be improved with increasing liquidity. The greater the liquidity of trading, the lower will be the producer and consumer surplus, because trading can potentially occur at more points along the supply and demand curves. All three cases show how modern computer and communications technologies can expand the number of customers’ (traders’) participation in the market, by greatly reducing the need for customers to be at a specific location at a specific time to trade. Where such limitations on auction location and time were imposed (as in the case of Russian industry vouchers), customers were greatly disadvantaged. Lesson 3: Electronic auction markets can, at low cost, provide exceptional levels of transparency of both market operations and of products quality. Transparency of a market implies that there exists some public source of information about products traded, and the activities of the market itself. This information must be current, must address product quality, and must be credible. In financial markets, audited financial statements and continuous news reporting provide a reliable source of information. Transparency of trading is another way of describing the demand functions of participants in the market. Three broad levels of transparency of trading on any single security are possible (with finer distinctions within each one of these levels): 1. Only the quotes of market makers may be broadcast to the trading public at large—This corresponds to the closing prices posted in newspapers;

114 Westland

2.

The price and volume on the last trade may be posted—This corresponds to the ticker information broadcast throughout the day; or 3. The price and volume on each order can be posted—This corresponds to a market maker’s or broker’s order book. Of the three markets described in this article, only the Santiago Exchange provides transparency of both fundamental business information in the firms whose securities are traded and market trading. All three exchanges provide trading transparency at level (1); the Santiago and Shanghai exchanges provide transparency at level (2) as well. It is rare for brokers or market makers to reveal their order books, though. Moscow’s stock market, because it processed only about 20% of trades in listed securities, lost the confidence of investors in part because market prices did not adequately reflect supply and demand. Lesson 4: Electronic auction markets are more efficient than traditional markets. This efficiency allows them to better provide information required to correctly price assets traded in the market. Efficiently is related to transparency in the sense that it has to do with the dissemination of information. Efficient market prices will accurately reflect underlying value, supply, and demand for a product. In efficient markets, the price will be quickly influenced by new information. The Moscow Exchange again provides an example of problems arising from failure to rapidly incorporate new information into prices. In the Moscow Exchange’s case, this arises because the majority of trades (around 80%) take place off the floor at transaction prices which are not posted. As a result, posted market prices tend to be ignored by traders. They also tend to be lower than the estimated book value of the shares. This illiquidity is reflected in bid/ ask spreads that are often an order of magnitude apart, whereas in most exchanges, it is a fraction of the smallest currency unit. Lesson 5: Electronic auctions can provide a market that, ceteris paribus, offers services at a lower transaction cost. Markets provide a specific bundle of services. Where consumers have a choice of two otherwise equivalent bundles, they will do business where transaction costs are lowest. Traditional securities markets did not need to be significantly concerned about transaction costs because an exchange’s dominant position in trading a listed security, the tendency of liquidity to draw in more transactions, and regional loyalties combined to limit the choice of venue offered traders. With the drop in price of hardware and software for creation of an exchange, more and more markets can enter the competition for order flow in the same securities, competing on convenience and transaction cost. Thus Bernard Madoff began an electronic exchange in securities over a decade ago that now does over 10% of the volume of the New York Stock Exchange on listed securities. More recently, E*Trade has opened its own exchange, competing for similar stocks but offering its own online order capture. The situation highlighted in the Bolsa Comercio de Santiago case involves the challenge to its dominance over Chilean market trades by the Bolsa Electrónica, a completely electronic exchange started three years earlier. The Bolsa Electrónica used a computer trading system very similar to the electronic system developed for the Bolsa Comercio, and in just three years it had captured 30% of transactions in securities of listed companies (that figure had risen to nearly 60% by the late 1990’s). The Bolsa Electrónica

Ten Lessons That Internet Auction Markets Can Learn 115

was cannibalizing transaction volume directly from the Bolsa Comercio and was able to process transactions less expensively, because it was able to do away with the trading floor and its associated maintenance and rental expense, displays, labor, and so forth. Not being able to compete on cost, the Bosa Comercio chose to improve service in order to attract business. The case focuses on the trade-offs between policy (e.g., fairness and transparency) of the market and the technology offerings required to make the market attractive (e.g., Ethernet communications lines and program trading). Its operations emphasized fault tolerance and online transaction processing and effective information dissemination systems to provide real-time transaction and summary data.

ELECTRONIC AUCTION MARKET CAVEATS (LESSONS 6 THROUGH 10) Lesson 6: Customers will abandon a market that is not perceived as fair, even though they may initially profit from “unfair” transactions in that market. By distancing customers from the traders in a market, they can provide a false sense of legitimacy to a market that allows unfair and opaque trading practices. Trading is fair if one trader cannot systematically profit from another trader. Fairness is related to information efficiency of a market in that it requires that information about the security, as well as information about offers, sales, and the trading environment, be available to all traders in the market. Even so, fairness may be difficult to define in general. Most traders assume that they can outperform the market and thus assume that they can systematically profit from less informed (and presumably less clever ) traders. Markets need to insure fairness in order to attract trading (otherwise the disadvantaged traders would not trade in the market, leaving the remaining traders without traders from which to systematically profit). The resulting loss of liquidity is likely to drive out even traders who have initially benefited from “unfair” transactions. Grossman and Stiglitz (1982) showed that, in fact, markets could not be efficient and fair in the sense that all traders are equally informed (otherwise trading would stop). What is needed is the perception of fairness, in that any differences in success in market trading are presumed to be due to trader risk preferences, luck, and competing philosophies about what drives performance in a particular industry or stock. Markets need to adapt them to the business environment around them, and fairness must be interpreted in a local context. Russia’s markets provide an example of the difficulties in defining fairness in any universal sense. Broker abuses, insider trading, and market failures are common in Russia. Some abuses (e.g., broadcasting prices via the exchange system but completing the transactions at a different price in a private transaction) could be avoided by greater integration of the brokerage community into the market making system. But some problems arise from the inefficiency and illiquidity of the market. Brokers may receive little information on firm assets or operations in Russia, and thus are at greater risk of misrepresenting the investment potential of a product. Illiquid markets make it difficult to assure that the next transaction takes place at a price close to the prior transaction, or to know whether the difference is due to manipulation. The Russian markets provide a telling example that your markets must monitor, enforce, and insure fairness in trading Lesson 7: Electronic auction systems must manage all aspects of trading activity, from

116 Westland

initiation to settlement and delivery. Markets that fail to integrate both price discovery and order completion (settlement) into their operations can encourage unfair trading behavior and opaque trading practices. This lesson is drawn from abuses at the Moscow Exchange where credibility of prices suffers significantly from failure to accurately record traded prices. In the Moscow Exchange’s case, this arises because the majority of trades (around 80%) take place off the floor at transaction prices which are not posted. As a result, posted market prices tend to be ignored by traders. They also tend to be lower than the estimated book value of the shares. Bid/ask spreads (the difference between the price at which a share can be sold on the exchange and that at which it can be purchased) may be an order of magnitude, whereas in most exchanges, it is a fraction of the smallest currency unit. Market prices may be set through two classes of mechanisms—(1) dealer quotes, and (2) price matching of bid and ask offers. The former are called quote-driven markets, and the latter order-driven markets. In an order-driven market, the stated price that a market sells or buys a particular commodity is typically the price at which the last trade took place. In a quotedriven market, quotations simply indicate that a particular market maker is willing to trade at specific bid or ask prices. Quote driven markets are similar to retail stores in that traders coming to the market must take the prices quoted. Retail stores typically serve only buyers, whereas quote-driven markets serve both buyers and sellers. Lesson 8: Because the delay in price response may have significantly faster completion and posting times, there is greater potential for feedback loops and instabilities that are a threat to orderly trading, and to fair and efficient pricing of assets traded in the market. Traders derive considerable information from price movements in markets. Thus, prices should not be overly influenced by imbalances in supply and demand (prevention of shortages and price fluctuations is one of the main reasons for using a market to trade). The last price at which a security was traded is close to the price at which they can trade. Where trading is not continuous, e.g., when a stock exchange closes at the end of the day, orderly trading may be insured by artificially setting the opening price of the stock to clear the outstanding buy and sell orders. All three of the exchanges in this article set opening prices to clear outstanding orders. In the case of the Shanghai exchange, this may often be considerably different from the last trade price of the prior day. Market operations must assure that prices do not vary greatly unless this truly reflects an ongoing trend or change in fundamental value. Specialized market-making functions and trading rules such as circuit breakers (which stop trading if an excessive drop in prices is detected) promote orderly trading on the underlying value of securities. Lesson 9: Electronic auctions may foment unfair trading practices through different relative speed of service through different parts of its network linking trading to customers. The issue of speed is a subtle one but is intrinsically tied to fairness and transparency. It appears in three places in trading: (1) order placement, (2) information dissemination, and (3) matching. Absolute speed will change continually, as new technologies are introduced. It is important that the speed offered by a market be competitive with competing markets on these three dimensions. An example of the importance of speed in order placement is provided in software of the Commercial Stock Exchange of Santiago when it was installed in Cali and Bogata, Columbia. To insure fairness of trading, the Bogata communications line actually had to be

Ten Lessons That Internet Auction Markets Can Learn 117

slowed down so that Cali traders couldn’t trade in advance of Bogata traders on the same information. A similar situation arose inside and outside the exchange building in Santiago, where the local network access was considerably faster than city-wide access. Santiago’s market was sensitive to speed of information dissemination as well. The market operators recognized this and provided several innovative ways of accessing realtime quote information, including a telephone dial-up system. At the opposite end of the spectrum was the Moscow exchange, where many stocks traded once or twice a week. In this case, daily newspapers were more than adequate for reporting. Matching proved to be the major bottleneck in the Bolsa transaction processing. With the increase in velocity and sophistication of trading—particularly the technology for program trading—comes increasing sensitivity to matching delays. Program trading and exceptional variability of volumes could significantly degrade matching speed at peak times, thus degrading market performance on other parameters, such as best price discover and orderly trading. Lesson 10: Order-driven electronic auction markets demand that the market clearly define when a sale has been made. In an order-driven market, customers place buy and sell orders in the market, in hopes of their being matched with other customers’ orders at the best price. At the point the orders enter the market, the traders lose control over the price negotiation process. Naturally, they will be concerned with the system that the market uses to complete their transactions. The algorithms used by the marketplace need to be clearly defined to the market’s customers in order to attract trading. In order-driven securities markets, order management can become very complex. In a survey of 50 electronic trading systems in 16 countries, Domowitz [1992] found 11 trade priority rules (e.g., best price has highest priority, time priority is first-in first-out); 7 levels of price discovery rules for information search and matching (e.g., manual exposure of orders to a market maker for price improvement; anonymous, direct negotiation); and 12 classes of information which define the transparency, type, and amount of information which is disseminated to traders (e.g., high and low price, best bid and offer, and quantities available at these prices). Markets can give customers some choice over the determination of the time of sale (and thus sale price, quantity, and so forth) by allowing a variety of orders (limit, stop limit, market, margin, and so forth).

ENDNOTES 1 2 3 4

Asiaweek, 28 February 1997, p. 62. Economist, 14 June 1997, p. 124. Economist. Capitals of Capital, Survey–Economist, 9 May, 1998 Delays are not just introduced by the length of wire (or fiber), which given the speed of light would be negligible. Rather, switching delays and error correction (which may require resending information and transmitting signals in two directions for confirmation) introduce the most significant delays. Indeed, the modified version of Bolsa Comercio’s electronic trading system that was sold to the Cali exchange in Columbia had to support a large number of transactions from both Cali and from Bogota 200 miles to the northeast. To accommodate this, signals at Cali were artificially delayed to allow the Bogota signals to “catch up,” insuring fairness in trading and diminishing the possibility of “front-running.”

118 Westland 5 6

“Linearly scalable” means, roughly, that if the Bolsa buys three Tandem CPUs, it will have a machine that processes roughly three times as many transactions in the same time interval. “Mirroring” means that copies of the same data or program code are maintained in two places, managed by the same system. This is necessary for mission critical systems, where information services cannot afford to be lost, as in the case of market transactions. It is a critical feature in providing “fault-tolerant” systems—i.e., systems which can recover quickly from faults, whether they be system or user induced.

REFERENCES Asiaweek. (1997). 28 February 1997, 62. Domowitz. (1992). Liquidity and Trading Rules. Northwestern University Working Paper. Economist. (1997). 14 June, 124. Economist. (1998). Capitals of capital, survey, 9 May. Economist. (1998). The cash don’t work, 19 December, 120-112. EDI Forum: Voucher Privatization Funds in the Russian Federation. http://www.tomco.net/ ~edinp/fsu/funds.html. Euromoney. (1994) Bulls go wild in a China shop, October, 56. Euromoney. (1996). Varying fortunes of China chips, March, 15. Euromoney, (1996). Varying fortunes of China chips, March. Federal Commission for the Securities Market. (1997). Report at the all Russian conference of professional capital markets participants, October 1996; and FSC to create regional self-regulatory network. Moscow Times, 26 July, 14. Frydman, R., Pistor, K. and Rapaczynski, A. (1995). Investing in insider-dominated firms: A study of voucher privatization funds in Russia. Oesterreichische Nationalbank Working Paper 21. Grossman, S. and Stiglitz, J. (1982). On the impossibility of informationally efficient markets. American Economic Review, 72(4). Harris, L. E. (1991). Trading rules, and electronic trading systems. Monograph Series in Finance and Economics. New York University Salomon Center. Kommersant. (1994). Better to buy gazprom, (10), 22 March, 32-34. Kommersant. (1994). Duma debates on privatization, (15), 26 April, 37-38. Kommersant. (1994). Voucher privatization is not likely to impact the gas industry, 3(1), 33-35. Kommersant. (1994). Privatization of voucher investment funds in reflection: Most important is not victory, but participation, (5), 15 February, 43-51. Moscow Times. (1997). VegaTech seeks market serving market-makers, 17 June, III. Moscow Times. (1997). VegaTech seeks market serving market-makers, 17 June, III. Potter, P. B. (1992). Securities markets opening to foreign participation. East Asian Executive Reports, April, 7-9. Rinaco Plus Brokerage House Web Site. (1999). Equity Index Methodology. http:// feast.fe.msk.ru/infomarket/rinacoplus/. Shama, A. (1997). Notes from underground as Russia’s economy booms. Asian Wall Street Journal, 30 December. National income statistics are reported in Asiaweek, 16 January 1998, 51.

Ten Lessons That Internet Auction Markets Can Learn 119

Thomas, W. A. (1993). Emerging securities markets: The case of China. Journal of Asian Business, 9(4), 90-109 Whitehouse, M. (1998). Shortchanged on the stock exchange. Russia Review, 8 May. Yergin, D. and Stanislav, J. (1998). The Commanding Heights, 282. New York: Random House

120 Mantelaers & van den Berg

Chapter IX

Transnational Information Systems: Development and Management Issues Paul Mantelaers and Wander van den Berg Delft University of Technology, The Netherlands

An increasing number of transnational information systems (TIS) is being developed and is operational. This paper describes a study that attempted to gather information from an empirical setting as a basis for theory building. The authors have studied three cases in an attempt to take some early steps towards guidelines for the development and management of TIS. General conclusions based on the analysis of the three cases are drawn.

INTRODUCTION The Problem Transnational information systems (TIS) are systems that cross national as well as company borders. Development and management of TIS are not straightforward. The interorganizational and international nature of these systems may introduce various organizational and technical problems. One major reason for this may be that the participants are independent in most aspects and autonomous with regard to their own proprietary information systems. Another potential cause of difficulties is the international context, which introduces cultural, legal, and language difficulties among participants. However, the globalization of the economy gives rise to an increasing demand for information systems that transcend national boundaries. Developments in information and communication technology now make it possible to support many processes and tasks that cross company and national boundaries. Alternatively, economic and political developments create demand for TIS. This demand is clearly evident within the European Union in the context of a European market for persons, goods, capital, and services. Similar developments can be seen on a global scale due to increased internationalization. Appeared in Journal of Global Information Management, vol. 8, no. 1, 2000. Reprinted by permission.

Transnational Information Systems

121

As to their international and interorganizational dimension, most individual TIS are being developed up to now on an ad hoc basis. Currently there are no known general techniques, tools, or guidelines to help develop and manage TIS. Practical experience and empirical research, however, show that many problems and solutions at first glance appear to be specific to a particular development effort, but often they are not unique at all. As the needs and opportunities for TIS become more prevalent, the call for generalized knowledge and guidelines increases. Organizations want to know how to avoid difficulties and, if problems should occur, how to address them. Organizations that consider to use, develop, manage, or participate in TIS may benefit from the availability of structured “dos and don’ts.”

The Literature Studies addressing the combined transorganizational and transnational IS setting are lacking. There is, however, a stream of literature concerning the development of information systems linking parts of multinationals located in different countries (Deans and Kane, 1992; Palvia and Palvia, 1994). Apart from research about global systems, there is also a substantial body of writing about systems crossing company boundaries. Much of the early discussion about interorganizational systems (IOS) focused on competitive benefits of developing IOS (Porter and Millar, 1985) and on facilitating and inhibiting factors in the development process (Reich and Benbasat, 1990). More recently research has taken on a more realistic approach, pointing out that not all IOS provide benefits to all participants (Webster, 1995), that IOS may involve risks and conflict (Kumar and van Dissel, 1996), and that IOS involve management of relationships among participants (Meier, 1995). The general impression is that findings from literature on IOS and global systems are relevant for TIS. The combination of systems crossing organizational as well as national boundaries has hardly been mentioned in the literature. The use of IT in TIS is a fairly recent phenomenon, and this may explain the dearth of research in this area.

Outline of the Paper This paper describes a multidisciplinary case study as three cases regarding TIS development and management. It is an attempt to take some early steps towards guidelines for the development and management of TIS. In view of the fact that very little existing research is available about TIS, this study focused on gathering information from an empirical setting as a basis for theory building. The paper proceeds as follows. First, the research approach is lined out and the three cases are discussed briefly. Then, the main TIS issues and other issues that emerged during the research are summarized. In the last paragraph, a few concluding remarks are added.

RESEARCH APPROACH Within the public sector in the European Union alone, about 30 TIS are operational or under development (Kroon, 1997). Experience and knowledge based on these systems tend to remain within the organizations involved unless an attempt is made to collate and generalize the lessons learned by individual organizations. For this reason it was decided to conduct an empirical study addressing a wide range of TIS issues in order to present this accumulated experience and knowledge to other interested parties, as well, using inductive case research methods and employing a multidisciplinary research team.

122 Mantelaers & van den Berg

Issues to Be Addressed in an Empirical Study One way of studying TIS issues is to distinguish issues in the informational, organizational, and institutional domains. Parsons (1960) first discussed these distinct but complementary areas of responsibility and control; he argues that they are three quite separate domains within an organization. We adopted this concept to categorize difficulties and potential solutions during the process of development and management of TIS. Parsons used the term “technical domain” to denote the technical suborganization; we have changed the term to “informational domain” to underline the fact that this area of attention concerns data, procedures, and people as well as the technical aspects of hardware, software, and communications. Organizational issues refer to TIS-induced adaptations of organizational structures, business processes, regulations, skills, standards, and ways of working. Institutional issues concern policy identification, political or competitive power struggles, role models, as well as the formulation and development of decision-making arrangements among participating organizations to ensure that the new order is accepted, is adhered to, and continues to operate effectively. A second dimension for studying TIS issues concerns the various phases of the systems development process—from identification of the need for a TIS through to the management of an operational TIS. Various authors (Turner, 1988; Martin, 1989) provide different names for separate development phases, divide the system development process into different numbers of phases, and describe the development process with or without iteration and feedback loops. However, authors generally agree that (1) the need for a system should be identified and requirements drawn up; (2) logical and physical design and building of the system has to take place; (3) the system has to be implemented; and (4) the operational system has to be managed, updated, and adapted over time. We used these four phases to categorize difficulties according to stages of TIS development. A third way of looking at the TIS area is to determine whether TIS issues lie in the interorganizational domain or the international one. A question was whether the same or similar difficulties would arise if the system would be developed on a national basis (inside a particular country), that is, without any complications as results from the international setting of the relevant information systems. It was important to determine whether TIS problems are caused primarily by the international dimension or by the fact that TIS require cooperation among and coordination of multiple autonomous organizations.

Case Studies As Research Method Case research was selected as the appropriate research method. Carrying out a number of case studies would provide rich data, which is important in the process of theory building (Yin, 1994). Observation of phenomena and relationships in a variety of settings forms the basis for conceptualization and generalization. Hence, studying the development processes of existing TIS—being guided by existing constructs, keeping an open mind to uncovering new relationships, and subsequently conceptualizing from finding—is a sound method for formulating guidelines for TIS development and management. Selection of cases was based on several criteria. Cases had to concern TIS but had to be sufficiently dissimilar to provide a rich picture of the problem area. Hence, cases were included from the private sector (TAPS), the public sector (TRANSIT) as well as a mixed sector case (EUCARIS). The focus in these three cases was on computerized transnational systems. TIS in the cases had to support a primary process in the organizations and cases had to involve systems that were operational even if only in an early version. Cases need

Transnational Information Systems

123

not be highly successful; in fact we preferred to have cases which had encountered or were facing difficulties. An additional criterion was the requirement that organizations had to be willing and able to participate in the study. The researchers had no relation with the developing organizations.

A Multidisciplinary Research Team Since this was an exploratory study, it was decided to use a multidisciplinary research team for carrying out the study. Using researchers with different backgrounds would strengthen the study by having a wider range of expertise in the team. In that way we hoped to avoid a onetracked study which would report on certain aspects in detail but would overlook other important points. The team that was brought together consisted of researchers from three different departments spread over two universities; backgrounds of the researchers were computer science, information systems, and management. The multidisciplinary nature of the project created its own challenges. It transpired that the different disciplines which were meant to strengthen the research also presented difficulties: each discipline had its own frame of reference and definitions, even for simple concepts like “information” and “system.” Similar difficulties with a multidisciplinary research team have been reported by Kaplan and Duchon (1988). Much time was spent discussing seemingly irrelevant details, but it was important to obtain a common sense of the concepts we wanted to study. After much discussion a list of “notions and terms” was drawn up, and this helped the team considerably. The participation of multiple researchers necessitated the writing of a detailed general proposal, a list of questions to be addressed in every case study, and a general framework for single-case analysis. For each case the available system’s documentation had to be studied carefully and indepth interviews were scheduled with various stakeholders. It was important that the different researchers would ask similar questions, address similar issues in the various cases, and analyze their data in a similar way (Miles and Huberman, 1984). Meticulous preparation and frequent meetings were necessary to enable analysis of data across the cases.

THE THREE CASES As mentioned, three (EUCARIS, TRANSIT and TAPS) cases were studied by the research team. In this section the three cases will be described in detail. They also will be compared along several dimensions.

Transit1 Several times, a TIS was under development to speed up and facilitate custom administration concerning goods that are imported into the European Union; such a system should also lead to more effective fraud detection and prevention. First, the transit procedure is lined out. Then the problems that triggered the systems development project are summarized. The New Computerised Transit System (NCTS) will be described, as well as the major specific problems that occurred during TIS development.

The Transit Procedure Goods from outside the EU that enter the EU must go through customs. Taxes and duties have to be paid and foreign trade requirements have to be complied with before these goods

124 Mantelaers & van den Berg

can enter the market. However, it is possible to apply a suspense regime to the goods. The community’s common transit arrangement is such a customs suspense regime. It facilitates the movement of goods within the European Union and EFTA countries without the relevant duty and tax having been paid. Since payment is not necessarily required at the point of entry, European customs officials must keep track of goods moving within Europe and, in particular, must be able to trace what duties and how much taxes have been paid. When goods from outside the EU are transported within the EU, a Single Administrative Document (SAD) has to be completed by a principal at the departure office. This SAD contains all kinds of numerical and textual data regarding the imported goods. The departure office is responsible both for departure formalities and for transit operation clearance. By law and regulations, the SAD consists of several copies. One remains with the departure office, and three copies go together with the goods to the office of destination. One of these three is for the destination office, the second is for the national bureau of statistics, and the third has to be sent back to the departure office through a central office. After that copy is received, clearance will take place for each article.

Triggers for Systems Development The main problems that are encountered by those involved in transit operations are the following: • The volume of transit operations handled by customs authorities is creating problems for them. Substantial resources are needed to handle around 17 million declarations in Europe each year and to process about one million inquiries within the context of search procedures. • Traders have to spend quite some time to produce and administer the various forms. • There are frequent problems as to the data quality of the documents supplied to customs and the language barrier between EU states makes problems worse. • The traditional mail system is used by the destination office to send the SAD copy to the departure office. This takes quite long, which makes it impossible to react adequately when there is something wrong. • The clearance at the departure office is done by hand; this is very labor intensive. • The office of destination does not know what goods to expect; they are not very well prepared to inspect the goods upon arrival. • The whole procedure is very fraud sensitive. Fraud has increased due to the disappearance of the borders between countries within the EU and due to the disappearance of the iron curtain (more trade with East European countries). • Guarantee certificates are difficult to check between countries. There is no overview of the total amount for which a principal is held responsible.

Towards a New Computerised Transit System Building a TIS to computerise this international customs process was first considered by the European customs organizations in the early 1980’s. A properly working TIS would facilitate tracking and tracing of goods; it would enable immediate comparison of goods and duties paid, speeding up the clearance process; it would produce readable, accurate, and upto-date information at any customs office. The private sector would also benefit because of faster clearance procedures. Disagreements among different EU partners kept the TIS from moving beyond the “nice idea” phase.

Transnational Information Systems

125

The current EU project to build a TIS, called TRANSIT, is the third attempt at developing computerised procedures. This time a serious effort has been put into TIS development: fraud problems have reached alarming proportions with fraud concerning goods in transit within the EU adding up to billions of U.S. dollars. Computerization of the transit procedures has now been placed high on the list of priorities of the EU. EDI messages have been designed, and national gateways are being set up. To account for diversity in local systems and to respect autonomy of individual customs agencies, no attempt has been made to integrate the various systems. Integration would have increased the complexity of the development process. Instead, for each country, one customs application host is built, which connects the central TIS system with local systems in various countries. In this way local systems and procedures need not be adapted to a single European-wide system; similarly, the diversity in local systems and procedures needs not affect the central TIS. At the end of 1997, the TRANSIT project was in its pilot phase; customs authorities aimed to have the system fully operational as soon as possible. By developing NCTS, the EU is trying to make the transit procedures more effective and efficient and to make provisions for more effective fraud detection and prevention by the authorities. A third objective is to provide traders with more rapid facilities for carrying out transit transactions.

Development Problems The present attempt to develop a system is the third one. According to the interviewees, the two previous attempts failed due to technical problems and problems regarding the autonomy of the participating countries. Due to technical limitations, IT professionals proposed to create one central database in Brussels containing all relevant data. The participating countries did not accept this proposal. It turned out to be very difficult to reach consensus between the countries about alternative architectures. Other reasons for failures in the past were the lack of political pressure (the amount of fraud was much smaller at that time) and the lack of user involvement in the systems development process. During the most recent development process, several problems had to be overcome. The first one was the difference between countries as to the degree of penetration of IT. Not every local customs office in every country has the necessary equipment at its disposal. In the Netherlands, for instance, all offices are connected through a LAN, but in Belgium, Germany, and France (let alone other EU countries) this is not the case. This problem forces the development team to lower the level of ambition: in the short run, only part of the problem can be solved. Next to that, there was a lot of discussion between countries about the fact that the benefits boil down in other countries than the costs. The investment a country is prepared to make depends on the benefits it expects to gain from it. A third problem was that there are many organizations involved; not only the central and local offices in the countries, but also the traders that play a role and have to be negotiated with. Last but not least are the cultural and language difficulties between countries that make a project, such as the development of NCTS, difficult to manage.

EUCARIS2 The incidence of car theft has increased alarmingly during the past 10 years. Often stolen cars are taken across European borders to be re-registered (and thus whitewashed) in another country. Registration authorities are obliged to register a car unless they can prove

126 Mantelaers & van den Berg

within a short period of time that the car is registered elsewhere and therefore is a stolen car. When in doubt about the legality of a car that is being offered for registration, authorities traditionally write, phone, or fax to foreign registration authorities. This process is inadequate, as it is time consuming and difficult to manage. Computerized access to registration databases in other countries would facilitate the process of checking for legal registrations. It would be much faster, more accurate, and more likely to combat re-registration of stolen cars. The Dutch car registration authority conceived the idea for building a TIS to enable instantaneous, electronic searching and checking of car registration data across Europe; the system was called EUCARIS (European Car Registration Information System). Very early on it became clear that it would not be possible or necessary to integrate the registration systems of various countries into one large system. There is, at present, no single European car numbering system to identify individual cars Europe-wide. Also, current registration systems each use different data definitions, which complicates the possible development of a single system. In addition, individual registration authorities value their autonomy and a central registration system would threaten that autonomy. Hence, it was decided to develop a relatively simple system: a common shell to provide a query interface to the separate systems of participating countries. EUCARIS has been operational since 1995. Currently six countries are online (the Netherlands, Belgium, Luxemburg, UK, Hungary and Poland); 14 other countries are preparing to go online; another 10 have enquired about participation.

Taps 3 “Cashless ambulatory payments” with checks or card-based (plastic) money comprise four systems, for payments and for collecting cash, that will be described in the next paragraphs.

Eurocheque (with identification by signature comparison) Eurocheques were introduced in 1974 as uniform and internationally guaranteed means of payment and gave rise to the establishment of a national clearinghouse and Eurocheque International for cross-border clearing. This policy represents a sound systems engineering approach by dividing Europe into subsystems (countries), or, when needed for historical reasons, into subsystems inside one country. A drawback in the Netherlands is having two different payment circuits, complicating transfers between accounts in these parallel circuits. The acceptor should verify signatures. An advantage is that national borders do not give problems, no electronic equipment is needed at the shops, and operations are cheaper than before. Disadvantages are still relatively high handling costs and fraud risk. Although an improvement at its introduction, the system will be abolished due to cost and fraud risk shortly after the year 2000. New payment systems reduce the use of the Eurocheque now with 25% per year. In 1996 Rabobank alone still processed about 12 million cheques.

The Europas (bank pass card with a magnetic strip and PIN identification) The Europas introduction, halfway into the 1980’s, aimed at lower transaction cost, better fraud prevention, and faster information transfer. These magnetic identification cards, using online identification by PIN, serve a dual purpose: cash dispensing at Automated Teller Machines and cashless payments at Point-Of-Sale terminals for Electronic Funds Transfer (EFT POS).

128 Mantelaers & van den Berg

lower cost (“electronic purse”) and in less time. A striking feature is the possibility of extra nonpayment functions, like loyalty programs. Other features are secure transactions via Internet or phone, due to enhanced fraud control. Safety and security of chip-based card operations (using card authentication, cardholder verification, and encrypting) can be guaranteed by executing all operations under control of the on-card microprocessor. Online PIN is used for larger payments and abroad as a magnetic strip card, because standardization of the chipcards for banking purposes could not yet be achieved across European borders. Interpay was tasked with planning, development, and operations of the Chipknip system. Again a dual system situation in the Netherlands developed. For both payer and payee this is an unsatisfactory situation, even while the terminals for chipcard payments accept both types of cards. No final conclusions can be given yet, because the actual use is still to come. Public acceptance of the system(s) and, as important, acceptance of the system(s) by retailers will bring more clarity. Interesting, as well, are chipcard security and safety issues.

Comparing the Cases In Table 1 the three cases are compared according to several characteristics. Keep in mind that the TAPS-case consists of four cases; that explains the number of columns.

MAIN TIS ISSUES The cases were compared along the three dimensions mentioned earlier: the informational, organizational, and institutional domains; issues according various phases of development; and issues categorized according to the interorganizational and international domain.

Informational, Organizational and Institutional Issues Table 2 presents a summary of the problems faced by the three TIS, categorized according to the informational, organizational and institutional nature of the problem. The table also shows which cases encountered which difficulties. Table 2 shows that development and management of TIS concerns difficulties in all three categories of problems: there are informational as well as organizational and institutional problems. Some of the problems are clearly created by and because of the TIS; other problems exist regardless of the TIS. The TAPS systems stand out as having encountered very few difficulties, especially in the informational and organizational domains. The information intensive and the cooperative nature of the banking sector can explain this. The banking industry is highly experienced in IT use and employs advanced IT systems in all its processes; hence adding new TIS or new functionality to TIS is not likely to be a problem for banks. In addition, payment systems have a tradition of cooperation among banks of different countries, as many banking products and services involve processes which cross company or national boundaries; hence, organizationally there is little resistance or difficulty when adding a TIS service. The other cases, having far less experience with IT and with interorganizational and international cooperation, encountered many more problems. Informational issues need not be permanent bottlenecks. Difficulties, such as lack of sophistication in data collection abilities or lack of IT infrastructure, tend to be dissolved over time as individual organizations or countries develop their systems. Similarly, technological

Transnational Information Systems

129

Table 1: Some key parameters of cases Number of countries Service provider Users

EUCARIS 14

TRANSIT ≅ 20

Eurocheque ≅ 20

Debit card ≅ 50

Credit card ≅ 80

Chip-card Chip: 1 Debit: ≅ 50

public public

private private

private private

private private

private private

private

private

private

private

≅ 100000 but decreasing ≅ 10000 but decreasing ≅ 10000, decreasing

millions

millions world wide millions world wide

millions in the Netherlands @ 10000 but increasing (Netherlands) @ 100000 (Netherlands) Increasing To millions (Netherlands) Netherlands

Beneficiaries

private

Number of users (persons) Number of users (organizations) Number of users (offices) Transactions or messages per day Region Transnational development organization Transnational operational organization

>1000

public public and private public and private >10000

2-5 per country needed for development

1 per country ≅ 500 per country

thousands

hundred thousands Europe European Customs Associations European Union

millions

millions

millions world wide

millions

Europe not available needed

hundred thousands Europe 1 per country, 1 transnat. 1 per country, 1 transnat

World 1 per country, 1 transnat. 1 per country, 1 transnat.

millions world wide World 1 per country, 1 transnat. 1 per country, 1 transnat

2 in Netherlands 2 in Netherlands

Table 2: Informational, organizational and institutional TIS problems

Institutional

Organizational

Informational

Category

Problem Technological limitations Data collection ability Insufficient local infrastructure System integration Diversity of systems to be connected Unequal cost/ Benefit sharing Transational organization Low priority fore some parties Responsibility issue Centralization disagreement Fear for autonomy loss National legal procedures Cross-border differences

EUCARIS

TRANSIT development capacity

Eurocheque

Debit card

Chip-card

number of EFT-POS terminals

ICT infrastructure technical complexity transorganizational and crossborder

Credit card

Number of POS terminals at national level

Between countries Between countries

at national level

separate chipcard architectures

small firms

small firms

needed for deployment ICT inexperience local power issue supranational level

national level

national level

national level

public bodies privacy issues

privacy issues cultural aspects

spoken language

spoken language

130 Mantelaers & van den Berg

limitations tend to be overcome as time passes and new developments become commercially available. Diversity of systems among partners is also overcome by increasing standardization of products. Although technical problems are real and occurred in three of the cases, they are not necessarily long-term difficulties. Next to that, solutions for informational issues are not necessarily only informational but can also be organizational or institutional. This also holds for organizational and institutional issues. Organizational issues are of a different kind. Inequality in costs and benefits of TIS and the lack of transnational bodies to develop and manage TIS are coordination and cooperation issues. Cooperation of individual organizations is necessary for the concept of a meaningful TIS to evolve, and coordination is necessary for the development and management of the actual TIS itself. In order to gain cooperation of individual organizations, it is important to ensure benefits are shared or, at least, costs are borne by those that stand to gain most of the TIS. The absence of transnational organizations inhibits fast or easy TIS development; a transnational body is particularly valuable for coordinating day-to-day TIS affairs. Institutional problems may have been latent prior to TIS development and are highlighted during the discussions about development and management of the TIS. Such problems are not necessarily directly concerned with the TIS and may affect TIS indirectly. Clearly, existing problems have to be addressed if the TIS is to be successful. An example of an existing problem that affects TIS indirectly is “disagreement among various local organizations.” Potential participants of the TIS can have disagreements about data, technical matters, or policy with parties outside the TIS, and such disagreements can affect the ability of the participants to participate fully. Political struggles or business competition between TIS participants or with outsiders are important determining factors for the success of the TIS. A system that is sound from a technical point of view may turn out to be not feasible due to institutional issues.

Issues During the Various Phases of the System Development Every system moves through a number of recognizable phases during the development process; each of the phases may cause, and may have to address, different problems. The problems (and the solutions) that were found in the cases are summarized in Table 3 according to the various system life-cycle phases. System identification. Difficulties in the identification phase appear to be inversely related to the apparent urgency of the problem that is to be addressed by the TIS. When the issue is considered important and urgent, identification of and agreement about the concept of a TIS is reached relatively easily. However, when the problem that TIS addresses is not considered a high priority, difficulties seem to mount up–and the TIS does not get beyond the “nice idea” phase. Designing and building of the TIS. The difficulties at this phase of development are primarily technical in nature. The technical issues are made more complicated because of the great number of participants and their different objectives. The difficulties are addressed by finding a technical solution or by improved organization of the development process. Although the problems in this development phase might seem difficult, finding solutions to the problems did not seem to be a major issue in any of the cases. Implementation. When the TIS has been properly identified and has been agreed upon by participants, implementation is a feasible matter, be it laborious and costly in the TAPS

Transnational Information Systems

131

case. In the TRANSIT case there is large diversity among the participants when it comes to informational and technical sophistication and facilities. This clearly inhibits implementation of the full TIS. Management and ongoing maintenance. Management issues include ensuring that participants observe their local systems so that they interface correctly with the TIS, enforcing standards and continuing use of the TIS, and developing new functionality and increasing the scope of the TIS. EUCARIS, for example, is a TIS that has reached this phase and faces the new challenge of keeping the earlier momentum going. For the TAPS case, this implies further integration inside each bank and at the national and transnational operational body level.

Issues Derived from Interorganizational or International Setting Difficulties encountered during TIS development or management can be caused by the interorganizational nature of the TIS or by the international component of the TIS. Table 4 lists and categorizes TIS problems from Table 2 according to interorganizational or international causes. International context. The usual difficulties of operating across national boundaries were evident in all cases. Although there are international differences and potential difficulties when collaborating internationally, many organizations and the people within them accept the differences and are willing to work towards overcoming such barriers. In the EUCARIS case where legal difficulties inhibited Germany from participating fully, the German car registration authority took steps to ensure the law was adjusted and the EUCARIS

Table 3: Issues during system life cycle phases Life cycle phase Identification

Designing and building

Problems Need for a TIS identified, but not possible to translate into TIS (T RANSIT first and second attempt) Possible TIS identified but too many organizational problems to begin actual development (TRANSIT second attempt) Need for TIS acknowledged, but fear for loss of autonomy (EUCARIS) Need for TIS acknowledged, but not seen as high priority (T RANSIT) TIS will lead to uneven sharing of costs and benefits (T RANSIT) TIS is turning into a technically complex system (T RANSIT ) Diversity of systems that need to be linked (EUCARIS, TAPS)

Implementation Ongoing management and maintenance

Time/manpower for TIS development difficult to find among participants (EUCARIS) Different abilities of participants (in terms of IT) do not allow implementation of full system (TRANSIT) Difficulty of ensuring that standards are enforced and TIS is used properly Time/manpower for coordination (EUCARIS) Long-term integration into existing systems is necessary for long-term success of TIS (EUCARIS, TAPS) Expanding the system to include more tasks/actors (EUCARIS, T APS)

Solutions delay plans for development indefinitely delay indefinitely opt for technical solution that is highly decentralized - delay indefinitely (T RANSIT ); (political) pressure to accept anyway - superior project management - discussion with various parties - participative development - opt for common interface and local gateway configuration - experimental development method - iterative development outsourcing of development - different entry levels - phased implementation (TRANSIT) - third party management of TIS itself; - monitoring committee from T IS general management body outsourcing of day-to-day management of TIS institutionalization of new processes by local organizations is required - discussion - prototyping - consensus building

132 Mantelaers & van den Berg

organization accepted Germany as a partner that would join as soon as legally possible. In the TRANSIT case, documents were translated into various languages to ensure that all partners would have access to documents in their own or in a very familiar language. In some cases there was no transnational organization, but the need for and importance of such an organization was recognized in all cases, and an organization for dealing with transnational issues was established. In all, the international context does present difficulties for TIS development and management, but those difficulties are recognized, can usually be overcome, and are addressed directly. Interorganizational context. Table 4 suggests that there are many more problems due to the interorganizational context than to the international nature of TIS. Some of these are technical and systems issues, which might be overcome fairly easy; other issues concern autonomy and responsibility of organizations and equality among organizations. Interorganizational information exchange requires common decisions, consensus, and commitment. If individual organizations feel their autonomy is threatened or they are not gaining benefits on par with other organizations, then they can withdraw participation or can become a difficult partner in the TIS. In each of the cases, difficulties of this kind were found, and each of these problems proved to be difficult to address.

OTHER ISSUES WHICH EMERGED DURING THE STUDY Apart from issues, which we had anticipated and had specifically looked for in the cases, issues emerged during the course of the study as empirical evidence was collated. A major issue which emerged was the level of ambition of the TIS. Under rational norms we would expect organizations to aim at maximizing the opportunity offered by a TIS, building a system to optimize efficiency and shared benefits. In practice, though, successful TIS organizations show satisfying behavior—adopting a TIS which is acceptable or “good enough.” TIS development involves difficulties, and TIS participants address difficulties during the development stages by lowering the level of aspiration of the TIS, be it in time constraints, in functionality, or in the cost/benefit ratio.

Table 4: Interorganizational vs. international issues Proble m Insufficient local sophistication in terms of IT infrastructure or applicat ions Technical co mplexity of integrated system Diversity of systems to be connected Unequal sharing of costs and benefits Disagree ment among various local organizations about responsibility fo r information (local power issue) Lac k of consensus about degree of centralization Fear fo r loss of autonomy Insufficient local sophistication in terms of ability to collect data Lac k of t ransnational organization to develop the TIS Lac k of t ransnational organization to manage TIS Legal procedure in various countries concerning data and privacy Cultural/language differences between countries

Interorganizat ional International X X X X X X X X X X X X

Transnational Information Systems

133

From a systems development perspective, an “optimum” TIS requires high levels of cooperation among partners and involves consensus on many aspects of the system. In the context of cooperation among independent partners, it is neither always feasible nor desirable to achieve high levels of consensus and commitment. There are many potential disagreements and mismatches, which may mean that the “optimum” TIS is never built and that a TIS at a lower level of aspiration is developed and implemented instead. A sound systems engineering approach in the development stages implies dividing a large system into more manageable subsystems (mostly on a national scale) with welldefined interfaces between them. Subsystems may have different architectures and/or realizations, for which the overall system acts as a unifying “linking pin.” Over time, applied system extensions, whether serial or parallel, tend to yield a complicated overall structure, and a major architectural redesign becomes a good solution. Practical barriers to a TIS with a high level of aspiration include the number of participants, diversity among participants, and disagreement among participants. Such difficulties and problems are intrinsic to the development and management of systems that cross organizational and national boundaries. Evidence from the three cases showed that such difficulties are avoided by lowering the level of aspiration of the TIS. It appears that for many organizations it may be more important to get a TIS of some kind in place than to build the best possible TIS. The level of aspiration can be adjusted in two ways: by reducing the need for consensus (for instance, by involving fewer participants or allowing participation at different levels of functionality) and by reducing the potential for conflict (for instance, by selecting enthusiastic participants only or by agreeing to a lower level of functionality). Each of the cases studied during this project showed evidence of one or more of these adjustments to the level of aspiration of the TIS. Participants may have agreed to a diminished level of aspiration for the TIS, but the initial problems resurface at a later stage. When the “simplified” system is operational, attempts are made to improve and enhance the system by adding functionality, adding further partners, and increasing the scope of the TIS (e.g., in the case of EUCARIS). At that moment the problems, which led to a reduced level of aspiration in the first place, can surface again. It would seem that reducing the level of aspiration at early stages of development may postpone the moment of facing difficulties with the TIS, but it does not eliminate the problems. But facing them head-on would lead to the failure, or at least serious delay, of the TIS. Two other issues emerged from all cases. The first is the fact that the introduction of a TIS also involves the adaptation of national systems and organizations. The second is that development and management of TIS requires the wisdom to make the right tradeoff between desirability and feasibility of solutions. This means that the right balance has to be found between issues like technical possibilities, costs, organizational competencies, power, etc. In an earlier section of the paper we described the main TIS issues and highlighted the difficulties in various domains and in different phases of development. Many of these difficulties are initially addressed by lowering the level of aspiration of the TIS, thus circumventing or avoiding the problems. This is clearly a successful strategy to help get a TIS beyond its early development stages. However, in the long run the difficulties of working together with many different partners of varying ability and background must be addressed head-on. In this respect development and management of a TIS is hardly different from cooperating with partners in any other sphere, and specifically from development and management of a non-cross-border information system.

134 Mantelaers & van den Berg

CONCLUDING REMARKS This study highlighted TIS development in an empirical setting. The three TIS cases showed many similar traits during the development of the system. Many of the issues were identified on the basis of the experiences of more than one case. However, there were also issues which were found in one case only, but which highlighted the importance of certain notions. This study was exploratory in nature. We hope that continued research will support these preliminary results. By organizing surveys in various countries it would be possible to better compare TIS with identical or similar functions from different countries, and also IS from different business perspectives but in the same country. Findings from such extended studies should enable formulating guidelines and concepts in support of planning and management of TIS development and management. A specific goal would be to compile a list of failure and success factors for TIS-type projects, taking into account the degree of stakeholder agreement and commitment, the availability of financial, human, informational and technical resources, the willingness to use common standards and methodologies, and the feasibility of creating new organizations and institutions when needed.

ENDNOTES 1 Details of this case can be found in Mantelaers (1996). 2 Details of this case can be found in Zuurmond (1996). 3 Details of this case can be found in van den Berg & Zuurmond (1997).

REFERENCES Deans, P. C. and Kane, M. J. (1992). International Dimensions of Information Systems and Technology. PWS: Kent. Kaplan, B and Duchon, D. (1988). Combining qualitative methods in information systems research: A case study. MIS Quarterly, 12(4), 570-586. Kroon, N. (1997). Europese informatiesystemen: grensverleggend? Ph.D. thesis, Erasmus University Rotterdam, The Netherlands. Kumar, K. and van Dissel, H. (1996). Sustainable collaboration: managing conflict and cooperation in interorganizational systems. MIS Quarterly, 20(3), 279-300. Mantelaers, P. (1996). The TRANSIT project. Working Paper, Department of Mathematics and Information Systems, Delft University of Technology (in Dutch). Mantelaers, P. A. H. M., van den Berg, W. G. and van der Zee, H. T. M. (1998). Developing transnational systems: The transit-case. In Banerjee, P., Hackney, R., Dhillon, G. and Jain, R.. Business Information Technology Management: Closing the International Divide, 463-476. New Delhi: Har-Anand. Martin, J. (1989). Information Engineering, Book I: Introduction. Englewood Cliffs, NJ: Prentice Hall. Meier, J. (1995). The importance of relationship management in establishing successful interorganisational systems. Journal of Strategic Information Systems, 4(2), 135-148. Miles, M. B. and A.M. Huberman (1984). Analysing Qualitative Data: A Source Book for New Methods. Beverly Hills, CA: Sage.

Transnational Information Systems

135

Palvia, S. and Palvia, P. (1994). Global Issues of Information Technology Management. Hershey, PA: Idea Group Publishing. Parsons, T. (1960). Structure and Process in Modern Societies. New York: Free Press of Glencoe. Porter, M. and Millar, V. E. (1985). How information gives you competitive advantage. Harvard Business Review, July-August, 149-160. Reich, B. H. and Benbasat, L. (1990). An empirical investigation of factors influencing the success of customer oriented information systems. Information Systems Research, 1(3), 325-347. Turner, W. S., Langerhorst, R. P., Hice, G. F., Eilers, H. B. and Uijttenbroek, A. A. (1988). SDM. Pandata BV, Rijswijk (in Dutch). van den Berg, W. G. and Zuurmond, A. (1997). International transfer of cashless ambulatory payments. Working Paper, Department of Systems Engineering, Policy Analysis and Management, Delft University of Technology, The Netherlands. van den Berg, W. G. and Mantelaers, P. (1998). International cashless ambulatory payments. Proceedings of the 4th International Conference on Information Systems, Analysis and Synthesis, ISAS ’98, 3, 486-493. Webster, J. (1995). Networks of collaboration or conflict? Electronic data interchange and power in the supply chain. Journal of Strategic Information Systems, 4(1), 31-43. Yin, R. K. (1994). Case Study Research: Design and Methods. (Second edition). London: Sage. Zuurmond, A. (1996). EUCARIS–The European car and driving licence information system. Working Paper, Department of Policy Analysis, Erasmus University Rotterdam (in Dutch).

136 Usoro

Chapter X

The Place of ICT in Global Planning Abel Usoro University of Paisley, UK There is a general agreement in the literature that social, technological, political, cultural, and economic factors encourage a greater number of businesses to globalize their operations and markets. To operate in more than one market or country involves complexity on a larger scale than to operate locally. The complexity is combined with increasing risks, a faster pace of change, and the difficulties of managing an organization in more than one country. Information and communication technologies (ICT) act not only as an imperative to globalize but also as a potential tool to help global managers to plan, and yet there is no conclusive study that these technologies are adequately providing assistance. Indeed, existing studies suggest that there is much room for improvement. At the same time, there is no coherent and well-tested theory to explain or predict the use of information and communication technologies for global planning. To bridge the knowledge gap, this paper presents a causal model that groups predictor variables of the use of ICT in global planning under organizational, ICT, personal, and infrastructural factors. It also reports on a pilot study in which one hundred questionnaires were distributed to multinational companies in the United Kingdom (UK) and in South Africa (SA) to collect information to examine the role of ICT in global planning using the model. The result suggested among others that the Internet forms the most popular platform for building global planning tools. Factors most important to managers include the provision of timely information, provision of report and presentation facilities, and support for group working and alternative (highly summarized and detailed) views of information. On the other hand, managers appear not to be very satisfied with the provision of technology for global planning, partly because it does not adequately provide for creativity needed in global planning. Recommendations are made based on the findings, and areas for further research to enhance the validity of the model are highlighted.

INTRODUCTION The increasingly general trend for business activities to cross national borders in order to gain competitive advantage is well documented in the literature (cf. Hull, 1987; Hax, 1989; Ohmae, 1989; Ietto-Gillies, 1997; Business Week, 2000, p. 113; Harvey et al., 2000; Groose, 2000, Copyright © 2002, Idea Group Publishing.

The Place of ICT in Global Planning 137

p. vii). The crossing of national borders, which is termed “internationalization” (Taggart & McDermott, 1993, p. 4), can take different forms of business involvement. At the lower end of the spectrum is indirect export whereby a company distributes its products abroad by using third parties (Toyne & Walters, 1993, p. 114). At the highest level is globalization whereby there is no distinction between domestic and foreign operations but there is freedom to source or allocate resources or operations to, and choose markets from, any strategically advantageous location (cf. Humes, 1993; Ball & McCullock, 1996). All international businesses irrespective of their cross-border involvement must constantly evaluate their positions to decide, firstly, whether to maintain their positions or to move up or down the spectrum. Secondly, they have to decide how to maintain or change their positions. In these two respects, we can conclude that all international firms perform some form of global planning in as much as nonglobal international companies have to constantly evaluate whether or not to globalize. Strong global, social, political and economic pressures provide drives for nonglobal international companies to develop global strategies (cf. Humes, 1993, p. 25; Govindarajan & Gupta, 1999, p. 5-10). Alongside these developments are advances in information and communication technologies (ICT) to bridge the geographical, time, and knowledge distances experienced in global operations1 (cf. Uenohara, 1992, p. 402; Cambell et al., 1999; Currie, 2000). This paper will (a) explain briefly what is involved in global planning; (b) discuss the role of ICT in global planning; (c) identify problems and justify the need for investigation; (d) propose a hypothetical model; (e) state the method of study; (f) present and discuss the findings of a pilot study, including its major conclusions and recommendations; and (e) highlight areas for further research.

WHAT GLOBAL PLANNING IS Searching for “global planning” in research databases would produce very little result. Most writers in this area assume that readers already know what the concept is about. The concept can be used in various contexts, for example, health care and the provision of public utilities (cf. Ryan, 1990, pp. 61-3; Yehia et al., 1995, p. 10). In the context of this paper it means strategic planning, the managerial act of setting long-term goals and objectives based on the strengths and weaknesses of the organization, on the one hand, and the threats posed and opportunities offered by the environment, on the other (cf. Ansoff, 1979; Johnson & Scholes, 1993; Wheelen, 2000). Much has been written on the subject of strategic planning for business since the 1920’s when Harvard Business School developed the Harvard Policy Model (Carter, 1999, p. 1). Strategic planning became very popular with organizations until the 1980’s, when it fell out of favor because of the increased pace of environmental changes that made obsolete and irrelevant most of the long-term planning done over a long period of time by a few organizational members at the top. Consequently, new concepts such as TQM (Total Quality Management), BPR (Business Process Reengineering), and Value Chain Analyses emerged as short-term planning techniques to match the changing environments and capture customer preference. Besides, in flatter2 organizations, the idea of central planning and its top-down process became unpopular with middle managers, who are increasingly exposed to the environment and need to respond rapidly without being tied to handed-down plans. Academics reasoned that rather than strategic planning, organizational learning, employee empowerment, and agility are what contribute to competitive advantage. Feurer and Chaharbaghi (1995) give a comprehensive review of the history of strategy development, with

138 Usoro

the main theme that “the dynamic environments of today require a more dynamic approach to strategy development” (p. 11). It is interesting to note that recently strategic planning has re-emerged as the unifying organizational process that directs its different members towards optimal achievements (cf. Carter, 1999; Desai, 2000; Quazi, 2001). New research proves that strategic planning does contribute to competitive advantage. For example, Desai (2000) confirmed that the explicit statement of companies about their strategic planning focus, function, or orientation has an immediate positive effect on their share values. Instead of being viewed as a discouraging factor to strategic planning, the increasing uncertainty of the environment provides the impetus for managers to use the tools to cope with the complex, changing environments and to shape organizational responses (Morgan and Piercy, 1993; Quazi, 2001). Global planning is strategic planning but emphasizes the need to take a worldview of the changing environments and the pooling together of organizational knowledge that may be spread across the globe. Global planning is affected more by environmental diversity and changes and requires simplification, speed and flexibility. Taking the current view of strategic planning, emphasis has to be placed on the process (the learning process) instead of the products (plans) themselves, which are snapshots and quickly become history (Carter, 1999, p. 48). A large amount of information has to be easily and rapidly pulled together, analyzed, and critical success factors quickly identified. Involvement of a greater number of organizational members is important to modern strategic planning and is therefore vital in global planning. Planning has to be integrated with action,3 and there has to be readiness to change course, when necessary (Feurer & Chaharbaghi, 1995, p. 5; Chakravarthy & Lorange, 1991). To establish global “strategic intent” and set major objectives, an organization needs to define and clarify its mission and philosophy. An organization’s philosophy should include its global stance (i.e., ethnocentric, polycentric, regiocentric, geocentric, or some mixture of these viewpoints) (cf. Holt, 1998, p. 236; Rugman & Hodgetts, 1995, p. 215; Buckley, 1998, p. 13). An ethnocentric stance places emphasis on the country of origin, for example, in personnel issues (Hill, 1997, pp. 448-453). A polycentric view localizes business in each global location, while a regiocentric position does the same for trade regions (e.g., EU). The geocentric philosophy emphasizes total globalization or standardization of goods, services, and practices with little or no adaptation to local needs. A host of global, national, and company factors should be taken into consideration to decide on and constantly review the philosophy in the light of rapidly changing global economic, political and cultural situations. Managers face a difficult balancing act between meeting the demands of the increasing global converged economy and the need to support national policies that provide distinctive competitive advantage (Doremus et al., 1999, p. 165). Recent developments in information and communication technologies present the potentials for performing these tasks easier and faster (Alkhafaji, 1991).

HOW ICT COULD HELP ICT is not a replacement for human judgement and imagination but can assist global planning in • quickly identifying internal strengths and weaknesses, • performing environmental scanning, • bridging the geographical distance between planners, and • simplifying the planning process.

The Place of ICT in Global Planning 139

Competitive advantage has been strongly linked to internal resources and capabilities (cf. Kay, 1993, pp. 17-37; Penrose, 1963). These resources and capabilities are discovered by organizational audit which, is part of the first stages of planning. The required information is typically distributed throughout the organization. If this information is in electronic form and the information systems are compatible, the information can be pooled together regardless of distance. The Internet technology along with developments in enterprise information systems (e.g., SAP) present the potential for achieving these aims. Without going into detail, it is worth mentioning that the recent developments in the Internet communication language, termed XML, promise to make possible the exchange of more meaningful information. Thus, information can be classified, analyzed, and presented at greater speed. This capability can also be of use when scanning the environment. Environmental (e.g., economic), changes tend to be the most disruptive to global planning efforts (Chae & Hill, 1996, pp. 880-891). It is therefore important that environmental scanning is constantly performed to monitor political, social, technological, economic, and market changes. Not every change is relevant to every organization. The relevant ones need to be quickly identified and acted upon. There are a number of Internet-based technologies and applications developed to perform these tasks. Apart from XML already mentioned, Fletcher and Donaghy (1994, pp. 4-18), for instance, write about computer systems for monitoring competitors’ moves. Groupware technologies make possible the linking of planners distributed around the globe. It provides electronic conferencing and enables users to share the same source of information and exchange ideas. It is interesting that Fulmer and Sashkin (1995, pp. 26-31) considered groupware also as an essential tool for global learning organizations. Sokol (1992) discovered that simplifying the strategic planning process yielded benefits. These benefits include saving time in developing a plan and shortening the lead time to implement it, making the plan easier to understand, focusing better on only the most relevant business issues, and assembling a more consistent strategy for the whole company. It is implicit in the foregoing that ICT should simplify the planning process. Moreover, business and planning models such as the SWOT, PEST, and Porter’s Value Chain, and 5point models can be built into ICT, thereby simplifying and quickening the processing of information (Rugman and Hodgetts, 1995, pp. 218, 221-222). The presentation facilities of ICT should make possible the display of information from different perspectives and the ability to easily jump between a highly summarized view and detailed information. ICT should facilitate the capturing and management of organizational knowledge needed for global planning. With the simplification of the global planning process, a relatively small business should be able to enter the global market with ease (Tetteh and Burn, 2001).

PROBLEM Reid (1989) reported on a face-to-face study of 100 chief and senior executives in Scotland. His report indicated that companies were failing to utilize effectively the creative talents of key people in the harnessing of information. As a consequence they are failing to win valuable insights and omitting to obtain essential collective interpretations of critical issues and events that affect their strategic planning process. Apparently, the situation has not completely changed, especially in global planning, despite the potentials that ICT offers to the planning process (cf. Hagmann & McCohon, 1993, pp. 183-192; Houben et al., 1999; Usoro, 2001, pp. 17-24).

140 Usoro

Besides, there is no theoretical framework to examine the use of ICT for global planning. The development of a framework will help in explaining the current use of ICT for global planning and also provide guidance on how to improve the use of ICT in this area. Along similar lines, Ho (1996) stated that “the connection between strategy and IT has not been clearly articulated with respect to a finite set of concepts, analytical framework, and normative prescriptions” (p. 77).

A HYPOTHETICAL MODEL: FACTORS THAT INFLUENCE THE USE OF ICT FOR GLOBAL PLANNING There is no existing and well-researched theory that explains the use of ICT for global planning. Usoro (2001, pp. 17-23) made an initial attempt at designing a model to explain and predict the use of ICT for global planning. From a literature review and an experience survey, he hypothesized that factors could be grouped under organizational, information technology, personal and infrastructural elements to form a relationship as follows: Use of ICT for global planning = f(organizational, ICT, personal, and infrastructural factors) Organizational factors refer to an organization’s profile and include items such as size, years in business, and the stance along the ethnocentric and geocentric continuum (cf. Özsomer, Calantone & Bonetto, 1997). Jeannet (1999, p. 28) has described this as a “global mindset” that will determine how successful global managers plan. The study takes a step further by examining the effect of the mindset on the use of ICT in planning. The closer an organization is to the ethnocentric end, the less the level of globalization; the closer it is to the geocentric end, the higher would be the level of globalization. Does the level of globalization co-relate positively with the use of global planning technologies? Another factor closely related to the level of globalization is involvement in global planning. Do companies that are more involved in global planning tend to use information technology more? Other organizational factors of interest are years in business and number of countries in which a company operates. Do these make any difference with regards to the use of information technology for global planning? ICT factors refer to the attributes of ICT itself. ICT should not only exist in the organizations but also be relevant and supportive of the needs of global planners. It should support group working as well as possess the following attributes of executive support systems (ESS) (Usoro, 1998)4 : • easy user interface for learning and using the system • ability to easily switch between highly summarized and detailed views of information • on-demand link to internal information for indication of strengths and weaknesses • statistical analysis tool • ad hoc query and sensitivity analysis handling • access to external data pools–non-company data • flexibility to solve diverse problems • constant review of decisions, before and after implementation • report and presentation facilities

The Place of ICT in Global Planning 141

• •

support for optimization, satisficing, and heuristic approaches use and provision of accurate, understandable, complete, and timely information Ives et al. (1993) have also pointed out the need for compatibility of systems when they state that “few multinational firms can boast of … globally integrated information processing environment” (p. 114). Incompatible ICT systems present a barrier instead of helping global planning, and it is one of the aims of emerging technologies such as the Internet to overcome this barrier. This study examines the effect of these factors on the use of specific technologies such as the Internet, video conferencing, and Groupware. Personal factors are the users’ attributes, for example, their acceptance of change and attitude towards new technology. Holt (1998, p. 69) has discussed how human factors can hinder the use of available information technology. Attitude to information technology is a likely personal factor that could influence the use of global planning technologies. Many psychologists have theorized attitude as a significant personal attribute that tends to predict behavior. For instance, Ajzen and Fishbein (1980) concluded in their study that, provided they are appropriately measured, attitudes are sufficient to predict intentions (behavior). Moghaddam (1998) presents both sides of research and arguments as to whether attitude predicts behavior. He tends to conclude that we can use attitude to measure behavior provided (a) we are relatively specific in our measure; and (b) we measure all the components to provide a better chance of capturing all the facets of the attribute. The components of attitude, which makes operationalization possible, are cognitive, affective, and behavioral. This study measures knowledge of ICT use for global planning (cognitive), feelings5 about them (affection), and compares these with actual use (behavior). The development of questions to measure these components is based on attitude scales developed by Kay (1989), who later used it to predict commitment to use computers (Kay, 1990). Infrastructural factors refer to basic facilities such as electricity and telephone systems which may be inadequately provided in less developed economies; this inadequacy could therefore hinder the use of global planning technologies (Holt, 1998; Barker, 1993, pp. 57-59). While availability of supportive telephone systems including some technologies as ISDN (integrated services digital networks) may present no problem in developed economies, it may not be adequate in developing ones such as SA (South Africa). And that may present significant problems for the networking aspect of information technology needed for global planning. Big multinationals such as the Mobil Oil Company provide for these facilities including electricity and internet backbone.6 Can all multinationals provide for themselves where the government cannot? The availability of infrastructure and the ability of companies to provide infrastructure for themselves when these are inadequate is examined to investigate their effect on the use of ICT for global planning.

METHODS The factors expressed in the theoretical model (see Section 5) were used to develop a questionnaire, copies of which were distributed to a sample of multinational companies in the UK (United Kingdom) and SA (South Africa). The sample size was 50 for each country. Only 16 companies responded at the time of writing. Three of the companies apologized that it is against their policy to answer to survey questions. One left large areas of the questionnaire uncompleted because, according to the respondent, “I don’t believe we use IT as a global planning tool.” This questionnaire was consequently excluded in the analysis to avoid undue

142 Usoro

bias. It will, however, be very interesting to follow up this respondent with an interview to discover how they carry out international trade in approximately 8 countries without using ICT to plan. The analysis performed in this paper is based on the remaining 12 returned questionnaire (12% response rate), and therefore it is best to consider the results as derived from a pilot study. Very interesting observations can be deduced from the returned questionnaires. Correlation analysis and averages (arithmetic means) are used to analyze data collected largely by Likert scale type of questionnaire. In the presentation of findings, correlation coefficients7 are widely used.

FINDINGS AND DISCUSSIONS The discussion of findings is organized around the major factors that the questionnaire sought to measure. However, use of specific planning technologies are considered first. These technologies are Internet, intranet, extranet, groupware, enterprising planning tools, video conferencing, data warehousing/mining, and others.

Use of ICT in Global Planning As shown in Figure 1, all the respondents use the Internet for strategic planning; and nearly all use video-conferencing facilities. Four of the respondents use systems other than those listed on the questionnaire. These, reflected under the “other” category in Figure 2, are external databases (e.g., Reuters) and others that are not available on the Internet; data collection and analysis applications (e.g., Holos; Hyperion; Market Modeler; Epic); and spreadsheet applications (e.g., Excel and Lotus, p. 123). Although all respondents use the Internet, for those who own groupware systems, groupware systems are most frequently (daily) used (see Figure 2). The Internet and intranet is the next most frequently used system for global planning. This suggests a great need for collaborative working and external information sourcing in global planning. The factors that influence the use of global planning8 tools are now examined.

Figure 1: Use of information technology 100% 90% 80% 70% 60%

Use No Use

50% 40% 30% 20% 10% 0%

de o co

ris

nf

e

e

is

t

s er g/ th O in us ho re wa ta g ng Da i n i n ci m en

Vi

rp

w

ne

up

te

ro

En

G

tra

t

et

ne

rn

tra

te

Ex

In

In

er

Pl an ni ng

The Place of ICT in Global Planning 143

Figure 2: Frequency of ICT use 100% 90% 80% 70%

Never

60%

Yearly

50%

HalfYearly

40%

Monthly

30%

Weekly

20%

Daily

10% 0% O er

o

a

de

at

th

D

Vi

s

ar ou si ng /m

ng

ng

ni

ci

an

en

Pl

er

eh

e

e

nf

ris

is

co

w

rp

t

w

t

ne

up

te

ro

En

G

tra

et

ne

rn

tra

Ex

te

In

In

in in g

Organizational Factors Strategic stance or level of globalization was operationalized in terms of (a) standardization of products, services, and practices irrespective of country of operation; (b) not exporting “home” standards to foreign countries; and (c) number of countries operated in. These three strands of globalization level as well as involvement in global planning and years in business are discussed below.

Standardization of Products, Services, and Practices Standardization of products, services, and practices correlate positively (0.31)9 with the use of ICT for global planning. Since standardization attenuates variety, it is expected that use of ICT would be facilitated. However, this does not imply that much original planning or exploration of new opportunities worldwide is taking place, hence the zero correlation between standardization and involvement in global planning (which was defined as identification of business opportunities irrespective of country). This indicates that ICT for global planning has need to give more support to creativity.

Not Exporting “Home” Standards to Foreign Countries A correlation coefficient of (-0.18)10 indicated that the more companies use original country standards as de facto abroad, the more they use ICT for global planning. Again, this situation suggests less originality and variety to deal within the planning process. Some companies in relatively stable industries and markets may be satisfied with that; but the picture would be different with companies in relatively turbulent industries. The problem might be that present ICT planning tools are not adequate for such situations.

Number of Countries Operated in Surprisingly, there is a negative relationship (-0.038) between number of countries operated in and the use of ICT for global planning. It also appears that the more the countries operated in, the less do the companies adopt a global standard (-0.45) and the more do they use original home country standards (0.19). It therefore appears that the greater the number

144 Usoro

of countries operated in, the more planning is delegated to individual countries or regions, therefore causing a reduction in overall global planning and use of information technology for that purpose. Also, this result may indicate that very large companies with a presence in large number of countries are not having adequate help to plan globally.

Years in Business There is a very high positive relationship (0.7) between the years in business and the use of ICT in global planning. This may be due to accumulated experience and financial backing, which enables acquisition and use of the systems.

Involvement in Global Planning Though not very large, there is a positive relationship (0.008) between involvement in global planning and the use of information technology. It appears the fewer the number of countries operated it, the more the involvement in global planning (-0.05). This may reflect the aim of smaller companies to explore more new opportunities, which was depicted in the definition of involvement in global planning. It is interesting also that companies which tended to be involved in global planning were also keen on not exporting “home” standards abroad (0.31). The small correlation, though positive, between involvement in global planning and use of ICT probably indicates that inadequate use is made of information technology in global planning.

ICT Factors ICT attributes (see Section 5) were measured. On average, the highest four ICT attributes that were found to be important to users are (a) provision of timely information (b) provision of report and presentation facilities (c) support for group working within the same site (d) support for group working in more than one site within a country These attributes weighted the least: on-demand link to internal information for indication of company strengths and weaknesses (b) access to external data pools–non-company data (c) adequate support given for use of systems (d) alternative views of information with highly summarized and detailed views It is paradoxical that systems that are weak on linking with internal and external data sources are described as providing timely information. This suggests that the timely information is less complete than would be desired, hence the absence of “provision of complete information” among the top four attributes. It is interesting to observe that when each of the factors is correlated with the use of information technology, almost all correlate positively. The two exceptions are (a) report and presentation facilities (0.11); and (b) adequate support given for use of systems (0.04). The four top correlated factors, in their order of priority, are (a) alternative views of information with highly summarized and detailed views (b) on-demand link to internal information for indication of company strengths and weaknesses (a)

The Place of ICT in Global Planning 145

(c) (d)

flexibility to solve diverse problems access to external data pools–non-company data

Personal Factors Cognitive and affective components of attitude were measured and compared to the behavioral component. There is a high positive relationship (0.46) between cognition and affection. This means that the more managers know the capabilities of global planning technologies, the more they feel positive about using them. Satisfaction with the use of global planning technologies was measured as part of affection. Correlating satisfaction with use results in a positive relationship of 0.18, which suggests that we can increase the level of use by increasing level of satisfaction with the systems. The fact that the positive relationship is not very high indicates that managers are not necessarily very happy with the systems they are using. This suggests the need to investigate how to provide more satisfactory systems. There is surprisingly a negative correlation (-0.34), though not large, between the behavioral component of attitude (use) on the one hand and cognition and affection on the other. This finding is at variance with findings of other studies, such as McGuire et al. (1999, pp. 54-55), that propose that cognition (knowledge) should be positively related to behavior. The smallness of the sample perhaps explains for this unexpected result. Otherwise, it may be another case of proof that attitude does not always predict behavior (Moghaddam, 1998). Other personal factors that were examined are age, gender, education and experience. Experience exhibits the highest positive relationship (0.41) with use. This result agrees with studies by Walters and Necessary (1996, pp 623-11) and Igbaria and Chakrabarti (1990, pp. 229-241). Age is the next in the value of positive relationship (0.25). This may be because with age comes experience generally, which is shown in the 0.37 relationship between the two factors. On the other hand, education co-relates negatively (-0.33) with use. This result is in contrast with earlier studies by McGuire and Hillan11 (1999, pp. 54-55) and Igbaria and Chakrabarti12 (1990, pp. 229-241); and this contrast might be because the question referred to the level of education rather than its content. It might also mean that the higher the level of education, the less information technology content. It made no sense to relate gender with the use of information technology because all the respondents were male, tending to confirm that top companies’ jobs are still dominated by males. A future study can verify the use by females of global planning tools by the use of quota sampling of respondents.

Infrastructural Factors As expected (see section 3.4), there is a positive correlation (0.2) between the use of ICT for global planning and the provision of infrastructural facilities. The correlation is more pronounced with provision of electricity (0.36) than telephone systems (0.012). The respondents were offered the opportunity to indicate other aspects of infrastructural facilities that may be relevant, but none was indicated.

Other Findings Of interest to this research was the difference between South African and non-South African companies. On average, use of ICT in global planning by South African companies is 2.5 (arithmetic mean), whereas non-South African multinationals score 3.8, indicating greater use. This outcome suggests that ICT planning tools are used more by UK-based

146 Usoro

companies than SA companies. The reason for this difference needs further exploration. However, this result is not conclusive given the small number of respondents upon which the calculations are based.

MAJOR CONCLUSIONS AND RECOMMENDATIONS The pilot study aimed at examining the use of global planning tools by multinational companies. The results indicate that Internet, groupware, enterprise planning, and video conferencing tools are very popular. What managers claim most important are the provision of timely information, report and presentation facilities, and support for group working. When level of use of global planning tools is compared to view of ICT attributes, other important ICT factors (attributes) emerged, for example, alternative (highly summarized and detailed) views of information. Another major finding is that managers are not very satisfied with the provision of technology for global planning. This tends to agree with the finding that the information technology they use for global planning does not provide adequate creativity needed in global planning. It is a major recommendation of this paper, therefore, that investigation be made on how to use ICT more creatively in global planning. Also, other ICT attributes that count highest to managers should be incorporated in any development of computerised global planning tools.

FURTHER RESEARCH Further research should examine how to use ICT more creatively in global planning. Moreover, while the major predictor variables of the theoretical model have been derived from secondary research, it is necessary to reconsider and possibly expand the subfactors. For instance, leadership style could be included within personal factors. It has been studied in relation to its effect on the use of strategic tools and models (Drago and Clements, 1999, pp. 11-18). Also, organizational factors should incorporate the control and regulation of planning process and the structure of the organization. Jarvenpaa and Ives (1993, p. 547) discovered a poor fit between organizational structure and global information technologies. This reworking of the theoretical model has to be performed before further primary data could be collected to validate its explanatory and predictive ability.

ENDNOTES 1

2 3

The International Monetary Fund (1997) defines “globalization” as “the growing economic interdependencies of countries worldwide through the increasing volume and variety of cross-border transactions in goods and services and of international capital flows, and also through the rapid and widespread diffusion of technology” (p. 45). In structure. Hertzberg (1995) stated that “every failure of implementation is, by definition, also a failure of formulation. The real blame [of unsuccessfully planned strategies] has to be laid, neither on formulation or on implementation, but on the very separation of the two.

The Place of ICT in Global Planning 147

4 5 6 7

8 9 10 11

12

…. It is the disassociation of thinking from acting that lies closer to the root of the problem” (p. 285). Similar lists have been compiled by Stair and Reynolds (1998, p. 459-460) and Clare and Stuteley (1995, pp. 55-59). Feelings include level of satisfaction. A major Internet infrastructure. Correlation coefficients represent relationships of two sets of data at a time. Their values range from –1 (perfect negative correlation) to + 1 (perfect positive correlation). The nearer the coefficients are to these two values, the stronger the relationship. The more the coefficients are close to 0, the less the relationship; at 0, there is no relationship (Carlson & Thorne, 1997). In many cases “use of global planning technologies” will simply be referred to as “use.” A positive correlation coefficient. The negative correlation coefficient is between the use of ICT and globalization as expressed by not exporting home standards abroad. McGuire and Hillan found out that although the midwives they studied had a positive attitude (feeling) about computers, they considered the lack of the necessary skills as a hindrance to using computers. Igbaria and Chakrabarti found computer training to be contributing strongly to decrease in computer anxiety.

REFERENCES Alkhafaji, A. F. (1991). Management challenges: A worldwide perspective. In Management Decision, June, 29. Ansoff, H. I. (1979). Strategic Management. London: Macmillan. Ball, D. A. and McCullock. (1996). International Business: The Challenges of Global Competition. Chicago, IL: Irwin. Barker, R. M. (1993). Information system development in a global environment. In Business Forum, Winter-Spring, 18(1-2), 57-59. Buckley, P. J. (1998). A perspective on the emerging world economy: protectionism, regionalization and competitiveness. In Mirza, H. (Ed.), Global Competitive Strategies in the New World Economy, 12-21. Cheltenham: Edward Elgar. Business Week. (2000). August 28, i3696, 113. Cambell, D., Stonehouse, G. and Houston, B. (1999). Business Strategy: An Introduction. Oxford: Butterworth Heinemann. Carter, H. (1999). Strategic planning reborn. Work Study, 48(2), 46-48. Chae, M. S. and Hill, J. S. (1996). The hazards of strategic planning for global markets. In Long Range Planning, 29(6), 880-891. Chakravarthy, B. S. and Lorange, P. (1991). Adapting strategic planning to the changing needs of a business. In Journal of Organizational Change Management, 4(2). Currie, W. (2000). The Global Information Society. NY: John Wiley. Desai, A. B. (2000). Does strategic planning create value? The stock market’s belief. Management Decision, 38(10), 685-693. Doremus, P. N., Keller, W. W., Pauly, L. W. and Reich, S. (1999). The myth of the global corporation. In Harvard Business Review, 77(1), 165-171. Drago, W. A. and Clements, C. (1999). Leadership characteristics and strategic planning. In Management Research News, 22(1), 11-18.

148 Usoro

Feurer, R. and Charharbaghi, K. (1995). Strategy development: Past, present and future. Management Decision, 33(6), 11-21. Fletcher, K. and Donaghy, K. (1994). The role of competitor information systems In Information Management & Computer Security, 2(3), 4-18. Fulmer, R. M. and Sashkin, M. (1995). Tools for the global learning organization. In American Journal of Management Development, 1(3), 26-31. Govindarajan, V. and Gupta, A. (2000). Analysis of the emerging global arena. In European Management Journal, June, 18(3), 274-284. Grosse, E. (Ed). (2000). Thunderbird on Global Business Strategy. Toronto: John Wiley & Sons, Inc. Hagmann, C. and McCahon, C. S. (1993). Strategic information systems and competitiveness: Are firms ready for an IST-driven competitive challenge? In Information and Management, 25, 183-192. Harvey, M., Griffith, D. and Novicevic, M. (2000). Development of timescopes to effectively manage global inter organizational relational communications. In European Management Journal, December, 18(6), 646-662. Hax, A. C. (1989). Building the firm of the future. In Sloan Management Review, Spring, 75-82. Hertzberg, H. (1995). The Rise and Fall of Strategic Planning. Upper Saddle River, NJ: Prentice Hall. Hill, C. W. L. (1996). Global Business Today. New York: McGraw Hill. Ho, C. (1996). Information technology implementation strategies for manufacturing organizations: A strategic alignment approach. In International Journal of Operations and Production Management, 16(7), 77-100. Holt, D. H. (1998). International Management: Text and Cases. London: The Dryden Press. Houben, G., Lenie, K. and Vanhoof, K. (1999). A knowledge-based SWOT-analysis system as an instrument for strategic planning in small and medium sized enterprises. In Decision Support Systems, August, 26(2), 125-135. Hull, C. W. (1987). Business in a global economy. In Hydrocarbon Processing, December, 61-66. Humes, S. (1993). Managing the Multinational: Confronting the Global-Local Dilemma. London: Prentice Hall. Ietto-Gillies, G. (1997). Internationalization trends. In Global Business Strategy. London: International Thomson Business Press, 73-90. Igbaria, M. and Chakrabarti, A. (1990). Computer anxiety and attitudes towards microcomputer use. In Behavior and Information Technology, 9(90), 229-241. The International Monetary Fund. (1997). World Economic Outlook, 45. Ives, B., Jarvenpaa, S. L. and Mason, R. O. (1993). Global business driver: Aligning information technology to global business strategy. In IBM Systems Journal, 32(1), 143-161. Jackson, T. (1999). Master of his art. In Financial Times Mastering Management Review, July, (25), 10-13. Jarvenpaa, S. L. and Blake, I. (1993). Organizing for global competition: The fit of information technology. In Decision Sciences, May-June, 24(3), 547-581. Jeannet, J. (1999). Strategies in the spider’s Web. In Mastering Global Business, 28-32. London: Pitman. Johnson, G. and Scholes, K. (1997). Exploring Corporate Strategy. London: Prentice Hall.

The Place of ICT in Global Planning 149

Kay, J. (1993). The structure of strategy. In Business Strategy Review, 4(2), 17-37. Kay, R. H. (1989). A practical and theoretical approach to assessing computer attitudes: The computer attitude measure (CAM). In Journal of Research on Computing in Education, Summer, 456-463. Kay, R. H. (1990). Predicting student teacher commitment to use of computers. In Journal of Educational Computing Research, 6(3), 299-309. McGuire, M. and Hillan, E. (1999). Obstacles to using a database in midwifery. In Nursing Times, January, 95(3), 54-55. Moghaddam, F. M. (1998). Social Psychology: Exploring Universals Across Cultures, 99129. New York: W. H. Freeman and Company. Morgan, N. A. and Piercy, N. F. (1993). Increasing planning effectiveness. In Management Decision, 31(4). Ohmae, K (1989). Managing in a borderless world. In Harvard Business Review, May-June, 67, May-June, 152-161. Özsomer, A., Calantone, R. J. and Bonetto, A. D. (1997). What makes firms more innovative? A look at organizational and environmental factors. Journal of Business & Industrial Marketing, 12(6), 400-416. Penrose, E. T. (1963). The Theory and the Growth of the Firm. Oxford: Blackwell. Reid, D. M. (1989). Data access and issue analysis in strategic planning. In Marketing Intelligence & Planning, 7(1). Rugman, A. M. and Hodgetts, R. M. (1995). International Business and Strategic Management Approach. New York: McGraw-Hill. Rugman, A. M. and Hodgetts, R. M. (1995). International Business and Strategic Management Approach. New York: McGraw-Hill. Ryan T. (1990). Dermatology–Global planning in relation to leprosy management. In Leprosy Review, September, 61(3), 209-212. Sokol, R. (1992). Simplifying strategic planning In Management Decision, 30(7). Taggart, J. H. and McDermott. (1993). International Business. London: Prentice Hall. Tetteh, E. and Burn, J. (2001). Global strategies for SMe-business: Applying the SMALL framework. In Logistics Information Management, 14(1), 171-180. Toyne, B. and Walters, P. G. P. (1993). Global Marketing Management: A Strategic Perspective (Second edition). London: Allyn & Bacon. Uenohara, M. (1992). Innovation on a global scale. In Chemtech, July, 22(7), 402-405. Usoro, A. (1998). A tool for strategic planning to support managers. 8th Annual BIT (Business Information Technology) Conference, 4-5 November. Usoro, A. (2001). Can information technology help managers plan globally? In Journal of Global Information Management, 9(1), 17-24. Walters, J. E. and Necessary, J. R. (1996). An attitudinal comparison towards computers between underclassmen and graduating seniors. In Education, Summer, 116(4), 623711. Wheelen, T. L. (2000). Strategic Management and Business Policy: Entering 21st Century Global Society (Seventh edition). Upper Saddle River, NJ: Prentice Hall. Yehia, M., Chedid, R., Ilic, M., Zobian, A., Tabors, R. and Lacallemelero, J. (1995). Global planning methodology for uncertain environments–Application to the Lebanese power-system. In IEEE Transactions on Power Systems, February, 10(1), 332-338.

150 Dirksen

Chapter XI

The Cultural Construction of Information Technology Vanessa Dirksen Universiteit van Amsterdam, The Netherlands

In this paper, I propose a research framework called the ‘“cultural construction of information technology.” It extends the more widely known concept of the social construction of technology (SCOT) and defines the concept of a cultural infrastructure for the application of information technology in other contexts than the one in which it is developed. Whether or not today’s economy is characterized best by the term globalization,1, the interweaving of economic processes and the growing number of internationally operating organizations in the world are matters of fact. The shaping of such a “global economy” is facilitated, among others, by information technology (IT) as it enables the crossing of both organizational as well as national borders. However, as in the case of so-called cultural universals, the universal applicability of IT is a myth also. It is therefore my contention that notwithstanding the blurring of borders, acknowledging the specificity of national cultures remains important, maybe even more than before, and should not be overlooked in the international application of IT. This is what we call the glocalization paradox2 of information technology. The paradox consists of both a global, border-crossing dimension of IT, as well as a local/national context dimension of IT thrust. Resolving this paradox is an important issue regarding the global application of IT. Thus, the application of IT across national boundaries stresses the growing awareness of cultural diversity, and ultimately, the need for the development of IT taking into account heterogeneous “target environments.” Apart from the fact that the target environment, or “host culture,” should not be conceived of as a cultural homogeneous environment, technology itself should not be conceived of as a culturally neutral phenomenon (Pacey, 1993). Rather, information technology is value-loaded; it reflects the values of the culture in which it is developed (Kumar and Bjorn-Anderson, 1990, in Jarvenpaa and Tractinsksy, 1995). Consequently, information systems can be conceived of as “carrying and communicating a worldview packed with assumptions, marked by the interests and ideologies that conceive them” (Depres, 1996, p. 17). The application of IT in another context than the one in which it emerged, that is, the country in which it is designed and developed, often causes problematic implementation and Appeared in Journal of Global Information Management, vol. 9, no. 1, 2001. Reprinted by permission.

The Cultural Construction of Information Technology 151

ultimately a low appropriation of IT. The reason for this is found in the fact that the values of its home country disagree with the norms and values of the target environment. Because culture significantly impacts the different effects of the introduction of the same information technology within different cultural contexts, we will address this cultural dimension of IT by introducing a research framework called “the cultural construction of information technology” (CCOIT). The cultural construction of IT extends the more widely known concept of the social construction of technology (SCOT) (Mackenzie and Wajcman, 1985). In the SCOT approach, “technology is treated as part of a system of different but interlocking elements–physical artifacts, institutions, and their environments” (Sahay et al., 1994, p. 249). The social constructionist approach to technology is drawn from the sociology of knowledge, i.e., “[t]he study of how styles of expression and the character of ideas or systems of thought are related to different social contexts” (Bullock et al., 1988, p. 457). It expresses the resistance to the existence of cultural universals and verges on the theories of cultural relativism.3 Thus, the social construction of knowledge argues that “knowledge is a social construction rather than a (more or less flawed) mirror held up to nature” (Bijker & Law, 1992, p. 13). In the same line of reasoning, the SCOT approach holds that “technologies and technological practices are built in a process of social construction and negotiation, a process often seen as driven by the social interests of participants” (Bijker and Law, 1992, p. 13). The CCOIT, then, not only views technology as an outcome of the working of society, but holds that viewing technology as part and parcel of a society entails investigating information technology in interaction with “culture.” Questions arising in the field of CCOIT concern, for one, the way that culture influences technology. Apart from this Weberian tradition, which questions how values constrain or enable technological practices and development, we extend the notion of SCOT in that it does not dismiss the deterministic properties of (information) technology (Orlikowski, 1992). This calls for a Marxist examination of the way that information technology also influences cultural values. In this tradition one could investigate, for instance, how cultural boundaries and conflicts arise under the influence of technological developments. A third area of research that CCOIT covers concerns the introspection of the technological community. Investigated in this area are the ways in which the internal working of the technological community and the norms and values of its members affect the technological outcome (cf. Alexandrov, 1994). It should be said that we limit ourselves to the introduction, implementation, deployment and maintenance phases4 of information technology infrastructures in “alien” target environments as divorced from their preceding processes of design taking place in the home country. This centers our focus on the ways culture influences the application of IT and how culture actually transforms the technology in use. This focus is based on the Duality of Structure as advocated by Gidden’s structuration theory which states that “the structural properties of social systems are both medium and outcome of the practices they recursively organize” (Giddens 1984, p. 25). The duality of technology consequently comprises the idea that technology is both created and changed by human action as well as used by humans to accomplish some action (Orlikowski, 1992). Furthermore, we posit the practice of IT as a Geertzian form of “cultural interpretation,” This acknowledges the role of the actor and stresses the need to study culture from within, reconstructing the native’s point of view as much as possible by trying to distill the meaning of the world as given by the people themselves who live in it. The “interpretative flexibility” of (information) technology is derived from the conviction that people in different contexts interpret the meaning of technology in different ways (Winner, 1993). These interpretations

152 Dirksen

of the technology are often referred to as ‘technological frames of meaning.’5 “The term ‘technological frame’ refers to the concepts, techniques, and resources adopted by technologists and others.6 It is a way of talking of the set of theories, expertise, values, methods of testing, and physical tools and devices available to communities as they negotiate about the putative character of innovation” (Bijker & Law, 1992, p. 19).7 In short, the explication of the different technological frames of meaning is of vital importance to the application of IT as they influence the actual interaction between people and the technology. One could say that they determine the interaction between culture and IT and, ultimately, the transformation of both.

INCORPORATING THE NOTION OF CULTURAL INFRASTRUCTURE An information technology infrastructure comprises a physical infrastructure made up of technological components, as well as a social and cultural infrastructure. The notion of cultural infrastructure is comparable to what Ciborra and Lanzara call “formative context,” “i.e., the set of the preexisting institutional arrangements, cognitive frames and imageries that actors bring and routinely enact in a situation of action” (Ciborra & Lanzara, 1994, p. 70). In order to comprehend the notion of cultural infrastructure, we must first outline its components. Firstly, our understanding of the rather vague and complex concept of culture is characterized best in terms of tradition, i.e., “a background within which we interpret and act; the historicity of our way of thinking–the fact that we always exist within a pre-understanding determined by the history of our interactions with others who share the tradition” (Winograd & Flores, 1986, p. 7). Culture in this respect is well represented by the Russian word byt, “… signifying “a way of life,” an aggregate of customs, habits, mores, and so forth, characterizing a particular people, social stratum, or group (Alexandrov 1994, p. 65).8 The majority of the information systems’ research publications refers to the concept of culture in terms of organizational or corporate culture only. We, however, explicitly stress the importance of studying IT in a context not confined merely to the boundaries of the organization. Although the area of research will be confined to IT application on an organizational level, the technological practice should be analyzed within the broader context of the national culture. This because “tools transferred from one country to a specific enterprise abroad suffer a ‘double-layered acculturation’: The technology is confronted with a foreign national and alien corporate culture” (Barkema et al., 1996, in Recht & Wilderom 1998, p. 8). The organizational culture is thus but one aspect of the cultural infrastructure. According to Schein’s definition of the term, organizational culture9 develops as a result of surviving in and adapting to the external environment and the gearing of internal processes to one another (Schein, 1992). The result is a shared frame of reference from which that same environment is defined (Van Muijen, 1994). Although we tend to agree with such a vision of organizational culture, we dispute its rather one-sided perspective in representing the management/leadership point of view only. In accordance with the Geertzian tradition as introduced in the preceding paragraph, we suggest that organizational culture be approached from the bottom-up perspective as well, since this will reveal not what “ought to be” but “what actually is.” Moreover, we disagree with the apprehension of corporate culture prevalent in the majority of information systems’ research publications, being used as an arbitrary classification. Pinpointing culture in general terms only implies it to be a static and uniform

The Cultural Construction of Information Technology 153

phenomenon. Rather, culture should be defined from a socio-historical development perspective, acknowledging it to be a context-dependent phenomenon. It should be said however, that we are not propagating any form of sterile relativism. That is to say, we do not rule out the possibility of systematic comparison with comparable phenomena. One of the rare studies assessing the broader context of IT in terms of “softer factors” influencing its apllication is a research concerning the implementation of GIS in India conducted by Sahay and Walsham (1996). In this study, apart from recognizing the organizational structure, Sahay and Walsham also consider the scientific tradition to be an important feature of the context responsible for contributing to the generation of homogeneous technological frames of meaning. The scientific tradition, or the scientific byt “relates to the prevailing beliefs about how science and technology is viewed in [a country], and the norms and values underlying these beliefs” (Sahay & Walsham, 1994, p. 394, cf. Alexandrov, 1994).10 The tradition, or the shared frames of reference of the various disciplines, the design, implementation, use, and management of IT, generates specific attitudes toward information technology in terms of, for instance, functions, possibilities, and goals.11 The different groups of people working with the technology, such as designers, users, and managers, may have different frames of meaning, depending on their individual backgrounds and interests (Barley, 1990 in Sahay et al., 1994, p. 248). The scientific tradition is thus another component of the cultural infrastructure. Taking the granulation of culture further adds yet another dimension to the cultural infrastructure of IT. One could also investigate the degree of “sharedness in meaning” according to the “historical positioning” of the people interacting with the technology, that is, a person’s place within contemporary society as an outcome of its specific historic tradition. Since “individual careers and collective developments are interwoven” (Hagendijk, 1996, p. 152) this could manifest itself, for instance, as a “generational phenomenon.” The notion of generation, in this respect, refers to personal courses of life related to the parallel development–life cycle12–of technical artifacts. This historical positioning of the people who interact with technology induces their specific conceptualization of this technology. Assessing the “personal histories”13 of these people against the life cycle of IT might thus reveal shared meanings that could be ascribed to a certain generation. The cultural infrastructure of information technology is thus composed of structures of accumulated historical and individual experiences that give rise to specific information technology practices characteristic of particular groups of people divided by nation, organization, scientific tradition, professional community, and/or generation. The investigation of the cultural infrastructure should therefore incorporate the analysis of several closely interrelated dimensions of national dominant values, organizational culture, expertise or professional culture, and individual histories, as they influence the practice of information technology. In order to study the interaction or interrelationship of IT and culture, the cultural infrastructure should also be specified in terms of the actual characteristics and attributes of the information technology. But also, the varied understandings of the role of information in an organization or a society underlie the social processes of acceptance or resistance to IT innovations and are inherent to the application and transformation process. For instance, “[i]n some countries the basic assumption that information is good and that all people should have access to public information is not as widely shared” (Mansell and Wehn, 1998, p. 145). This conviction will most probably constrain a positive attitude toward the use of IT (Mansell & Wehn, 1998). Focusing on “how people approach and handle information” (Davenport,

154 Dirksen

1997, p. 83)–information behavior–could thus also be insightful to the understanding of the interrelationship between IT’s physical and cultural infrastructure. Encapsulated, information culture is “… the pattern of behaviors and attitudes that express and organization’s [and society’s] orientation toward information” (Davenport, 1997, p. 84). To sum up, technological frames of meaning are prompted by the various cultural histories, those of the organization, professional/expertise community, and of generations, as related to the technological, economic, and ideological development of a country in general, and as related to the state of the “informational society” 14 in particular. The explication of the technological frames of meaning requires the compilation of dialectical histories, i.e., linking personal, corporate, and national histories to the lifecycle of information technology. And so, ideally, “the historical specificity or tradition of people and organizations converge with the socio-technical logic of the informational paradigm” (Castells, 1996, p. 21).

THE CULTURAL CONSTRUCTION OF IT IN RUSSIA The significance of relating technological frames of meaning to cultural histories within the big picture of the ruling ideological and economic situation and technological development of a country is illustrated by some rather stereotypical or superficial insights in Russian information technological culture.15 With the “opening up” of the Soviet Union to the West, new market forces and the information technology revolution make their way into Russia. Consequently, new and sophisticated technologies from the West are increasingly introduced in Russian firms to upgrade or replace the Soviet machinery. However, the application of IT originating from the West can be expected to be problematic for various reasons. The problems that can be anticipated are those resulting from a deviant or insufficient technological infrastructure and most likely also those resulting from variations in organizational structure (Palvia et al., 1992).16 The problematic application of IT as a result of the “cultural infrastructure,” however, is considerably less predictable. As an outgrowth of Russia’s rather turbulent history, we expect the “cultural framing” of IT in Russia to be subjected to a high degree of controversy due to the nature of contemporary Russian society. It follows that the acceptance or rejection of IT could depend on (the perseverance or disappearance of) certain controversies, persisting as logical outcomes of Russia’s ideological, economic, and social history. These “struggles” could become apparent conform to the division by generation, manifested as a cleavage between Russians born, raised, and educated during communist times and the younger generation who grew up after the ideological, economic, and discursive turn. Firstly, the older generation may, for example, be inclined to be more influenced by the tradition of Soviet thought in their perception of technology, still applying to some extent the norms and values of Soviet culture [sovyetskaya kultura]. For instance, the notion of the purpose of IT could be influenced by the Soviet dogma that “production, technology and innovations, and science have meaning and are justified only insofar as they promote military might, the expansion of territory, and the strengthening of the state” (Rakitov, 1993, p. 80). Secondly, the old-time distinction between “ours” and “not ours” [nashi–ne nashi] might become apparent in the Russian framing of information technology. This distinction

The Cultural Construction of Information Technology 155

reflects the idea of technological progression in which Russian tools, or Soviet for that matter, are preferred over Western tools in fear of technological and cultural dependence on the West (cf. Cats-Baril & Jelassi, 1994).17 A third factor in the interpretative scheme of the “Soviet generation” may be the association of information systems with control. “[E]nterprises that played by the rules in the Soviet economy risked cutting their own throats; computerized management information systems are designed to help enterprises play by the rules” (McHenry & Goodman, 1986, 1040). A fourth example of a generation-stereotypical attitude toward IT may stem from the rather specific Soviet reluctance to information sharing, that is, “the voluntary act of making information available to others” (Davenport, 1997, p. 87).18 The perseverance of this tradition could generate an averse attitude in terms of information transparency and therefore also toward the use of the computer as an informatory device.

CONCLUSION I reject the idea of truly universally applicable IT on the same grounds as I reject the existence of cultural universals. It is my contention that “global IT” should also recognize the importance of the relevant aspects of the national contexts of IT which create heterogeneous target environments for its application. Hence, I propose a construction of IT with emphasis placed on the concept of culture, that is, in terms of the history and tradition of a particular country. The interpretation of information technology in contemporary society should then be understood against the historical background of the country in which the technology is implemented, and also, against the history of computing. In short, I hold that the different meanings attributed to technology are generated by people’s specific cultural and “historical positioning” within this wider context. I therefore emphasize the need to study the cultural histories that shape local IT practices. The global adoption of IT could, for example, be enhanced by considering the “generational” effects in the national sub-field of information technology and the personal contexts of agents interacting with IT during application of so-called global IT. For example, information analysis for global IT should not be restricted to functional requirements but must also acknowledge the existence of different technological frames of meaning and therefore incorporate cultural requirements. A socio-cultural-historical analysis during information analysis could be imperative. The question of how to incorporate these dimensions in the development of global IT remains unanswered, however, and forms a subject for future research. With respect to the cultural constructivism of science, Hagendijk (1996) also discusses the discourse coalitions in scientific and social change, and public controversies regarding science. Interesting future research in the cultural constructivism of IT could be directed toward demonstrating the relevance of discourse coalitions in and public controversies about IT. Furthermore, as IT reflects the values of its culture of origin, it should be productive to question to what extent IT can be said to be an ethnocentric device to begin with, that is, a device in which Western rationality is the criterion by which other cultural bodies of knowledge are being evaluated.19 Then it rather becomes a question of technological displacement.

156 Dirksen

ENDNOTES 1

2

3

4

5 6 7

8 9

10 11 12 13

14

15

16 17 18 19

In general, the term “globalization” refers to the free transfer of goods, capital and information across a growing number of accessible continents, assuming eventually the abolishing of national borders. “Globalization in this context involves not the leavening impact of universal processes but, on the contrary, the calculated synthesis of cultural diversity” (Gordon, 1994 in Castells, 1996, p. 393). “The social constructionist view originated as a reaction to the early misuse of Darwinism: Using biology, the analogy of natural selection, to explain and justify everything from class stratification to the domination of White Europeans. Boas and his students [Margaret Mead and Ruth Benedict] argued that differences between human groups could be ascribed to culture and not to nature” (Fukuyama, 1998, p. 28). Deployment is the phase “in which the resulting system is installed in its target environment” and maintenance is the phase “in which the system is continually modified, upgraded and debugged” (Goguen, 1994, p. 176). A term of Collins and Pinch (1982). Comparable to Heidegger’s “Gestell” or “enframing”; “it is that attitude to the world that is the foundation for, yet wholly present in, technological activity” (Durbin, p. 320). “Bijker presses the view that both social groups and technologies are generated in the contingent arrangements of the concepts, techniques, and resources brought together in the relevant technological frames” (Bijker & Law, 1992, p. 19). The term byt is comparable to Pierre Bourdieu’s understanding of “habitus.” The concept of culture represented in the first wave of the corporate culture literature, “a move away from the hard aspects of technology and structure” (Van Muijen, 1994). Alexandrov studied “how the evolution of a discipline was linked to the byt, the culture of the scientific tradition” (Alexandrov, 1994, p. 80). One could, for example, see a computer primarily as an automating or an informating device (Zuboff, 1988). The notion of lifecycle refers to the progression of technology as a consequence of new technologies competing with old ones (Rip, 1995). The trend of micro-histories was initiated as a reaction against the search for “hard structures” and is characterized by localizing the subject and encouraging more existential studies of society. Informational society refers to “a specific form of social organization in which information generation, processing, and transmission become the fundamental sources of productivity and power, because of new technological conditions emerging in this historical period” (Castells 1996, p. 21). Drawn from an unpublished preliminary study conducted by the author concerning the explication of the Russian understanding of the term “culture” as related to the idea of technological progression. The decentralized information systems from the West will most likely be inappropriate for use in most Russian organizations (Palvia et al., 1992). This introduces the notion of IT as an ethnocentric and colonizing device. Information sharing is, according to Davenport (1997), apart from handling overload and dealing with multiple meanings, one of the three critical types of information behavior. This perspective would entail viewing information technology as a potential colonizing device, subjugating the traditional way of life.

The Cultural Construction of Information Technology 157

REFERENCES American Anthropological Association/ Computing Research Association. (1995). Culture, Society and Advanced Information Technology, AAA/CRA, June 1-2. Alexandrov, D. A. (1994). The historical anthropology of science in Russia. Russian Studies in History, Fall, 62-91. Bijker, W. E. and Law, J. (Eds.). (1992). Shaping Technology/ Building Society: Studies in Sociotechnical Change. Cambridge, MA: MIT Press. Bullock, A., Stallybrass, O. and Trombley. S. (Eds.). (1988). The Fontana Dictionary of Modern Thought. London: Fontana Press. Castells, M. (1996). The Information Age: Economy, Society and Culture, Volume I. The Rise of the Network Society. Oxford: Blackwell Publishers. Cats-Baril, W. L. and Jelassi, T. (1994). The French videotext system minitel: A successful implementation of a national information technology infrastructure. Management Information Systems Quarterly, March, 1-19. Ciborra, C. U. (1999). Hospitality and IT. PrimaVera Working Paper 99-02, Universiteit van Amsterdam. Ciborra, C. U. and Hanseth, O. (1998). From tool to Gestell: Agendas for managing the information infrastructure. PrimaVera Working Paper, 98-03, Universiteit van Amsterdam. Ciborra, C. U. and Lanzara, G. F. (1994). Formative contexts and information technology: Understanding the dynamics of innovation in organizations. Accounting, Management & Information Technology, 4 (2), 61-86. Davenport, T. D. (1997). Information Ecology: Mastering the Information and Knowledge Environment. Oxford: Oxford University Press. Depres, C. D. (1996). Information, technology and culture: An ethnography of information technology and modernist business organization. Technovation, 16(1), 1-20. Durbin, P. T. (1980). A Guide to the Culture of Science, Technology, and Medicine. New York: the Free Press. Giddens, A. (1984). The Constitution of Society. Cambridge: Polity Press. Gobbin, R. (1998). The role of cultural fitness in user resistance to information technology tools. Interacting with Computers, 9, 275-285. Fukuyama, F. (1998) Women and the evolution of world politics. Foreign Affairs, September/ October, 24-40. Hagendijk, R. P. (1996). Wetenschap, Constructivisme en Cultuur. Amsterdam. Hirschheim, R., Klein, H. K. and Lyytinen, K. (1996). Exploring the intellectual structures of information systems development: A social action theoretic analysis. Accounting Management & Information Technology, 6 (1-2), 1-64. Ives, B. and Jarvenpaa, S. L. (1991). Applications of global information technology: Key issues for management. Management Information Systems Quarterly, March, 33-49. Jarvenpaa, S. L. and Tractinsky, N. (1995). Information systems design decisions in a global versus domestic context. Management Information Systems Quarterly, December, 507-534. Kumar K., van Dissel, H. and Bielli, P. ( 1998). The merchant of prato -revisited: Toward a third rationality of information systems. Management Information Systems Quarterly, June, 199-226. Mansell, R. and Wehn, U. (1998). Knowledge Societies: Information Technology for Sustainable Development. New York: Oxford University Press.

158 Dirksen

McHenry, W. K. and Goodman, S. E. (1986). MIS in Soviet industrial enterprises: The limits of reform from above. Communication of the ACM, 29(11), 1034-1043. Montealegre, R. ( 1998). Managing information technology in modernizing “against the odds”: Lessons from an organization in a less developed country. Information and Management, 34, 103-116. Orlikowski, W. J. (1992). The duality of technology: Rethinking the concept of technology in organizations. Organization Science, 3(3), 398-427. Pacey, A. (1983). The Culture of Technology. Oxford: Basil Blackwell. Palvia, S., Palvia, P. and Zigli, R (1992). The Global Issues of Information Technology Management. Hershey, PA: Idea Group Publishing. Rakitov, A. I. (1993). Civilization, culture, technology, and the market. Russian Social Science Review, July/August, 73-84. Recht, R. and Wilderom, C. (1998). Kaizen and culture: On the transferability of Japanese suggestion systems. International Business Review, 7, 7-22. Rip, A., Misa, T. J. and Schot, J. (1995). Managing Technology in Society: The Approach of Constructive Technology Assessment. London/ New York: Pinter Publishers. Sahay, S., Palit, M. and Robey, D. (1994). A relativist approach to studying the social construction of information technology. European Journal of Information Systems, 4(3), 248-258. Sahay, S. and Walsham, G. (1996). Implementation of GIS in India: Organizational issues and implications. International Journal of Geographical Information Systems, 10(4), 385404. Schein, E. H. (1992) Organizational Culture and Leadership. San Franciso, CA: JosseyBass. van Muijen, J. J. (1994). Organisatiecultuur en Organisatieklimaat: De Ontwikkeling van een Meetinstrument op Basis van het ‘“Competing values” Model. Amsterdam. Winner, L. (1993). Upon opening the black box and finding it empty: Social constructivism and the philosophy of technology. Science, Technology & Human Values, 18(3), 362374. Winograd, T. and Flores, F. (1987). Understanding Computers and Cognition: a New Foundation for Design. Reading, MA: Addison-Wesley. Zuboff, S. (1988). In the Age of the Smart Machine: the Future of Work and Power. New York: Basic Books.

The Cultural Construction of Information Technology 159

Section II Regional Perspectives

160 Lacity & Willcocks

Chapter XII

Survey of IT Outstanding Experiences in US and UK Organizations Mary Cecelia Lacity University of Missouri–St. Louis, USA Leslie P. Willcocks Oxford University, UK

The global IT outsourcing market is estimated to exceed $121 billion by the year 2001. To assess current market practices and experiences, a survey was distributed to 600 US and UK CIOs. The 101 US and UK respondents are generally pleased with information technology (IT) outsourcing. In particular, respondents rated overall supplier performance as “good,” respondents mostly realized the benefits they expected from IT outsourcing, and respondents characterized the majority of problems/issues as only “minor” in nature. The healthy IT outsourcing report card is likely explained by the scope and type of IT outsourcing practiced by responding organizations. The vast majority of respondents pursue selective outsourcing which is less risky than total outsourcing. Most respondents also use multiple suppliers rather than a single supplier, which allows for best-of-breed supplier selection. The healthy report card may also be explained by the types of IT activities selected for outsourcing. Respondents generally targeted IT infrastructure activities—such as disaster recovery, mainframe operations, network management, midrange operations, PC support, and help desk operations—rather than IT development or IT strategy. UK and US practices and outcomes were very similar, although a few exceptions are noteworthy. On average, UK organizations (30%) totally insourced IT more frequently than US organizations (8%). US organizations (29%) more frequently used a single supplier than UK organizations (9%). UK organizations (50%) use only one stakeholder to negotiate/define contracts compared to US organizations (9%). Differences may be explained by a more matured approach to outsourcing in the USA together with the higher preponderance of larger deals and organizations studied. Findings are compared to prior survey and case study research. Ever since Kodak’s landmark decision to outsource the bulk of their IT functions in 1989, IT outsourcing has been a widely publicized practice. Most of us are familiar with a Appeared in Journal of Global Information Management, vol. 8, no. 2, 2000. Reprinted by permission.

Survey of IT Outstanding Experiences 161

number of other high-profile IT outsourcing mega-deals besides Kodak. Companies that have outsourced significant portions of their IT functions by transferring their IT assets, leases, licenses, and staff to outsourcing vendors include British Aerospace, British Petroleum, Chase Manhattan Bank, Continental Airlines, Continental Bank, DuPont, Enron, First City, General Dynamics, McDonnell Douglas (now Boeing), and Xerox. Mega-deals have also been signed in the public sector, including a £1 billion contract between the United Kingdom’s Inland Revenue and EDS and a $600 million contract between the South Australia Government and EDS. Overall, global market revenues have grown from $US 9 billion in 1990 to a projected $US 121 billion by year 2001. The underlying compound annual growth rate has been 15-20 percent in the 1992-98 period, with the leading markets in the US and UK. Other countries are also showing market increases. The Australian market, for example, has grown from $AUS 2.2 billion in 1998 to an estimated $AUS 3.87 billion in 2002 with a 24 percent annual growth rate from 1998-2002. A recent conference with Japanese CIOs sponsored by DEC also indicates that Japan is interested in IT outsourcing as a turnaround practice in response to the Asian monetary crisis (Lacity and Willcocks, 1999). Clearly, IT outsourcing has outlived the fiveyear period typical of a management fad. But is the rapid growth of the IT outsourcing market primarily attributable to the wellpublicized and studied mega-deals? Are high-profile, immense contracts indicative of the sourcing practices of most organizations? Are customers satisfied with their IT outsourcing practices and outcomes? These questions prompted a recent survey of IT outsourcing experiences to supplement the ever-growing number of mega-deal case studies. The survey was targeted at the two countries which have the largest and most established IT outsourcing markets—the US and UK. The survey was distributed to 600 US and UK Chief Information Officers (CIOs). This paper presents the findings of this survey and compares findings—where relevant—to previous surveys. The vast majority of the 101 respondents pursue selective outsourcing of IT services rather than total outsourcing. Overall, the respondents are generally pleased with selective outsourcing. In particular, respondents rated overall supplier performance as “good,” respondents mostly realized the benefits they expected from IT outsourcing, and respondents characterized the majority of problems/issues as only “minor” in nature. Overall, US and UK outsourcing practices and outcomes are similar, but notable differences are discussed.

RESEARCH METHODOLOGY In 1994, Willcocks and Fitzgerald (1994) developed an IT outsourcing survey for distribution in the UK. In 1996/97, Currie and Willcocks (1997) replicated the survey in the UK to assess how the practices and outcomes have evolved over the three-year period. For this paper, the survey was replicated, with amendments, for the first time in the US and for a third time in UK. By using the same survey instrument, the long-term research goal is to track IT outsourcing practices over time and to compare practices over time across countries.1 Respondents were asked to identify the percentage of IT budget outsourced, the types of IT activities outsourced, the anticipated benefits of outsourcing, people and processes used to evaluate IT outsourcing, contract details, contract duration, and outcomes in terms of actual benefits received, supplier performance, and extent of problems encountered. In particular, the questions targeted at outcomes were deemed important because few prior

162 Lacity & Willcocks

surveys assessed outcomes—most merely assessed practices (Collins & Millen, 1995 and Grover et al.,1996 are notable exceptions). While most of the thirty-plus questions were closed-ended, respondents were also asked to cite the most important lessons they have learned. The open-ended questions provided an opportunity for respondents to express views not identified in the close-ended questions. To replicate the survey in the US, the UK version was shortened because US CIOs— the target audience—would not likely complete an eight-page questionnaire. The US version is six pages long and differs from the UK version only in minor ways, such as adapting the survey for differences in language. For example, the UK survey uses the terms “turnover,” “redundancy,” and “federal structure.” Corresponding US terms are “revenue,” “attrition,” and “mixed/matrix structure.” For the US survey, a list of 1,000 of the largest US organizations across various manufacturing and service sectors was purchased from Dun & Bradstreet Information Services. This list included the names, addresses and phone numbers of the organizations, but not the names or exact titles of the Chief Information Officers. Every organization was called on this list to obtain the name and correct title of the head of the information systems department, in hopes that personally labeling envelopes and cover letters would increase the response rate. In the spring of 1997, surveys were mailed to the top 500 names on the US list. The UK sample was based on the Financial Times Top 100 list and a list provided by the research sponsors, British Intelligence, and the Oxford Institute of Information Management (OXIIM). Because the Financial Times list focuses on private sector organizations, a subset of the sponsors’ list was added to include public sector organizations. The 100 UK surveys were mailed in early 1998 to a contact person responsible for IT within the organization, usually an IT director but sometimes a senior business manager with IT responsibilities. A total of 101 usable surveys were returned, providing an overall response rate of 17%. The response rate compares favorably with similar surveys of this nature, especially given the number of questions and the detailed nature of the requested responses. Thirty-eight surveys were returned from the US and 63 surveys were returned from the UK.

SURVEY RESPONDENTS Job Titles of Survey Respondents. CIOs or senior executives were the target response group because respondents needed to have enough stature to answer strategic and IT management-related questions. To ensure the target person responded to the survey, the people filling out the surveys were asked to give their names, titles, and addresses. Respondents were motivated to complete this information so they could receive survey results. Indeed, the job titles of respondents indicates a high level of authority. Respondents are Directors (33%) of Computers and Computer Systems, Computing and Communication Services, Corporate Information Technology, Information and Information Services, IT Enterprises, IT Services, or Technical Services; Managers (24%) of Information Resources, IS Administration, IT Services or MIS, Chief Information Officers (21%) or Head of IS; and Vice Presidents (11%) of Division, Computers and Information Systems, Information Systems, or MIS (See chart “Job Titles of Survey Respondents”). Industries Represented. The 101 respondents represent a number of manufacturing and service industries (see table “Industries Represented”). Three industries—consumer products, financial services, and industrial products—represent nearly half of the sample.

Survey of IT Outstanding Experiences 163

Figure 1: Job titles of 101 US and UK survey respondents Did not indicate title 11.9% Vice President 10.9%

IT Director 32.7%

12 11

33

21 24 CIO/Head of IT 20.8% IT Manager 23.8%

Table 1: Industries represented Industries Represented Consumer Products Financial Services (Insurance, Investment, Banking) Industrial Products Energy (Petroleum, Chemicals) Public Services (Government, Hospitals, Police) Leisure Services (Restaurant Chains, Hotel Chains, Travel Agencies) Transportation (Rail, Air) Healthcare Utility Other Total

US

UK

Total Number of Respondents

7 5 7 7 0

9 11 7 1 7

16 16 14 8 7

4 3 0 1 4 38

2 3 5 3 15 63

6 6 5 4 19 101

Industries which only have one respondent represented are grouped in the category “other.” There is a greater diversity of industries represented in the United Kingdom (15 UK companies categorized as “other”) compared to the United States (4 US companies categorized as “other”). Average Revenues. Respondents reported annual revenues for private sector companies, total annual operating budgets for public sector organizations, or total assets for banks and insurance. Seventy-eight people answered this question. To compare US and UK size of firms, UK dollars were converted to US dollars using the exchange rate .5998. The average US annual revenue is nearly $11 billion compared to the average UK revenue of $1.3 billion (See table “Annual Revenues”). Despite the large difference in means, one-tailed ttest assuming heteroscedasticity (p = .03) indicates that the US and UK revenues are only marginally different. This finding is explained by the large standard deviation. Annual IT Budget. Respondents were asked to indicate their annual IT budget for the organization they are representing. Ninety-eight people answered this question. To compare US and UK size of firms, UK dollars were converted to US dollars using the exchange rate .5998. The average US IT budget is $200 million compared to the average UK IT budget of $6.9 million (see table “Average IT Budget”). A one-tailed t-test assuming heteroscedasticity

164 Lacity & Willcocks

Table 2: Annual revenues Minimum Maximum Average Standard Deviation Number of Respondents

United States Revenues $30 million $168,800 million $10,995 million $29,158 million 32

UK Revenues converted to US$ $1 million $12,000 million $1,311 million $2,728 million 46

Table 3: Size of annual IT budget Minimum Maximum Average Standard Deviation Number of Respondents

United States IT Budget $3 million $4,000 million $200 million $657 million 36

UK IT Budget converted to US$ $.1 million $111.0 million $6.9 million $16.0 million 62

Table 4: Scope of sourcing decisions Total Outsourcing Selective Outsourcing Total Insourcing Total

US 3 (8%) 32 (84%) 3 (8%) 38 (100%)

UK 3 (5%) 41 (65%) 19 (30%) 63 (100%)

Total US & UK 6 (6%) 73 (73%) 22 (22%) 101 (100%)

(p = .04) indicates that US and UK IT budgets are only marginally statistically different. Again, this finding is explained by the large standard deviation.

SCOPE OF SOURCING DECISIONS Finding: 73% of respondents pursue selective outsourcing The 101 respondents reported a wide spectrum of sourcing decisions, ranging from exclusive use of internal IT functions to large-scale IT outsourcing. For the purposes of the analysis of this survey, we adopted the sourcing definitions used in Lacity and Willcocks (1996): Total Outsourcing: the decision to transfer IT assets, leases, staff, and management responsibility for delivery of IT services from an internal IT function to an external IT provider which represents more than 80% of the IT budget. Selective Outsourcing: the decision to source selected IT functions from external provider(s) while still providing between 20% and 80% of the IT budget internally. This strategy may include single or multiple suppliers. Total Insourcing: the decision to retain the management and provision of more than 80% of the IT budget internally. Using these definitions, six organizations engage in total outsourcing. On the opposite end of the spectrum, 22% of organizations exclusively use internal IT functions to provide IT services. Seventy-three assume a “middle-of-the-road” approach by contracting for only select subsets of IT activities (see table “Scope of Sourcing Decisions”).

Survey of IT Outstanding Experiences 165

Total outsourcing is pursued by the following six organizations: A US airline entirely outsources IT to a subsidiary company. This subsidiary company operates a world-famous reservation system that is used by many other airlines and travel agencies. • A US automotive manufacturer has a 12-year contract with two large suppliers. One of these IT suppliers used to be a wholly-owned IT subsidiary until it became an independent company. • A US $1.6 billion manufacturing company outsources all of IT to multiple suppliers for five years. • A UK export company has a five-year contract with one supplier for all IT infrastructure and applications. It also has a three-year contract with another supplier for support services, such as the helpdesk. • A UK life insurance company has a three-year total outsourcing contract with one supplier for mainframes, networks, applications development, and applications support. This company insources midrange computers and desktops. • A UK insurance company that is currently insolvent has a one-year, total outsourcing contract for 100% of IT services. Concerning total insourcing, there is a difference across countries: only 8% of US respondents claim to totally insource, compared to 30% in the United Kingdom. The greater propensity of US organizations to outsource might be explained by a greater maturity of the market on the supply side—many of the largest vendors developed from a US base, and by the proportionately greater number of larger organizations in our sample. Of the 22 total insourcing cases in both countries, all but two organizations have never even considered outsourcing IT. Thus continued IT insourcing may be based on precedent rather than a rational evaluation of market options. By far, the most common practice is selective outsourcing, as indicated by 84% of US and 65% of UK respondents pursuing this sourcing strategy. This finding is important because the majority of press releases focus on mega-contracts associated with total outsourcing. The news-worthiness of multimillion and multi-billion dollar deals may skew perceptions of the frequency of the practice (Lacity and Hirschheim,1995). Our finding is similar to many other sample surveys: • In a survey of 300 IT managers in the US, on average less than 10% of the IT budget was outsourced (Caldwell, 1996). • A survey of 110 Fortune 500 companies found that 76% spent less than 20% of the IT budget on outsourcing, and 96% spent less than 40% (Collins and Millen, 1995). • A survey of 365 US companies found that 65% outsourced one or more IT activities, but only 12 outsourced IT completely (Dekleva, 1994). • A survey by IDC found that “in the United States, outsourcing takes around 17 percent share of the IT services market” (Foley, 1993). • A survey of 48 US companies identified domestic and global IT sourcing practices of America’s most effective IT users, as determined by Computerworld’s Premier 100 list (Apte et al., 1997). Seventy-seven percent of the respondents outsourced at least one domestic IT function, but outsourcing was generally targeted at select activities such as support operations, training and education, disaster recovery, etc. Thus, although the large multimillion/billion dollar, long-term deals make headlines, past surveys have systematically unveiled that selective IT outsourcing is the more common practice. With selective outsourcing, discrete IT activities, representing a much smaller portion of the IT operating budget, are outsourced. •

166 Lacity & Willcocks

IT ACTIVITIES OUTSOURCED Finding: The most commonly outsourced functions involve IT infrastructure. Prompted by a list of 13 IT activities, respondents were asked to indicate whether they currently outsource part or all of the IT activities. Thirty-five US and 44 UK respondents answered this question. (See table “IT Activities Outsourced” for the most and least commonly outsourced IT activities). The patterns of IT activities outsourced are generally similar across US and UK organizations. In general, infrastructure operations are the most commonly outsourced activities, including disaster recovery (68%), clients servers and personal computer operations (67%), mainframe operations (61%), networks (57%), and mid-range computing (54%). The least commonly outsourced IT activities involve IT management and applications. Only 5% of respondents outsource IT strategy, 10% outsource procurement, and 28% outsource systems architecture. These activities are often considered strategic in nature, and even in the few instances of outsourcing, the activity is usually only partially outsourced. Another interesting finding is that 27 (34%) respondents outsource IT project management, but only three UK respondents totally outsource project management—the other 24 organizations only partially outsource. Several differences in US and UK IT outsourcing are apparent. Overall, UK organizations that selectively outsource outsource IT activities more frequently than the US organizations for nine of the 13 areas. In particular, there is a large difference in mid-range computer outsourcing—31% in US compared to 73% in UK. In the four areas that US figures are higher than UK figures (end-user support, help desk, project management and IT strategy), closer scrutiny reveals that US organizations primarily only partially outsource these activities. Compared with other surveys, we see much overlap in results as far as the most commonly outsourced and insourced IT activities. In the US, surveys indicate that the types of services most commonly outsourced are contract programming, education and training, PC maintenance, disaster recovery, applications development, and data entry (see table “Global Surveys on IT Outsourcing”). Thus US and UK organizations are tending to outsource or keep in-house similar activities, suggesting that similar principles for making IT sourcing decisions are being applied in both countries. Although surveys rank-order the most commonly outsourced IT activities differently, these same activities surface again and again. And like our survey, no surveys found that companies systematically outsource strategic planning or customer liaisons. And surveys generally indicate that 60% of applications are still developed in-house.

SINGLE VERSES MULTIPLE SUPPLIERS Finding: 82% of respondents use multiple suppliers Respondents were asked whether they use a single supplier or multiple suppliers. Of the 79 organizations that engage in IT outsourcing, 65 (82%) use multiple suppliers (see table, “Single verses Multiple Suppliers”). US organizations have less propensity to use multiple suppliers. Explanations may lie with organizations having a less fragmented, more mature approach to outsourcing, and/or their larger size attracting larger vendors able to accomplish a wider portfolio of activities. There are several advantages to using multiple suppliers, in fact. First, customers can select “best-of-breed” suppliers for each outsourced IT activity (Chapman and Andrade, 1997). For example, a customer may choose one supplier to manage the networks while choosing another supplier to man the help desk.

Survey of IT Outstanding Experiences 167

Second, customers mediate risk because no one supplier has sole monopoly power. Third, customers can motivate supplier performance through an environment of competition (Lacity et al., 1995; 1996). However, there are also two disadvantages with a multiple supplier strategy. Companies incur higher transaction costs because they conduct multiple evaluations and negotiate multiple contracts. During post contract management, the administration overhead is higher when dealing with multiple suppliers, particularly if supplier activities are integrated and thus require additional coordination. Two respondents cited the need to integrate services among multiple suppliers as a major lesson learned: “Develop better framework for lead vendors to service integrate” — UK Head of IT, Chemicals Company “Define multiple vendor relationships” — UK Service Manager, Insurance Company

DECISION SPONSOR Finding: IT managers are the sole sponsors of 47% of outsourcing decisions. Respondents were asked which stakeholder group(s) in their organizations initiate/ sponsor major IT outsourcing evaluations—steering committees, board members, senior business managers, and/or IT managers/directors. Thirty-five US and 44 UK people responded to this question. In 58% of the cases, only 1 stakeholder sponsors the decision, typically the IT Director (47%). In 35% of the cases, two stakeholders sponsor the decision, and again IT Directors are almost always one of the two participants. In only 6% of the cases are three or more stakeholders involved. (see table “Decision Sponsor”). Three prior studies found similar results. Apte et al. (1997) found that in the US, MIS executives initiate 95% of outsourcing decisions, top management participates 20% of the time, and departmental executives participate 15%. Willcocks and Fitzgerald (1994), in a survey of 162 UK organizations, found that in 47% of the companies, IT managers initiate outsourcing alone. Lacity and Willcocks (1998) in a multiple case study of 61 decisions found that 41% of decisions were sponsored by IT managers alone and 29.5% of decisions were jointly sponsored by senior executives and IT managers. Despite survey findings, several respondents indicated on the open-ended question that more stakeholders should have been involved in the decision: Table 5: IT activities outsourced IT Activity

Disaster Recovery Client/Server & PCs M ainframe Networks M id-Range End-User/PC Support Help Desk Project M anagement Systems Analysis Systems Design Systems Architecture Procurement IT Strategy

US Partial

US All

15 16 1 8 1 17 16 13 10 8 8 3 2

6 7 20 8 10 2 6 0 0 0 1 0 0

US Partial and All 21 23 21 16 11 19 22 13 10 8 9 3 2

US Partial/ All% 60% 66% 60% 46% 31% 54% 63% 37% 29% 23% 26% 9% 6%

UK Partial

UK All

13 22 19 18 21 15 5 11 10 12 10 4 1

20 8 8 11 11 5 9 3 4 3 3 1 1

UK Partial and All 33 30 27 29 32 20 14 14 14 15 13 5 2

UK Partial All% 75% 68% 61% 66% 73% 45% 32% 32% 32% 34% 30% 11% 5%

Total Partial and All 54 53 48 45 43 39 36 27 24 23 22 8 4

Total Partial/ All% 68% 67% 61% 57% 54% 49% 46% 34% 30% 29% 28% 10% 5%

168 Lacity & Willcocks

“More detailed request for proposal by consulting more of the people who do the work.”—US Senior Financial Analyst “Make sure everyone is involved that needs to be involved. Make sure all parties understand the agreement.”—US Director of Corporate Information Technology “In-house staff should be fully involved in the decision process.”—UK IT Director, Travel Company

STAKEHOLDERS RESPONSIBLE FOR DEFINING OUTSOURCING CONTRACTS Finding: In 32% of the cases, only one stakeholder defines and negotiates outsourcing contracts. Once the decision has been made to outsource, how many stakeholders actually negotiate and define the contract? Contract negotiations typically require highly specialized knowledge about contract law and IT technical specifications for service levels to ensure delivery (Halvey and Melby, 1996). Thus, we would expect that more stakeholders are involved in contract negotiations than in making the initial sourcing decision. Thirty-four US and 44 UK people responded to this question. Overall, 32% of outsourcing contracts are only negotiated by one stakeholder group. However, this finding is attributable to the large preponderance of UK cases involving only one stakeholder in Table 6: Global surveys on IT outsourcing Author(s) Arnett and Jones (1994)

Survey Survey of 40 US Cios

Collins and Millen (1995)

Survey of 10 US companies

Dekleva (1994)

Survey of 365 CIOs and CFOs

Grover, Cheon and Teng (1994; 1996)

Survey of 63 US companies Survey of 188 companies

Willcocks and Fitzgerald (1994)

Survey of 162 UK CIOs

Sobeland Apte (1995); Apte, U., Sobol, M., Hanaoka, S., Shimada, T., Saarinen, T., Samela, T. and Vepsalainen, A. (1997)

Survey of 48 US companies, 86 Japanese companies and 141 Finnish companies

The most commonly outsourced IT functions were: Contract programming (67%) Mini/Mainframe maintenance (67%) Software support and training (56%) Workstation/PC maintenance (39%) Systems integration (28%) Education and training (50%) PC support (49%) Network services (33%) Applications development (33%) Application maintenance (26%) Data centers (24%) Software maintenance (39%) User training (37%) Applications development (35%) Microcomputer support (35%) Disaster recovery (22%) Data centers (7%) % growth over 3 years: Systems operations (36%) Applications development and maintenance (30%) Telecommunications management (17%) End user support (16%) UK Outsourcing: Hardware maintenance (68%) User training and education (42%) Data centers (38%) PC Support (34%) US Outsourcing: Finnish Outsourcing: Support operations (48%) Software development (48%) Training & education (48%) Support operations (46%) Disaster recovery (40%) Software maintenance (42%) Software development (33%) Data network (39%) Data entry (22.9%) Training & education (38%) Japanese Outsourcing: Software development (61.6%) Data center operations (44.2%) Software maintenance (38.4%) Support operations (33.7%)

Survey of IT Outstanding Experiences 169

Table 7: Single vs. multiple supplier

Single Supplier Multiple Supplier Total

US 10 25 35

US % 29% 71% 100%

UK 4 40 44

UK % 9% 91% 100%

Total 14 65 79

Total % 18% 82% 100%

US % 46% 3% 3% 0% 51% 23% 6% 9% 6% 43%

UK % 48% 7% 7% 2% 64% 7% 14% 7% 0% 29%

Total % 47% 5% 5% 1% 58% 14% 10% 8% 3% 35%

3% 0% 3% 3%

2% 5% 7% 0%

3% 3% 5% 1%

Table 8: Decision sponsor Stakeholder: IT Director Only Board Only Senior Executive Only Steering Committee Only Only 1 Stakeholder Senior Executive and IT Director Board and IT Director Steering Committee & IT Director Steering Committee & Senior Executive 2 Stakeholders Steering Committee, Senior Executive, IT Director Board, Senior Executive, IT Director 3 Stakeholders All 4 stakeholders

contract negotiations (50% of UK cases). In the US, we see that more stakeholders are involved in negotiating and defining contracts, probably because the outsourcing deals are relatively bigger in terms of effort and price (see table “Number of Stakeholders Involved in Defining Contracts”). Once again, several respondents indicated on the open-ended questions that more stakeholders should have been involved in negotiations: “More experienced personnel involved in contract negotiations.”—UK IT Director, Travel Company “Use more time of a more experienced manager to negotiate contract.”—UK IS Director, Public Library “More IT involvement in contract.”—UK IT Director, Rental Car Agency “Negotiate aggressively, competitively, through seasoned heavyweight business negotiators.”—UK Head of IT, Chemicals Company Finding: IT managers (82%) and internal lawyers (64%) are the most common stakeholders involved in defining/negotiating IT outsourcing contracts. Respondents were asked to indicate all the stakeholders typically involved in detailing their outsourcing contracts, such as external lawyers, internal lawyers, consultants, suppliers, board members, senior business managers, and IT managers. Thirty-four US and 44 UK people responded to this question.

170 Lacity & Willcocks

In all of the US cases and in 68% of the UK cases, the IT director is involved. Internal lawyers are the next most commonly involved stakeholder, participating in 64% of negotiations (see table “Percentage of Negotiations Involving Stakeholder”). The board, outside consultants, and the suppliers are the least common participants. Outside experts are brought in more frequently in the US than in the UK. US organizations hire external lawyers in 38% of the responding organizations as compared to 20% in the UK. US organizations also hire outside consultants (35%) more frequently than UK organizations (5%). Country differences may be due to the more litigious nature of US society and the availability of such experts in the US. It may also be because, again, many contracts being dealt with in the US part of the sample are relatively bigger and attract more attention and concern. Like our current survey, Willcocks and Fitzgerald (1994) found in a survey of 162 European companies that 53% of companies use multiple stakeholders to define the contract. Unlike our current survey, however, Willcocks and Fitzgerald (1994) found that the most commonly involved stakeholder was internal lawyers, not IT managers. They noted from their case research, however, that “internal lawyers may have little experience of IT outsourcing contracts, as opposed to contracts generally” (p. 294). Perhaps we are witnessing an organizational learning effect. Early customers may have learned not to rely solely on internal lawyers to negotiate IT outsourcing contracts.

CONTRACT COMPLETENESS Finding: In 34% of the cases, respondents have complete contracts by including all important contractual clauses. We asked participants to indicate the types of clauses they include in their IT outsourcing contracts. Our list was generated from prior research on 61 case studies which found that the contract was one of the most important determinants to outsourcing success (Lacity and Willcocks, 1996). People who negotiate outsourcing deals are not necessarily the ones who manage the contract. Thus, it is vital to document the terms of the relationship. Participants were asked which of the following 10 types of contract clauses are typically included in their IT outsourcing contracts: 1. Service level agreements to precisely define the service delivery and quality. 2. Penalty clauses for supplier nonperformance. These clauses do not typically compensate for a customer’s loses, but rather are designed to ensure supplier attention to critical services. 3. Confidentiality clauses to protect customer’s data and information. In the case of public sector outsourcing, citizens are guaranteed that their privacy will not be invaded. 4. Named contract managers to ensure approved supplier staffing of key roles. 5. Intellectual property right clauses to protect the customer’s intellectual property. 6. Warranty clauses on hardware and software to protect assets. 7. Liability and indemnity clauses to hold the supplier accountable in cases of breach of contract or negligence. 8. A force major clause to protect customers in the case of natural disasters, civil wars, strikes, or other causes of work stoppages. 9. Specified arrangements for adapting the contract in the future to help parties adjust to changes in volumes, business requirements, or technology. 10. Early termination provisions to ensure the supplier will help the customer transition to another sourcing option upon contract termination.

Survey of IT Outstanding Experiences 171

Table 9: Number of stakeholders involved in defining contracts One Stakeholder Group Two Stakeholder Groups Three Stakeholder Groups Four Stakeholder Groups Five Stakeholder Groups Six Stakeholder Groups All 7 Total

US 3 11 5 9 3 3 0 34

US% 9% 32% 15% 26% 9% 9% 0% 100%

UK 22 11 5 4 1 1 0 44

UK% 50% 25% 11% 9% 2% 2% 0% 100%

Total 25 22 10 13 4 4 0 78

Total % 32% 28% 13% 17% 5% 5% 0% 100%

Table 10: Percentage of negotiations involving stakeholders US 34 26 13 13 12 10 1

IT Director Internal Lawyers External Lawyers Senior Executive Consultant Supplier Board

US% 100% 76% 38% 38% 35% 29% 3%

UK 30 24 9 12 2 6 3

UK% 68% 55% 20% 27% 5% 14% 7%

Total 64 50 22 25 14 16 4

Total % 82% 64% 28% 32% 18% 21% 5%

Table 11: Contract completeness

1-2 Clauses Defined 3-4 Clauses Defined 5-6 Clauses Defined 7-8 Clauses Defined 9-10 Clauses Defined Total

US 1 4 14 6 9 34

US% 3% 12% 41% 18% 26% 100%

UK 1 5 4 14 16 40

UK% 3% 13% 10% 35% 40% 100%

Total 2 9 18 20 25 74

Total% 3% 12% 24% 27% 34% 100%

Thirty-four US and 40 UK people responded. To assess contract completeness, we counted the number of clauses typically included in IT outsourcing contracts (see table and chart “Contract Completeness”). In general, UK contracts are more complete than US contracts, which is surprising given the previous finding that the US involves more stakeholders, particularly outside experts, to help define contracts. However, the number of clauses may still not translate into comprehensiveness within each clause, which is where use of a wider group of experts, as in more US organizations, may be making a difference. Nevertheless, the fact that so many US and UK organizations are not focusing on some important contractual clauses must be a common matter for concern.

CONTRACT CLAUSES Finding: The most common contractual clauses are confidentiality clauses (95%) and service level agreements (88%). Most participants have very detailed contracts with confidentiality clauses (95%), specific service level agreements (88%), early termination provisions (84%), and liability

172 Lacity & Willcocks

Table 12: Percentage of contracts with contract provisions Confidentiality Clauses Service Level Agreements Early Termination Clauses Liability and Indemnity Clauses A Force Major Clause Adapting to Change Clauses Intellectual Property Right Clauses Non-performance Penalty Clauses Named Contract Managers Warranty Clauses

US 30 29 29 25 15 24 23 21 15 16

US% 88% 85% 85% 74% 44% 71% 68% 62% 44% 47%

UK 40 36 33 36 34 24 24 25 28 23

UK% 100% 90% 83% 90% 85% 60% 60% 63% 70% 58%

Total 70 65 62 61 49 48 47 46 43 39

Total% 95% 88% 84% 82% 66% 65% 64% 62% 58% 53%

and indemnity clauses (82%) (see table “Percentage of Contracts with Contract Provisions”). We do note some systematic weaknesses, however. Forty-two percent do not name the representatives who will serve as account managers on the supplier side. These key roles must be filled by dynamic individuals with strong leadership, business, negotiating, and problem-solving skills. In a previous study, naming representatives in the contract was cited as a key success factor to the relationship (Lacity and Hirschheim, 1993). Thirty-eight percent of respondents do not include penalty clauses for supplier non-performance. Thus, there are no contractual remedies sought if suppliers fail to deliver on the service levels. Onethird of respondents generally do not include a force major clause to cover disasters such as civil wars, earthquakes, or strikes. Although the odds of a force major are small, the consequences are potentially devastating.

CONTRACT DURATION For each activity outsourced, we asked respondents to indicate the number of years of the contract. Again, the length of a contract is deemed important because prior research has found that customers cannot typically define IT requirements past a three-year time horizon. Contracts longer than three years in duration may become obsolete. If the relationship has power asymmetries, the weaker partner may suffer financial consequences of change (Lacity and Willcocks, 1998). The average US contract is nearly four years long, and the average UK contract is 4.7 years (see table “Contract Duration”). This is a year longer than recommended in the US, and nearly two years longer than recommended in the UK. A two-tailed t-test assuming homoscedasticity (p=.273) and heteroscedasticity (p=.266) indicates there is not a significant difference in sample means. The longest contract signed was a 20-year contract for mainframe operations at a UK public library. The respondent rated the supplier’s performance a “3” (very poor) on a scale of 1 to 10. The respondent cited deteriorating service, lack of supplier responsiveness to client needs, poor supplier staffing of the contract, and unforeseen additional costs not specified in the contract as serious problems. Clearly, the customer’s expectation that a 20year contract could articulate requirements in the long run was unrealistic. The shortest contract duration was six months, used by three customers. These three respondents rate supplier performance an “8-9,” “8,” and a “6.” Overall, short time horizons can generally realize expected benefits more frequently than long-time horizons and may thus lead to higher supplier performance ratings.

Survey of IT Outstanding Experiences 173

Table 13: Contract duration US 3.968 3.190 0.5 15 47

Mean Stand Dev. Minimum Maximum Number of Contracts

UK 4.6779 3.483 0.5 20 68

Overall 4.388 3.384 0.5 20 115

Figure 2: Contract duration in US and UK Contract Duration in US and UK

Number of contracts

25

19%

18%

20

15% 13%

15

8%

7%

10

6%

6%

3%

5

2% 1% 0%

0 .5-1

2

3

4

5

6

7

8

9

10

11

12

0%

0%

13

14

1%

15

1%

20

Number of Years

BENEFITS VERSUS ACTUAL BENEFITS Respondents were asked to check all anticipated benefits and actual benefits of outsourcing from a list of 13 possible benefits. Thirty-five US organizations and 44 UK organizations responded to this question. Expected Benefits. The number one expected benefit was cost reduction (67%). Of the respondents expecting cost savings, two-thirds expected only some cost savings, while onethird expected significant cost savings (see table “Expected Benefits” and chart “Percentage of Respondents Expecting Benefits from IT Outsourcing”). Other commonly expected benefits included better quality service (61%), access to scarce IT skills (56%), and improved flexibility of IT (51%) . It is also interesting to note that very few respondents expected that outsourcing would provide access to new IT (24%). This generally indicates that IT suppliers take over the customer’s in-house functions rather than develop/infuse new IT. And only 15% expected outsourcing to help their cash flow problems. This reason was commonly found in total outsourcing cases (Lacity and Hirschheim 1993), but selective outsourcing lacks the critical mass to significantly affect income statements.

174 Lacity & Willcocks

Other research has also identified cost savings as the major reason companies outsource. Collins and Millen (1995) received 110 questionnaires from CIOs of the top Fortune 500 companies and found that the number one stated reason for outsourcing was headcount cost savings. Willcocks and Fitzgerald (1994) found that “some cost reduction” was the number one ranked anticipated benefit of outsourcing. Although outsourcing is often cited as a tool for focusing on core competencies, it is clear that cost considerations supersede a focus strategy for most organizations. Actual Benefits. What benefits did customers actually receive from outsourcing? In total, over half of respondents achieved cost savings (see table “Actual Benefits”). But of those who achieved cost savings, 69% achieved some cost savings rather than significant cost savings. The second most commonly cited benefit was the refocusing of in-house IT staff on more value-added IT work. Given that most respondents outsource infrastructure and operations, perhaps the IT staff was able to focus more on business applications. The third most commonly realized benefit was improved IT flexibility (41%). This often occurs because IT costs become more variable with outsourcing, thus contract prices adapt to volume fluctuations more easily than insourcing, in which most costs are fixed. Better quality service was the forth most commonly realized benefit, occurring in 39% of responding organizations. It is noticeable that more US organizations are getting benefits on the more specifically IT targets, while more UK organizations are getting relatively better benefits on financial and business dimensions. It is not clear why this is the case, though it may reflect where client effort was actually placed during post-contract management. Expected versus Actual Benefits. There is a noticeable gap between “Expected Benefits” and “Actual Benefits.” Specifically, among the list of all 13 benefits, expected benefits always exceed the actual benefits. Previous research has also generally indicated a gap in expectations (Chapman and Andrade, 1997). Prior to the signing of a contract, both parties are typically enthusiastic and optimistic. Suppliers have often noted how difficult it is to foster realistic expectations among customers (See November 1998 edition of Infoserver: The Journal for Strategic Outsourcing Information on http://www.infoserver.com). One current IT outsourcing customer notes: Table 14: Expected benefits EXPECTED BENEFITS Cost Reduction Some Cost Reduction Significant Cost Reduction Better Quality Service Access to Scarce IT Skills Improved IT Flexibility Refocus In-house IT Staff Improved Use of IT Resource Focus on Core Business Improved Business Flexibility Better Management Control Access to New IT Balanced Processing Loads Assist Cash Flow Problems

US 23 17 6 20 18 17 17 16 13 11 9 6 7 4

US% 66% 49% 17% 57% 51% 49% 49% 46% 37% 31% 26% 17% 20% 11%

UK 30 18 12 28 26 23 23 20 21 19 19 13 5 8

UK% 68% 41% 27% 64% 59% 52% 52% 45% 48% 43% 43% 30% 11% 18%

US & UK 53 35 18 48 44 40 40 36 34 30 28 19 12 12

UK & UK % 67% 44% 23% 61% 56% 51% 51% 46% 43% 38% 35% 24% 15% 15%

Survey of IT Outstanding Experiences 175

“I’ve actually been an outsourcer before. I’ve been on a bidding side. I understand the challenges a supplier faces. There is generally an expectation of management on the user side that here is this knight in shining armor, I’ll get three times better service at half the price. And also what happens, that expectation grows as you get closer to contract, so you have this large gap in expectations from the start.” — Contract Administrator, Public Sector Organization However, despite the gap in expected and actual benefits, the next finding shows that customers gave their suppliers good performance ratings overall.

CUSTOMER RATING OF SUPPLIER PERFORMANCE Finding: Overall, customers’ mean rating on 113 contracts indicates a “good” performance by suppliers. UK respondents were asked to rate the supplier performance of their largest IT outsourcing contract, and US respondents were asked to rate the performance of their three largest contracts. A Likert scale from 0 to 10 was used, with 0 and 1 indicating poor performance, 2 to 4 indicating satisfactory performance, 5 to 7 indicating good performance, and 8 to 10 indicating excellent performance. Seventy-six respondents rated 113 contracts. The overall mean response is 6.47, indicating that respondents are generally pleased with the performance of their outsourcing suppliers (See table and chart, “Customer Rating of Supplier Performance”). Five respondents rate the performance of their suppliers as a perfect 10. Only 7 respondents give suppliers a “0” performance rating. T-tests assuming Figure 3: Percentage of respondents expecting benefits from IT outsourcing 60 67%

50

61% 56% 51%

40

51% 46%

43% 38%

30

36%

15%

15%

Balanced processing loads

Assist cash flow problems

24%

20 10

Access to new IT

Better management control

Improved business flexibility

Focus on core business

Improved use of IT resource

Refocus in-house IT staff

Improved IT flexibility

Access to scarce IT skills

Better quality service

Cost reduction

0

176 Lacity & Willcocks

Table 15: Actual benefits Actual Benefit Cost Reduction Some Cost Reduction Significant Cost Reduction Refocus In-house IT Staff Improved IT Flexibility Better Quality Service Improved Use of IT Resource Access to Scarce IT Skills Improved Business Flexibility Focus on Core Business Better Management Control Access to New IT Balanced Processing Loads Assist Cash Flow Problems

US 14 12 2 13 16 14 15 15 9 10 6 9 4 2

US% 40% 34% 6% 37% 46% 40% 43% 43% 26% 29% 17% 26% 11% 6%

UK 28 17 11 22 16 17 16 16 16 15 11 7 4 6

UK% 64% 39% 25% 50% 36% 39% 36% 36% 36% 34% 25% 16% 9% 14%

Total 42 29 13 35 32 31 31 31 25 25 17 16 8 8

Total% 53% 37% 16% 44% 41% 39% 39% 39% 32% 32% 22% 20% 10% 10%

both homoscedasticity (p=.17) and heteroscedasticity (p=.16) indicate no significant differences between US and UK respondents. How do we reconcile the findings that show customers do not receive the level of anticipated benefits, yet report an overall good supplier report card? One plausible explanation is that managing information technology is a significant challenge regardless of the sourcing strategy. While customers may initially hope for radical improvement under a supplier’s management regime, customers may not be surprised when the supplier encounters the same obstacles, such as a scarcity of IT skills, pressures to provide excellent service with a limited budget, changing user specifications, etc. Indeed, one CIO for a US Petroleum company noted, “They bitched about IT before [outsourcing], they bitch now—but at least it’s costing me less.”

EXTENT OF PROBLEMS/ISSUES ENCOUNTERED WITH OUTSOURCING Finding: Overall, 76% percent of responses indicated no problems or only minor problems with outsourcing. In addition to assessing overall supplier performance, participants were asked to indicate whether they had experienced any strategic, cost, managerial, operational, contractual, or technical difficulties with IT outsourcing. In total, respondents were asked to comment on 24 potential problem areas. For each of these potential problems, respondents were asked to indicate whether it was a “continuing serious problem,” a “problem that was difficult to resolve,” a “minor problem,” or “no problem/not a difficulty.” Thirty-two US and 44 UK people responded (See table “Extent of Problems/Issues Encountered with Outsourcing”). In total, 76 respondents commented on 24 potential problem areas, yielding a total of 1,824 check-marks. Overall, only 8% of check-marks indicated “continuing serious problems,” 17% indicated “problems that were difficult to resolve,” 33% indicated “minor problems” and 44% indicated “no problems/not a difficulty.” Below we discuss some of the more serious

Survey of IT Outstanding Experiences 177

problems encountered with outsourcing. It is important to keep in mind that overall, 76% percent of responses indicated no problems or only minor problems with IT outsourcing. At the same time, it is clear that sizeable minorities of organizations really do have serious issues in certain areas. Strategic Issues. Considering all five potential strategic problem areas, the vast majority of responses (79%) indicate no strategic problems or only minor strategic problems. Of the few respondents that do experience serious strategic problems, 11 complain that the supplier does not understand their business, 7 complain that the corporate strategy is no longer aligned with the IT strategy, and 4 complain of poor strategic planning for IT. Cost Issues. Considering three possible cost problems, 36% of responses indicate no cost problems and 39% of responses indicate only minor cost problems. However, 23 (31%) Figure 4: Percentage of respondents actualy ahcieving benefits from IT outsourcing

50 53%

40

44% 41% 39%

30

39%

39% 32%

32% 22%

20

20%

10

10%

10%

Table 16: Customer rating of supplier performance Mean Standard Deviation Number of Contracts Rated

US 6.72 2.57 68

UK 6.09 2.11 45

Overall 6.47 2.42 113

Assist cash flow problems

Balanced processing loads

Access to new IT

Better management control

Improved business flexibility

Focus on core business

Improved use of IT resource

Access to scarce IT skills

Better quality service

Improved IT flexibility

Refocus in-house IT staff

Cost reduction

0

178 Lacity & Willcocks

responses indicate severe/difficult problems with services beyond the contract. And 20 (26%) responses characterize “cost escalation due to contract loopholes,” as serious or difficult problems, indicating a weakness in cost implications of contracts. For example, several organizations did not realize they were accountable for software license fees, which can eat away hundreds of thousands of dollars—even millions—in expected savings. Sixteen (22%) respondents experience “difficulties in monitoring/controlling costs” as severe/difficult to resolve. Thus, some respondents may not be able to track whether cost savings are realized or not. These cost findings may provide insight into why 67% expected cost savings but only 53% achieved cost savings. Perhaps expected cost savings are not always realized because customers fail to fully define/understand requirements in the contract. Any additional service not defined in the contract will typically trigger excess fees, thus deteriorating cost savings. Managerial Issues. Once again, most managerial problems are only of a minor nature. Seventy-two percent of responses indicate no managerial troubles or only minor managerial troubles. However, 32 (42%) responses indicate that the supplier does not properly staff the contract, causing serious or difficult problems. Other serious or difficult problems are attributed to the customer, such as 21 responses (28%) complaining that the in-house staff was resistant to outsourcing and that in-house managerial skills to manage the contract are inadequate (21 responses; 28%). Operational Issues. Overall, most respondents experience no operational problems (34%) or only minor problems (40%). Twenty-nine responses (38%) indicate serious or difficult problems attributable to service level definitions. This is surprising, given that 88% have service level agreements defined in the contract. However, Chapman and Archade (1997) found that “today’s metrics don’t provide the information needed to identify and resolve problems” (p. 47). Contractual/Legal Problems. Due to differences in the US and UK surveys, we analyzed the only contractual issue addressed in both surveys—defining the contract. The majority of respondents cite only minor (33%) or no difficulties (42%) with defining the contract. In the lessons learned section, many respondents cited a well-defined contract as critical to the success of the relationship: “Get fixed pricing and thoroughly define deliverables.” —Director of Computing and Communication Services “[Have] specific service levels.”—Manager of IT Services “[Have] specific service levels, define the baseline, and have an out clause.”—CIO “Clearly define objectives and detailed contractual arrangements.”—Corporate Manager of Corporate Systems “Ensure performance measures are clear and calculable.”—Manager of IS Administration “Get good detailed help—legal, metrics, and consultants. Keep senior management out of all but the final negotiations.” — Project Manager “You need measurable milestones, performance incentives and penalties, and a good contract manager.—Project Manager

Survey of IT Outstanding Experiences 179

“Contract performance specs.”—VP of Information Services “Contracts must be customer focused with rewards and penalties linked to customer satisfaction.”—Director of IT Services “Understand what you want to accomplish with the contract. Move slowly to start with—pilot first before you give away the keys.”—CIO And one dissenting comment: “Flexibility is the key to success.”—CIO Technical Issues. Most respondents have minor (24%) or no technical troubles (54%). But 24 (32%) complain of a shortage of supplier’s IT skills. On the open-ended questions, several respondents reiterated their complaints of poor supplier skills and staffing of the contract: “The vendor lacked technical knowledge.”—UK Head of IS, Financial Services Company “Should have ascertained staffing commitment from the vendor.”—UK Manager of IT, Financial Services Company “The vendor did not understand our business.”—UK Head of IS, Financial Services Company “Watch out for poor talent assigned to your project.”—US Director of Information Services “Beware of high turnover of vendor staff.”—US Manager of Information Resources “Suppliers gave us IT expertise, but they are too driven to make [a] profit.”—no title provided Twenty-three percent complain outsourcing led to duplicate systems. Only 17% complain that the supplier recruits inexperienced staff. Overall, we would expect any IT activity—whether it was insourced or outsourced— to have at least some problems. When problems do occur, they stem from both the customer and supplier side. Although most problems are only of a minor nature, this question provides opportunities for improving the success of IT relationships (see section Managerial Implications of Findings).

REDUCTION IN IT STAFF Finding: 63% of companies reduced IT headcount as a consequence of IT outsourcing. We asked respondents whether outsourcing led to reduced IT staff, and if yes, what percentage among several methods of staff reduction tactics was used. Of the 75 respondents that have outsourced and answered this question, 47 (63%) resulted in a reduction of IT staff. US and UK results are nearly identical (see table “Percentage of Organizations in Which IT

180 Lacity & Willcocks

Figure 5: Customer rating of vendor performance Customer Rating of Vendor Performance 113 Contracts Rated Mean Response = 6.47 30

Number of contracts

25%

25 19%

20 14%

14%

15 10 5%

5

4% 2%

6% 4%

4%

2%

0 0 1 Poor

2

3 4 5 Satisfactory

6

7 Good

8

9 10 Excellent

Outsourcing Reduced Staff”). Given that IT salaries are the largest portion of IT budgets, one would expect staff reductions were required to realize expected cost savings. Nearly a tie for the most common method of IT staff reduction were transferring staff to the supplier (32%) and natural attrition/termination (i.e., redundancy) (31%). Internal staff transfer was the least commonly used method (17%) (see table “Percentage of Organizations Using Each Method to Reduced IT Staff”). Most respondents (70%) report no problems with transferring staff to the supplier. Thirty percent, however, reported difficulties. For example, a UK Assistant Group Director for a transportation company transferred 11% of his in-house staff when he outsourced almost all of the IT infrastructure. He cited the following problems: “Not all the staff wanted to go. Pension rights were not equal. Employees did not want to change locations.” One UK IT Director for a large record company wrote that “outsourcing is generally bad for morale. That hurt us because our current systems are too people-specific because we have poor in-house documentation.”

REASONS FOR REJECTING OUTSOURCING Finding: The most common reason for rejecting outsourcing is expense (23%). Respondents were asked whether their organizations had ever considered outsourcing, but rejected the option. Twenty-two US organizations and 44 UK organizations responded to this question. Respondents reported on 125 instances of rejecting outsourcing. We also asked respondents to indicate the reason(s) they rejected outsourcing. (See table “Reasons for Rejecting Outsourcing”).

Survey of IT Outstanding Experiences 181

Table 17: Extent of problems/issues encoutered with outsourcing S trategic Issues: 1. Supplier’s lack of understanding of your business 2. Failure to align corporate strategy with IT strategy 3. Poor strategic planning of IT 4. Defining intellectual property rights 5. Defining data protection procedures Total # of Strategic Issues Cost Issues: 6. Cost escalation as a result of contract loopholes 7. Difficulties in controlling/monitoring costs 8. Costs for additional services beyond contract Total # of Cost Issues Managerial Issues: 9. Loss of control over IT operations 10. In-house staff resistance to outsourcing 11. Poor supplier staffing of contract 12. M anagerial skills shortage 13. Lack of supplier training for staff Total # of M anagerial Issues Operational Issues: 14. Getting different contract suppliers to work together 15. Defining service levels 16. Coordinating IT work with supplier 17. Communication with supplier 18. Lack of supplier responsiveness to client needs 19. Deteriorating service Total # of Operational Issues Contractual Issues: 20. Defining the outsourcing contract Technical Issues: 21. Supplier failure to upgrade IT 22. Duplication of systems 23. Policy to recruit inexperienced IT staff 24. IT skills shortage affecting supplier’s service Total # of Technical Issues Overall Total Overall Percentage

Serious Problem

Difficult Problem

Minor Problem

Not a Problem

11 7 4 0 2 24 6%

14 12 15 8 5 54 14%

30 18 16 23 27 114 30%

21 39 41 45 42 188 49%

76 76 76 76 76 380

10 8 8 26 11%

10 8 15 33 14%

31 34 23 88 39%

25 26 30 81 36%

76 76 76 228

3 8 9 5 5 30 8%

8 13 23 16 10 70 18%

24 22 21 29 19 115 30%

41 33 23 26 42 165 42%

76 76 76 76 76 380

9 6 2 2 8 7 34 7%

14 23 14 10 18 6 85 19%

27 33 39 33 23 28 183 40%

26 14 21 31 27 35 154 34%

76 76 76 76 76 76 456

7 9%

12 16%

25 33%

32 42%

76

3 5 3 6 17 6% 138 8%

9 13 10 18 50 16% 304 17%

20 16 16 20 72 24% 597 33%

44 42 47 32 165 54% 785 43%

76 76 76 76 304

Total

1824

The top reason for rejecting outsourcing is that the outsourcing option was deemed more expensive than insourcing (23%). This finding makes sense because respondents cite cost savings as the number one expected IT outsourcing benefit. Thus, if cost savings can not be expected, outsourcing is likely to be rejected. The next most common reason for rejecting outsourcing is that a supplier appeared to provide no additional benefits over insourcing (19%). Presumably, such benefits would not be limited to costs, but to service as well. Few respondents (6%) reject outsourcing because they could not find a suitable supplier. The outsourcing market has clearly matured over the last decade, with many small niche suppliers as well as mega-suppliers offering services. And only 1 respondent indicates that his organization has a policy against outsourcing.

182 Lacity & Willcocks

Respondents were also invited to provide additional reasons/comments not listed on our survey. Only two respondents wrote comments: “The supplier was unable to convince us that they could meet our service standard.”—US Director, IT Enterprises “Potential of IT department to provide strategic capability.”—US Senior Financial Analyst for IT Financing

CANCELED CONTRACTS Respondents were asked whether they canceled any IT outsourcing contracts in the past five years. If they did, we asked about the outcome of the cancellation. Of the 79 people who responded to this question, 32% of organizations have canceled one or more IT contracts. Across countries, the United Kingdom has a higher percentage (41%) of organizations that have canceled contracts compared to the United States (20%) (see table “Number of Organizations That Have Canceled Contracts”.) The seven US organizations canceled a total of 15 contracts and the 18 UK organizations canceled a total of 20 contracts. The most common outcomes were to change suppliers (51%) and bring the activity back in-house (34%). The least common outcomes were litigation (3%) and renegotiation (11%). These results are somewhat contrary to other findings. The Gartner Group, based on a survey of 250 CIOs, estimates that 75% of all IT outsourcing customers will renegotiate their deals rather than bring the activity back in-house. Of this 75%, Gartner expects 10% of customers to terminate their contracts early and 20% to switch suppliers (Caldwell and Table 18: Percentage of organizations in which IT outsourcing reduced staff Number of Organizations That Reduced IT Staff as a Consequence of Outsourcing Number of Organizations That Did Not Reduce IT Staff as a Consequence of Outsourcing Total Number of Respondents

US

US%

UK

UK%

Overall

Overall %

21

62%

26

63%

47

63%

13 34

38% 100%

15 41

37% 100%

28 75

37% 100%

Table 19: Percentage of organizations using each method to reduce IT staff Method of Reducing Staff Transfer to Supplier Redundancy (Attrition and Termination) Replace In-house Staff with Supplier Staff Internal Staff Transfer Total

US 13

US% 29%

UK 14

UK% 36%

Overall 27

Overall % 32%

17

38%

9

23%

26

31%

8 7 45

18% 16% 100%

9 7 39

23% 18% 100%

17 14 84

20% 17% 100%

Survey of IT Outstanding Experiences 183

Table 20: Extent of problems with transfer of IT staff No problems transferring staff Yes, problems transferring staff Total number of respondents

US 9 4 13

US% 69% 31% 100%

UK 7 3 10

UK% 70% 30% 100%

Overall 16 7 23

Overall % 70% 30% 100%

McGee, 1997; McGee, 1997). Experts argue that renegotiation is preferred over switching suppliers or terminating the contract: “When the contract no longer fits the users’ needs, both sides will sit down and renegotiate the contract. Very few users exercise the termination of the contract; they’re too dependent on the outsourcer.”—Harry Glasspiegal, partner at Shaw, Pittman, Potts, and Trowbridge

DISCUSSION Overall, US and UK respondents are pleased with IT outsourcing: • Respondents rated overall supplier performance as “good” • Respondents mostly realized the benefits they expected • Respondents characterized the majority of problems as “minor” in nature. The healthy IT outsourcing report card is likely explained by the scope and type of IT outsourcing practiced by responding organizations. The vast majority of respondents in both countries pursue selective outsourcing rather than total outsourcing. Prior research has found that most companies are successful with their selective outsourcing strategies: • In a survey of 110 Fortune 500 companies, Collins and Millen (1995) found that 95% realized increased flexibility, 95% focused in-house staff on IT core competencies, 86% realized personnel cost savings, and 88% improved service. • Lacity and Willcocks (1998) studied 61 sourcing decisions, including total outsourcing, total insourcing, and selective outsourcing. That study found that 85% of selective outsourcing decisions were successful, whereas only 29% of total outsourcing decisions and 67% of total insourcing decisions were successful. The healthy report card may also be explained by the types of IT activities selected for outsourcing. Respondents in both countries generally targeted IT infrastructure activities— such as disaster recovery, mainframe operations, network management, midrange operations, PC support, and help desk operations—rather than IT development or IT strategy. Our findings are consistent with prior research. Chapman and Andrade (1997) found that the three most successful IT outsourcing activities are data processing operations, network management, and help desk functions. Grover et al. (1996) conducted two surveys (n=68 and n=188) and correlated the types of IT functions outsourced with perceived success. They found a high rate of perceived success associated with outsourcing systems operations and telecommunications, but outsourcing applications development and systems management “did not lead to increased satisfaction” (p. 103). Willcocks and Fitzgerald (1994) found that it was easier for participants to outsource “technically mature” activities. Customers understood how to cost and evaluate such activities and therefore could negotiate a sound contract. IT infrastructure, such as mainframe operations, networks, and telecommunications are often technically mature and may be successfully outsourced.

184 Lacity & Willcocks

Table 21: Reasons for rejecting outsourcing

Outsourcing Deemed Too Expensive In-house IT Able To Achieve Same Benefits Lack of Identifiable Benefits Concern About Loss of Operational Control I.T. Activity Deemed As Too Strategic No Suitable Supplier Was Found Company Policy Not To Outsource Total

US 14

US% 24%

UK 15

UK% 22%

Total 29

Total % 23%

13 11

22% 19%

11 12

16% 18%

24 23

19% 18%

10 7 3 0 58

17% 12% 5% 0% 100%

12 12 4 1 67

18% 18% 6% 1% 100%

22 19 7 1 125

18% 15% 6% 1% 100%

Table 22: Number of organizations that have canceled contracts US Number of Organizations That Have Canceled at Least 1 Contract 7 Number of Organizations That Have Never Canceled a Contract 28 Total Number of Respondents 35

US%

UK

UK%

Overall

Overall %

20%

18

41%

25

32%

80% 100%

26 44

59% 100%

54 79

68% 100%

Table 23: Outcome of canceled contracts Change of Supplier Brought Activity In-house Renegotiation Litigation Total Number of Canceled Contracts

US 5 8 2 0

US% 33% 53% 13% 0%

UK 13 4 2 1

UK% 65% 20% 10% 5%

Overall 18 12 4 1

Overall % 51% 34% 11% 3%

15

100%

20

100%

35

100%

Overall, the survey findings across the US and UK are similar. In addition to the similar supplier report cards and types of IT activities outsourced mentioned above, similar US and UK practices include: • Decision sponsors (typically IT director) • Contract negotiators (IT directors and internal lawyers) • Contract provisions (Confidentiality clauses, service level agreements, and early termination clauses) • Contract duration (4.388-year average) • Expected benefits (cost reduction, better quality service, access to scarce IT skills) • Supplier ratings (6.47 on a 10-point scale) • Extent of problems encountered (mostly minor or no problems reported)

Survey of IT Outstanding Experiences 185

• • • •

Percentages of IT outsourcing resulting in staff reduction (63% overall) Methods to reduce IT staff (transfer to supplier and redundancy) Percentage experiencing problems with transferring staff (30% overall) Reasons for rejecting outsourcing (too expensive; in-house staff can achieve same benefits) Because the US and UK samples showed only marginal statistical differences in terms of annual revenues and IT operating budgets, similar results may be based on the sample spreads selected for each country. But more likely, similar practices and outcomes may be a result of a shared Anglo-Saxon culture. Prior studies have found other IT-enabled management practices such as business process reengineering (Willcocks & Currie, 1997), client server implementations (Subramanian & Lacity, 1997), and system development practices (Taylor-Cummings & Feeny, 1997) to be similar in the US and UK. One explanation for the similarities in US and UK findings is an institutional isomorphic effect where outside experts seed client organizations with similar standards and methods. In the IT outsourcing arena, for example, Technology Partners (a Houston-based outsourcing consulting firm), Shaw & Pittman, or Millbank & Tweed (US-based IT outsourcing legal firms) participated in 3 out of 4 US and UK billion dollar contracts studied. Furthermore, the Gartner Group and Compass were commonly used as the benchmarking firms to help assess IT performance (Lacity and Willcocks, 1999). In addition, three out of four US and UK organizations studied by Lacity and Willcocks (1999) belong to the International Information Technology Users Group (IITUG), which provides many opportunities for information exchange. In this way, organizational learning is transferred across organizations, and practices are quickly disseminated among US and UK organizations. However, some US and UK differences were noted and explained throughout the paper: • US respondents (8%) totally insourced less frequently than UK respondents (30%). • US respondents (31%) outsource midrange computing less frequently than UK respondents (73%). • US respondents (63%) outsource help desks more frequently than UK respondents (32%). • US respondents (29%) use a single supplier more frequently than UK respondents (9%). • US respondents (9%) use only 1 stakeholder to define contracts less frequently than UK respondents (50%). • US respondents (44%) include 7 to 10 important contract clauses less frequently than UK respondents (75%). • US respondents (20%) have cancelled IT outsourcing less frequently than UK respondents (41%). • US respondents (33%) changed suppliers less frequently than UK respondents (65%). Many of these differences may well be due to one or combination of factors, the main ones being–a more matured approach to outsourcing in the USA, larger outsourcing deals, and relatively larger organizations represented in the sample. As we plan to replicate this survey every three years, we will be able to track how IT outsourcing practices and outcomes evolve over time and across countries.

MANAGERIAL IMPLICATIONS OF FINDINGS Despite the overall good report card on IT outsourcing, there are opportunities for improvement. Several respondents pointed to key areas they would improve in the future. Below are lessons for customers and suppliers.

186 Lacity & Willcocks

Lessons for Customers: What Needs to Be Improved? 1. Should more people be involved in evaluation? In 58% of organizations, only 1 stakeholder makes sourcing decisions. Case study research has found that successful sourcing decisions require a mix of political power and technical skills. Political power helps to enforce the larger business perspective—such as the need for organization-wide cost cuts—as well as the “muscle” to implement such business initiatives (Lacity and Willcocks, 1998). Technical expertise on IT services, service levels, measures of performance, rates of technical obsolescence, rates of service growth, price/performance improvements, and a host of other technical insights are needed to develop requests-for-proposals and to evaluate supplier bids. The question is, how many stakeholders are needed to cover these skillsets? The survey found that IT managers are the sole sponsors of 47% of outsourcing decisions. Because most respondents in this survey pursue selective outsourcing on a piecemeal basis, IT managers may have enough power, authority, and knowledge to initiate and sponsor sourcing decisions without involving other stakeholders. In 11% of responding organizations, only the board, senior executive, or steering committee makes the decision. Most research agrees that these stakeholders should not be the sole evaluators. Lacity and Willcocks (1998) found that senior executives realized their expected benefits only 40% of the time when they outsourced IT without IT’s input. Chapman and Andrade (1997) note: “Several major outsourcers make a point of marketing their services only to executive management. They are introduced to the MIS organization only after upper management has decided to do the outsourcing. When they get to MIS, the outsourcers are not offering a service; they are implementing a fiat. How can an outsourcer know more about a client’s business needs than the internal MIS organization? They can’t.” (p. 134) 2. Better communications with in-house IT staff during evaluation. A common problem was in-house staff resistance to outsourcing (43%). Outsourcing evaluations almost always affect IT staff moral. When senior managers are evaluating IT outsourcing, they often debate whether they should inform the IT staff. Lacity and Hirschheim (1995) found that IT staff always found out, anyway. Lack of formal communication bred distrust and often premature departure of the most talented staff. What can be done? Lacity and Hirschheim (1995) have shown that inviting the IT staff to submit an internal bid was an effective way to prevent staff resistance. Internal bids can be a catalyst for creativity and a galvanizing force among employees. The staff focuses on the task of improving IT to compete with external supplier bids rather than on opposing management’s evaluation. And often times, internal bids proved superior to external bids. 3. Better defined contracts. The survey indicates that 25% of customers experienced trouble defining contracts. And certainly the barrage of open-ended responses indicates a need for better contracts, particularly better service level agreements. Perhaps organizations might benefit from the use of more external consultants and lawyers to help define contracts, given that only 28% of respondents hire external lawyers and only 18% use external consultants (see also Willcocks and Kern, 1998). 4. Be careful of outsourcing too many key people. One of the most common customer complaints is poor supplier staffing of the contract (42% experiencing severe or difficult problems). Customers must realize that when they outsource, they lose control over human

Survey of IT Outstanding Experiences 187

resources. Transferred employees now work for the supplier, and the supplier has a right to assign employees as they see fit. To protect themselves, customers can demand clauses that focus on outputs, such as productivity and quality measures, or inputs, such as requiring a certain skillset, a specified number of years experience, and specified training for different job categories. 5. Better post-contract management infrastructure. The survey also indicates that customers may not create a proper management infrastructure to support the supplier relationship. Customers claimed they had severe or difficult trouble coordinating work among multiple suppliers (30%) and lack of in-house managerial skills (28%). Organizations must plan to retrain or recruit people to properly manage contracts, coordinate user demand and supplier supply, resolve conflicts, monitor supplier performance, etc. (See Willcocks and Feeny [1997] and Feeny and Willcocks [1998] for nine core IT capabilities that cannot be outsourced.)

Lessons for Suppliers: What Needs to Be Improved? 1. The key lesson for suppliers is to help their customers set realistic expectations. The survey found that customers expected more benefits from IT outsourcing than were achieved. Although suppliers focus on the advantages of outsourcing to win contracts, suppliers need to help customers be realistic. Suppliers typically cannot reduce their customer’s IT costs while radically improving their service, although customers apparently expect this. When customers ask suppliers, “Can you do X?” suppliers need to do more than say yes—suppliers need to explain what doing X would require in terms of costs and other resources. 2. Explain staffing policies to the customer. Forty-two percent of respondents complained about poor supplier staffing. Many customers may have unrealistic expectations because they assume suppliers can obtain scarce IT skills for less money. But if the supplier pays an SAP expert $120,000 per year, they must certainly charge the customer a proportional amount for the expert’s time. Suppliers need to explain who will staff what functions on the account. Will new hires work on the account? What training will be given to transferred employees? How many years of experience are required to fill each job category? Will industry experts or technical experts be moved from other accounts to this account? Does an infusion of scarce IT skills or emergency IT skills trigger customer excess fees? 3. To help ensure the success of the contract, suppliers should help make sure the customer has a proper infrastructure to manage and coordinate the contract. Suppliers need to help the customers create an environment for success. This includes agreeing on how to monitor supplier performance, explaining how customers are charged for services, and showing customers what reports they will receive. Suppliers and customers must work together to define how users will communicate with supplier employees—do they call them directly? Do they use a liaison? Customer contract managers must understand how user requests will affect volumes and subsequently affect costs.

CONCLUSION The sourcing market for information technology continues to grow and evolve. The more pervasive (but less audible or visible) trend continues to be the selective outsourcing of information technology to multiple suppliers for a specified subset of IT activities. This

188 Lacity & Willcocks

survey found that selective outsourcing, particularly of IT infrastructure activities, is generally successful. The benefits of selective sourcing included: • cost reduction (53%) • refocus of in-house staff (44%) • improved IT flexibility (41%) • better quality service (39%) • access to scarce IT resources (39%) Why is selective outsourcing successful? Information technology spans a variety of activities in terms of business contribution, integration with existing processes, and level of technical maturity. Such diversity often demands tailored solutions. Typically, no one supplier or internal IT department possesses the experience and economies of scale to perform all IT activities most effectively. While some activities, especially stable IT activities with known requirements, may be easily outsourced, other IT activities often require much management attention, protection, and nurturing to ensure business success. However, selective sourcing does have some downsides, which include: 1) increased transaction costs associated with multiple evaluations, multiple contract management; and 2) potential lack of integration, cooperation, and coordination among multiple sources. The findings on selective outsourcing are consistent with prior studies. Despite the media attention on mega-deals and new multi-billion dollar contracting options such as “value-added” deals, cosourcing, and equity exchange-deals, surveys continue to show that IT sourcing decisions are targeted in scope and are driven by cost considerations. These practices—while warranting less press coverage—have certainly fueled the ever expanding market for global IT services.

ENDNOTE 1

Readers interested in the survey instrument are asked to contact the authors directly.

REFERENCES Apte, U., Sobol, M., Hanaoka, S., Shimada, T., Saarinen, T., Salmela, T. and Vepsalainen, A. (1997). IS outsourcing practices in the USA, Japan, and Finland: A comparative study. Journal of Information Technology,12(4), 289-304. Arnett, K. and Jones, M. (1994). Firms that choose outsourcing: A profile. Information & Management, 26, 179-188. Caldwell, B. (1996). The new outsourcing partnership. InformationWeek, 24 June, 585, 50-64. Caldwell, B. and McGee, M.(1997). No big savings—Too many outsourcing deals don’t pay off as expected. InformationWeek, 10 March, Issue 621, TechWeb News. Chapman, R. and Andrade, K. (1997). Insourcing After Outsourcing. New York: Amacom. Collins, J. and Millen, R. (1995). Information systems outsourcing by large American industrial firms: Choices and impacts. Information Resources Management Journal, 8(1), 5-13. Currie, W. and Willcocks, L. (1997). New strategies in IToOutsourcing. Business Intelligence. London. Dekleva, S. (1994). CFOs, CIOs, and outsourcing. Computerworld, 28(20), 96.

Survey of IT Outstanding Experiences 189

Feeny, D. and Willcocks, L. (1998). Core IS capabilities for exploiting information technology. Sloan Management Review, 39(3), 9-21. Foley, A. (1993). Hong Kong busks Asia services trend. ComputerWorld, May, 13(1), 56. Hong Kong. Grover, V., Cheon, M. and Teng, J. (1994).A descriptive study on the outsourcing of information systems functions. Information & Management, 27, 33-44. Grover, V., Cheon, M. and Teng, J.(1996). The effect of service quality and partnership on the outsourcing of information systems functions. Journal of Management Information Systems, 12(4), 89-116. Halvey, J. and Melby, B.(1996). Information Technology Outsourcing Transactions: Processes, Strategies, and Contracts. New York: John Wiley & Sons. Lacity, M. and Hirschheim, R. (1993). The information systems outsourcing bandwagon. Sloan Management Review, 35(1), 73-86. Lacity, M. and Hirschheim, R. (1995). Beyond the Information Systems Outsourcing Bandwagon: The Insourcing Response. Chicester: John Wiley & Sons. Lacity, M. and Willcocks, L. (1996). Best practices in information technology outsourcing. Oxford University Executive Briefing, Templeton College, Oxford University. Lacity, M. and Willcocks, L. (1998). An empirical investigation of information technology sourcing practices: Lessons from experience. MIS Quarterly, September, 22(3), 363-408. Lacity, M., Willcocks, L. and Feeny, D.(1995).IT outsourcing: Maximizing flexibility and control. Harvard Business Review, May-June, 84-93. Lacity, M., Willcocks, L. and Feeny, D. (1996). The value of selective IT sourcing. Sloan Management Review, Spring, 37(3), 13-25. Lacity, M. and Willcocks, L. (1999). Global Information Technology Outsourcing: In Search of Business Advantage, Unpublished Manuscript. McGee, M. (1997). Measuring outsourcing’s ROI. InformationWeek, Issue 641, 28 July, TechWeb News. Menagh, M. (1995). Driving a hard bargain. Computerworld, 29(32), 69-76. Sobel, M., and Apte, U. (1995). Domestic and global outsourcing practices of America’s most effective IS users. The Journal of Information Technology, 10(4), 269-280. Subramanian, A. and Lacity, M. (1997). Managing client server implementations: Today’s technology, yesterday’s lessons. Journal of Information Technology, 12(3), 169-186. Taylor-Cummings, A. and Feeny, D. (1997). The development and implementation of systems: Bridging the user-IS gap. In Managing IT as a Strategic Resource, 169-198. New York: McGraw Hill. Willcocks, L. and Currie, W. (1997). Does radical re-engineering really work? Emerging issues in strategic projects. In Managing IT as a Strategic Resource, 238-270. New York: McGraw Hill. Willcocks, L. and Feeny, D. (1997). Core capabilities in IT. Oxford Institute of Information Management Working Paper, Templeton College Oxford University, England. Willcocks, L. and Fitzgerald, G. (1994). A Business Guide to Outsourcing Information Technology: A Study of European Best Practice in the Selection, Management and Use of External IT Services. Business Intelligence, London. Willcocks, L. and Kern, T. (1998). IT outsourcing as strategic partnering: The case of the UK inland revenue. European Journal Of Information Systems, 7, 29-45. Willcocks, L. and Lacity, M. (Eds.). (1998). Strategic Sourcing Of Information Systems. Chichester: John Wiley & Sons.

190 Kidd & Yau

Chapter XIII

Management Integration Through Software Applications: Japanese Manufacturing Firms in the UK Exert Control John Kidd and Tessa Yuk Lan Yau Aston Business School, Birmingham, UK

Through the 1970’s to the present time the governance of the Japanese firm overseas has changed, moving towards a more international form of operation rather than operating as a peripheral organization based solidly on the Japanese headquarters office. Notwithstanding the evolution of their formal and informal management structures this research questions in detail the nature of computer software that may have been imported from Japan to the UK to control imported production systems and/or to control management data flows. These research questions stem from evidence noted in the 1980’s that imported Japanese production machines contained Japanese language instructions beneath their cover plates: so when maintenance was to take place, the UK engineers became baffled and then frustrated at their inability to translate the “instructions.” In the research reported here, we are essentially asking if the computer programs used to control this imported machinery carry embedded Japanese documentation which may prove difficult to interpret by software engineers in the UK? As an adjunct to this question, we note the origin of their control software (and other software used by the Japanese production subsidiaries in the UK), its modification in the UK (by whom), and the natural language used to communicate with the HQs in Japan. As an emergent finding, we report on the tensions arising at that time from the data-integration of the Japanese firms’ operations in Europe through the use of Enterprise Resource Planning software.

Copyright © 2002, Idea Group Publishing.

Management Integration Through Software Applications 191

INTRODUCTION Japanese firms have been represented in Europe since the mid-19th century following the Meji revolution. However, the Japanese did not manufacture any goods in Europe until 1966 when YKK and Pentel opened their factories in the UK and France respectively. Since that time there has been a regular investment in their European productive capacity, slowing only as the global recession became an internal factor in Japan: as a consequence there are approximately 850 Japanese production subsidiaries in Europe at the present time. During early surveys of the Japanese production subsidiaries in the UK through the 1980’s, it was noted that their imported production machinery had embedded instructions for the first-line maintenance crews written in Japanese (see Kidd [1994], Kidd & Teramoto [1995b]). It is not surprising that this is so, as the machinery in the UK was derived from a KND (knockdown and reassemble) program which brought viable machinery systems from Japan to be the bedrock of their production in the UK. However, the embedded Japanese instructions for first-line maintenance hindered the development of harmonious relations between the UK work force and their managers who, in turn, were often controlled by an expatriate Japanese chief executive officer. Through the 1970’s to the present time we have seen the nature of the Japanese firm change as it moved towards a more international form of operation rather than seeming to run a peripheral organization based solidly on the Japanese headquarters office. Notwithstanding the evolution of their formal and informal management structures, we wished in the research to question in detail the nature of any computer software that may have been imported from Japan to control their imported production systems and their local managerial control systems. Did they have embedded Japanese documentation which might prove difficult to interpret by software engineers in the UK? Thus, our primary questions were upon the origin and subsequent modification of the control software, and where such modification took place. Subsequently, we explored the origin of all software in use and, during interview, found many firms were considering using Enterprise Resource Planning (ERP) software to integrate their systems. We report on the use of computer software in the Japanese production subsidiaries (JPS) only in the UK from research conducted through 1998-99. This sample restriction arose from pragmatic reasoning–there were about 250 Japanese production subsidiaries in the UK, and nearly 900 across Europe. We restricted the search to the UK, avoiding the expense of a pan-Europe postal survey, so as to obtain a response to our UK postal survey in a reasonable time frame, and so we could undertake interviews and/or telephone discussions in our natural language, English, with our sample firms. It was found that there was little software that was still substantially the same as that in use in Japan: most was modified in the UK though outsourcing contracts, and much of the other operational software was purchased from UK or US sources. Overall, managers said they were looking for their software to be able to be able to be integrated– and thus they were also looking to use Enterprise Resource Planning applications (like SAP, Baan, PeopleSoft, etc).

The Japanese Investment in Europe In recent years, as the world recession deepened, there has been a deceleration in the investment by the Japanese in productive capacity in Europe, although investment in R&D centres continues unabated. However, leading up to our research deadline, there were 825 Japanese manufacturing firms in Western Europe, of which 331 firms indicated they were

192 Kidd & Yau

involved in some form of R&D; and there were 78 independent R&D establishments (JETRO, 1998). Figure 1 indicates the current numbers of Japanese manufacturing firms across Europe based on data published by the Japan External Trade Organization (JETRO, 2000). We note that the investment in three countries (the UK, Germany, and France) tends to have absorbed about 55-60 % [in numeric terms] of the overall European Inwards-FDI from Japan–a figure that has held reasonably constant from 1983. We note also that many Japanese firms operate in more than one European country either as a major manufacturer (e.g., Sony), or as a keiretsu-like supplier to major manufacturing firms, following an organizational norm developed though their evolutionary history in Japan. Overall in 1998, the average capital per Japanese company across Europe was US$ 22.3m, with 182 firms having less than US$ 5m (i.e., 42% of the total firms); and were on average 1.2 factories per company that yield sales of US$ 92m. These data should be viewed against the fact that the European Union saw US$ 129 billion inwards foreign direct investment in 1997 (UNCTAD, 1998). The current data is not much different–the average capital per firm was US$ 21.26m (JETRO, 2000), but the FDI inflows to the European Union are now US$ 305 billion (UNCTAD, 2000). With respect to the UK we now find: • English-speaking countries account for 73% of all FDI in the UK • The USA and Canada account for 69% of all FDI in the UK • The EU accounts for 20% of all FDI in the UK • Japan accounts for under 1% of all FDI in the UK Given Japanese firms had invested earlier and more extensively in the US than in Europe many early academic studies upon these “foreign” firms were based on US apocrypha and observations. It is not pertinent to report these finding here, but the interested reader may Figure 1: Numbers of Japanese production subsidiaries in Europe (Source : 16th Annual Survey of Production in Europe, JETRO, 2000) Manufacturing firms firms with with at at least least 10% 10% Japanese Japanese ownership ownership Manufacturing 1

Long-term investment in European Union 10 1

14

3 29

235

Recently more investments in Central Europe

58 43

10 115 12 3

125

10

7

4 21 3

57 20

65

Total Firms = 800 (West), 50 (Central)

JETRO 2000

4

12

Management Integration Through Software Applications 193

be referred to the American Association of Japanese Business Studies (http://www.ajbs.org) for detail. Inevitably, the growth in European investment by the Japanese through the 1980’s led to more home-based European studies upon the Japanese personnel management methods, their managerial methods, and their quality improvement programs. These studies sometimes questioned if there were regional differences between the management practices in the US or in Japan, compared to those observed in Europe: see, for instance, Oliver & Wilkinson (1992). The studies generally yielded interesting conclusions, especially when noting the detailed involvement of senior Japanese managers in the minutia of their firms’ operations. However, and pertinent to this research, we found there was a lack of detailed research on the use of computer-based support for management and control in the Japanese manufacturing sites in Europe. Of course, the use of CAD/CAM systems had been noted under the general consideration of inward technology transfer, but not to the depth of reviewing the code of the programs themselves. This paper addresses this issue: we considered, as an initial “hypothesis,” that many computer systems would have migrated from Japan to help control the machinery that was imported into the Japanese production subsidiaries in the UK. Further, we wondered if these computer systems may have had difficult-to-interpret Japanese documentation embedded within their code.

METHODOLOGY Sample Frames A two-stage approach was to used for the study. First, a postal questionnaire was addressed in person to the Chief Executive Officers of the Japanese production subsidiaries in the UK. Then, following an initial analysis of the returned data, interviews were requested to ascertain finer detail and to elicit “softer” information that respondents were unwilling to commit to paper while presenting their answers in the questionnaire. The population of Japanese production subsidiaries in the UK was not too large - some 203 questionnaires were sent out to all the firms listed by JETRO in their annual surveys. Certain standard data were obtained from the Directory of Japanese-affiliated Companies in the European Union (1996-97)–also published by JETRO. The “Japanese” firms have different ownership modes–they may be fully Japanese owned or owned through a majority share-holding; or there may only be a Japanese minority capitalization (down to, but not below 10%). It was noted at the time that the UK Government Department of Trade and Industry (DTI) listed 273 Japanese production subsidiaries in the UK–perhaps the JETRO list did not count firms having multiple sites. It was decided to concentrate on manufacturing firms, rather than service firms (or others), since the Japanese manufacturing firms and the management systems therein are well known and well studied–again note Oliver & Wilkinson (1992)–but they omit to mention the software used by the Japanese firms. Further, the manufacturing sector in general has shown good evidence of integrated and coordinated approaches to manufacturing on a world wide base: see many papers on the use of Just-in-Time systems or other “Japanese” management systems as portrayed in tomes on Technology Management (for instance, Khalil et al, 1999). Even so, any mention of “software” is directed to algorithms, not the importation of potentially difficult-to-read ‘management control’ software from one country to another: as in our case, from Japan to the UK.

194 Kidd & Yau

Questionnaire Framing We were to collect from the production firms a mix of general qualitative and quantitative data as well as focused data relating to their computer-based hardware and software. In part, this was to help us judge the worth of requesting an interview with a given respondent. Even so, the questionnaire was kept simple and easy to complete. It was sent to the CEO or President allowing him to delegate the task of completion. Iit asked for responses on • The company’s general trading information • The company’s IT strategy • The company’s manufacturing-related computer-based programs • The company’s general hardware and software • The company’s communication network The questionnaires asked if the firm would be willing to engage in follow-up interviews. These were eventually arranged with several firms–but due to time and cost constraints (given these firms could be scattered across the UK), some interviews were conducted by telephone or by e-mail.

A Caveat There is always a concern raised when using the “survey method”– there are questions raised about bias and validity; and further difficulties arise with respect to the response rate. In our survey the response rate is fairly low at 26%: Kidd earlier had quoted a survey conducted by himself and Teramoto of the Japanese manufacturers in the UK which had a 23% rate (Kidd, 1994; Kidd & Teramoto, 1995b). Furthermore, he noted other surveys on this target group of firms (Oliver & Wilkinson, 1992) who stated their response rates were 28% (a 1984 survey) and 34% (a 1991 survey). Kidd suggested that this rate of response over this particular target group would be “usual” given these firms are over exposed to questionnaires and requests for interviews. Indeed, McCormick (1997) stated that 8% of his total responses were explicit refusals from firms stating they were inundated by this form of survey or interview request. In one sense, it is rather nice they took time to reply to him so positively; however, the general anxiety remains. We note the annual publication of the “best and biggest firms,” such as the Times 500 in the UK or the Fortune 500 in the US, simply adds names to the databases of a vast army of business students searching for data. Their continued demands harden the attitudes of responsive managers against responding. In other words, we have no solution to raise the low(ish) response rate for this form of survey analysis. The target CEO was directed to complete and return the document or to delegate the document to a competent subordinate. The questions were quite simple and elicited factual information relating to the broad sectors noted above. In so far as this was exploratory Table 1: Response rates of the postal survey

Replies from Manufacturing Firms Returned useable questionnaires Returned with no answer Returned by the Post Office–“gone away” Total issued

Number 53 5 3 203

% 26 2 1

Management Integration Through Software Applications 195

research, we did not take a strongly positivist stance–to name hypotheses, and subhypotheses in order to undertake detailed statistical analyses. Frankly, we were looking to answer a simple question–was Japanese language embedded in the documentation of software applications code imported from Japan which might prove difficult to “maintain” by UK software engineers. We saw this problem as being analogous to the maintenance engineers in the UK factories in the 1980’s when faced with instruction panels written in Japanese located within the imported machinery.

RESULTS Company Size and IT Strategy It would be reasonable to imagine that it is only large firms that would be able to formulate and implement a company-wide IT strategy based on the surmise that they would have several IT staff, general knowledge about IT, and so on. In the questionnaire we asked bluntly if the firm had an Information Technology strategy (or not) and accepted their answer without further questioning of their perception of “strategy.” Similarly we accepted their perception and thus response when asking if they had a “major IT system.” Our lack of further questioning is based upon our knowledge of the many initiatives from the European Commission to support the SMEs (the Small and Medium Enterprises). These SMEs have been targeted by the EU and by the UK Department of Trade & Industry, as they constitute the bulk of feeder firms that service major enterprises, probably delivering goods on a Justin-Time (JIT) basis. Currently the DTI notes that in the UK, of the entire business population of 3.7 million enterprises, only 24,000 were “medium” sized (50 to 249 employees), and less than 7,000 were “large” having 250 or more employees (Source: DTI Statistical News Release, 7 August 2000). In the small SMEs (being about 99.1% of the UK population of enterprises) we see little absorptive capacity, that is, they have an innate inability to investigate and absorb innovation (Cohen & Leventhal, 1990; COST, 1998). In our sample we received responses from small through to large firms (see Table 2). It is clear in this respondent group that even small SMEs say they have IT strategies; and herein we noted that the 14 firms without a formal IT strategy reported they operated at least one major IT system. We thus looked to the origin of the IT systems in use (see Table 3). After, we note the sourcing of these IT systems in Table 4. One of our prior “hypotheses” was now confounded: of 43 companies which either had an IT strategy and/or used at least one major IT system, there were only 8 firms which had systems that migrated from Japan. Moreover, only four of these systems were reported to be basically identical to the systems in their Japanese headquarters. Thus, while one might observe the massive inward transfer of production machinery from Japan to the UK, 81% of the firm’s IT software systems do not originate in Japan. It would seem that outsourcing of IT systems (in the 43 respondent companies which stated they had IT systems) is popular–as popular in these Japanese production subsidiaries in the UK as in general studies of outsourcing (see, for instance, Lacity & Hirshheim, 1995). We also noted within the respondent firms (using major IT systems) that more than half had their systems outsourced from British suppliers, and the rest originate in firms based in the US, Japan, Germany, or The Netherlands. It would appear that the Japanese production subsidiaries in the UK have moved strongly towards localization strategies for IT software supply–perhaps this is to cope better with localized requirements of legislation (for instance, trade laws or taxation).

196 Kidd & Yau

Table 2: Respondent company size and IT strategy presence

Size of firm: employees Less than 50 50 - 100 101 - 200 201 - 250 251 - 500 501 - 1000 1001 - 2000 Over 2001 Total Numbers

% with IT strategy 18 6 26 3 3 21 10 13 39 (100%)

% without IT strategy 50 7 14 14 7 7 14 (100%)

Table 3: The origin of major IT systems

Of 43 companies with a major IT system: Systems from Japan Systems not from Japan 8 (19%)

35 (81%)

Table 4: The sourcing of IT systems IT systems developed though outsourcing From services suppliers From consultancies 29 (62%) 2 (4%)

Number of companies without a major IT system 10

IT systems developed internally 16 (34%)

Note: The apparent total is 47 firms, not 43, since 4 firms quoted joint internal/external development of software and so became double counted.

Manufacturing Control Systems Localization is also seen in the development and use of software specifically to control the manufacturing systems. 30 firms had such systems, running programs such as MRP, MRPII, MAPICS, CAD, CAM and so on. Yet, only 3 firms had transferred these programs from Japan–each of these systems needed some modification.

Communications Systems Through the 1970’s and the 1980’s many non-Japanese managers in the JPS of the UK expressed their concern at the prevalence of “Japanese-style” communication within their firms–that is informal, face-to-face, verbal discussion, often in Japanese, between the Japanese expatriates. These managers also reported the frequent use of the Japanese language over the telephone or in faxes between their Japanese HQ and the firm (Kidd & Teramoto, 1995b). In this research program we wished to see if digital communications systems (e-mails, the Internet, etc.) were the prevailing inter- and intra-firm mode of communication. We also asked if natural language was a barrier to communication. First we looked to the availability of computer-based communications networks (see Table 5).

Management Integration Through Software Applications 197

Table 5: The availability of a computer-based network

Communications net with the Japan head office Yes No 36 17

Communications net internally in the firm Yes No 41 12

Communications net with the suppliers Yes No 12 41

Table 6: Natural language in use (in firms with a link to their Japanese HQ)

Dominant language English & Japanese English Japanese No answer Firms with a communications net to Japan

Number of firms 14 9 4 9 36

% of firms 39 25 11 25 100

The links back to their HQ were based essentially on Internet technology and so were not restricted to the Japanese language (in fact, at the time, these messages had to be in English). The links with suppliers were invariably for EDI (Electronic Data Interchange) in pursuit of order and supply chain management within a JIT system. Most respondents said there were no serious problems now with their EDI systems, but they had experienced teething troubles. The firms without an obvious communications network readily used facsimile and the telephone for direct [synchronous] conversation, and thus the Japanese managers might have had a greater tendency to use their natural language when communicating with a fellow Japanese. We found the firms used an eclectic mix of computer hardware and controlling software: mainframe (12 firms), Unix (15), client/server (27), PCs (38), and Macintosh (2). They had a mix of networks, of which Internet dominated (38 firms), and Intranet followed in popularity (23 firms). In one firm we noted that the communications network “point of presence,” as it were, terminated in the laptop PC of the CEO–hardly a public service. We noted, with respect to natural language communications, most firms used the bilanguage of English and Japanese (see Table 6). Generally there were few reservations expressed, even in interview, with respect to the need to communicate in Japanese. Often this need was met by the senior expatriates using Japanese word processing software, or using direct speech via telephone, or even by handwriting via a fax–but they readily communicated with their UK staff in English.

Software Type and Origin As well as software being needed for the control of their production systems, firms have many other functions that could benefit from the application of information technology and its software. We questioned our firms on these aspects (see Table 7). With respect to the criteria for their choice of software (Table 8).

198 Kidd & Yau

Table 7: The country of origin and functional use of software (numbers of respondents) Country of origin UK US Internal development Japan Germany Holland A bureau Other

Sales & accounting 42 6 1

Payroll 43 1 1

Customer database 33 4 1

Word processing 25 26 6

Manufacturing control 28 7 6

0 0 0 0 0

1 0 0 2 3

1 1 1 0 0

12 0 0 0 0

1 2 2 0 0

Table 8: Criteria for software selection (number of firms responding)

Efficient 34

User friendly 29

Cost 29

Integration 23

Language 7

The selection criteria seem hardly surprising. The low score on the language criterion further emphasizes the ease by which the staff in the Japanese production subsidiaries in the UK manage to communicate internally and externally without strict reliance on the Japanese language. In Table 7 we have noted the Japanese sourcing of word processing software, which suggests there is a specific, but not overwhelming, need for Japanese language text. We suggest that the 26 respondents who nominated the US origin of word processing software may simply reflect the acceptance of Microsoft as a vendor, and they are known to be US-based.

DISCUSSION Localization Almost all of our respondents have a free choice of their hardware and software, and thus their IT systems as a whole–such purchases are not now controlled by their HQ in Japan. Only 8 firms (15% of our sample) have transhipped their IT from Japan, and no system is absolutely like the one “back home.” According to Porter (1986), Bartlett & Ghoshal (1989), Dunning (1986), and Kidd (1994), a localization policy can yield better flexibility and better customer orientation in overseas markets. Herein we should consider the UK to be the overseas market with respect to inwards investment from Japan in terms of the technology transfer. Furthermore, this localization policy should also yield cost effectiveness through purchasing locally, by reducing system delivery time, and by shortening the technical support chain. For many Japanese expatriates this is a fearful solution, since they are breaking new ground in splitting from their known and trusted suppliers in Japan, their traditional keiretsu. They are, however, being courageous and looking to European solutions for their local IT needs.

Management Integration Through Software Applications 199

Cross-Language and Cross-Culture In the early days of “Japanization” in the UK and in Europe, there were many mistakes made due to misperceptions of language and the nonrecognition of cultural modes and inclinations: see Kidd and Teramoto (1995a), Bartlett and Ghoshal (1989), Hunt and Targett (1995). Many of these differences can be understood by reference to the research of Hofstede (1980, 1991), Trompenaars (1993), and others. It is understood that through more frequent exposure to people from different nationalities, and through the media and television, we in the West have come to understand better some of the Oriental attitudes. Yet Hampten-Turner and Trompenaars (1998) have suggested we should absorb more of the ways of the Oriental into our Occidental management routines. Herein we wish to say that, in our sample of managers (in the interviews), they claimed to suffer little, since each side of the cultural divide seems now to be well educated about the other. We would say now they have become “culturally literate” (Merry, 2001). Both sides seem more tolerant by 1998-99 than during the early years of the Japanese production subsidiaries in the UK. Many of the managers in interview suggested that a good way of easing cross-language and cross-culture differences was to initiate better coordination and better integration of database systems based on open and fully accessible data. Formerly it was noted that data, being the bedrock of knowledge, tended to be gathered by the Japanese and shipped back home. While it was not held exclusively in the domain of the expatriates, it was nevertheless not easily assessed by local managers–in part because it may have been encoded in Japanese in formal Japanese-style reporting documents like a ringi. Opening the database to all should reduce ambiguity. It is a cultural norm across Asia to utter ambiguous statements, and the Japanese are no exception to this: but Westerners (UK persons, northern Europeans, and US persons) don’t like ambiguity–there are large differences in the Hofstede indices of Uncertainty Avoidance, for instance. If one is an “insider,” then access to data will be open, but to outsiders the access is resisted. This is a natural effect in all cultures, but it is made more difficult if languages and organizational practices differ greatly, as between the UK and Japanese firms, hence the need expressed by local managers for open database access in the Japanese production subsidiaries in the UK.

Coordination and Integration The research showed the Japanese production subsidiaries in the UK were using their own IT and software systems that were said to be not well integrated one with the other, or integrated with their HQ systems. The basic coordination is done by the HQ after collecting global data and collating it for their own purposes–it was the Japanese tradition to employ a mass of middle managers to make sense of their global data. Nowadays it is easy to purchase or pay-on-demand for a high bandwidth communication system; further, in-house IT systems may fairly easily be integrated, thus enabling enhanced data flow. As Gupta & Govindarajan (1991) note, “more intense communications patterns create higher information processing capacity, and these patterns become especially desirable in contexts where such capacities are needed” (p. 779). High data flow requirements are noted by Hedlund (1986), Bartlett & Ghoshal (1989), and earlier by Galbraith (1973) in order to reduce uncertainty in complex contexts and to sustain the flexibility of response to changes in the environment. Such data flows are a fundamental requirement in the support of the “Law of Requisite Variety” (Ashby, 1958). At one time, say in the 1970’s and 1980’s, it was sufficient for the Japanese CEOs in Europe to fly to Japan once per month to embed themselves in their HQ nemawashi processes and thus integrate, in a sense, their data flows from Europe into the

200 Kidd & Yau

centre in Japan. But now, under the more intense pressures of globalization and the general acceptance of global supply chain integration, there is a need for faster and more formal data exchange. These aspects were noted in 1998 by 23 firms who said it was important to consider “integration” when assessing software acquisition. We noted from our questionnaire data that the sampled firms said their software (other than their WP software) was often sourced from the UK (see Table 7). This seems to be another misinterpretation. More realistically, in interview, the respondents correctly identified some software as being “sold” from a UK site, while it might have originated in Germany (as for SAP, for instance), or the US (e.g., Microsoft products). What was salient was that several of our sample firms said in interview they were progressing to full integration of their data systems–to use integrated applications developed by firms like SAP, Baan, PeopleSoft, or Oracle. The acceptance of Enterprise Resource Planning systems (ERPs) for these firms was more than an academic concept–it was a contemporary response to integrate their pan-European data flows. Maybe it was their reaction to the increasing one ness of Europe; yet, it was more likely that these managers see data and administration integration as a way of making their pan-European operations more efficient and effective: incorporating the Euro currency into their accounting systems is just one aspect. Their newly integrated systems offer empirical evidence for the “differentiated network” proposed by Nohria & Eccles (1997), which “allows MNCs simultaneously to achieve decentralized authority and initiatives, lateral coordination and sharing, central control, and overall organizational cohesiveness” (p. 129). This is a big claim, but we thought we were observing such a sea-change from a multinational strategy with decentralized organizational structures to a transnational strategy with globally integrated networks of operation (Karimi & Konsynski, 1991). However, there are at least two issues opposing the generality of this scenario. The first is that the SMEs who are in the [European] supply chains simply can’t afford the high costs of total organizational systems integration, even if they acknowledge its academic benefit. They haven’t the spare resource capacity or capability in terms of people skills or cash flows. Secondly, for the major players there is some resistance expressed by their associate firms outside Europe–in the US, and in Japan. This last factor has two aspects: the first is the “not invented here” syndrome which can be overcome by discussion and reeducation. The second is simple, and again a reeducation issue. There was limited knowledge, and especially experience, about the benefit of SAP-like applications in the US, and almost zero knowledge in Japan, as quoted by our sampled firms. This may be due to the slow and natural diffusion process of the knowledge of the ERP vendor’s success and their sales drive. The market leaders, SAP and Baan, commenced in Europe. For example, Baan was incorporated in The Netherlands in 1978, its first programs were ready in 1982, and slowly as it became successful, it moved its sales effort to the rest of Europe and by 1990 was selling to 35 countries through intermediate sales offices. It was only in 1994 that they opened a US office in California (http:// www.baan.com). This is not to say there was not knowledge that resided outside Europe. We were saying simply that knowledge diffusion processes are slow–and, through being in Europe, the European branches of the Japanese production subsidiaries became aware of the benefits of SAP-like applications systems (i.e., those of SAP [the market leader], Baan, PeopleSoft, or Oracle). Thus, some firms in our sample were implementing ERPs, which perturbed their US associates and the senior Japanese HQ managers in Japan.

Management Integration Through Software Applications 201

Reverse Technology Transfer Bartlet & Ghoshal (1989) state one of the three most important capabilities of a successful international firm is its ability to transfer knowledge and competence. This is vital for R&D continuance, but it is also vital in IT systems diffusion to maintain the continued vitality of an enterprise (Kay, 1993). Here, stemming from the UK, and from Europe, we see a force being expressed on the Japanese subsidiaries in the US, and upon the HQ in Japan, to join in the pan-European implementation of enterprise resource planning using SAP-like applications. As it is the continuance of innovation that ensures the vitality of a firm, we suggest both the Japanese HQ and the US affiliates needed to embrace the European initiatives in ERP to achieve greater global strength. This would inevitably cause cultural conflicts to be created in Japan where for a long time the senior managers in Japan have considered that innovation originating outside Japan is worthless, or almost so: the Japanese call this ringeisho (Peters et al., 1997).

Further Remarks Even without pressure to implement ERP systems, there is a need to integrate and harmonize accounting methods in multinational enterprises (MNEs) so that the calculation of profits, losses, and transfer prices become more transparent. This is not only for internal accountability in an MNE, to compare between subsidiaries, but for the government agencies which are demanding transparency to ensure compliance with agreements on tariffs and fair trading within the global World Trade Organization (WTO). In a similar vein, there is a need also to be able to exchange data within the MNE-based on GAAP (Generally Agreed Accounting Principles), to which many countries say they will accede, including China and Japan. Yet it is prudent to consider the meaning of “generally acceptable,” as it may apply in Asian countries. It has been shown that accounting disclosure, at least historically, is strongly correlated with cultural measures (cf. with those measures of Hofstede) and that Oriental cultures are biased towards secrecy (non-transparency) (Gray, 1988; Gray & Radebaugh, 1993; Gray & Vint, 1995; Salter & Niswander, 1995; Gray, 1996; Zarzeski, 1996). Thus one may now ask if different regularity practices and the development of more open financial markets in these regions will force firms to be more transparent (in a GAAP sense)? In fairness, we should note that opacity is not a unique East/ West issue, since the Channel Islands, Belgium, Spain, and Switzerland all practice low levels of financial disclosure (Gray, 1996). The ERP systems being implemented by some of our sampled firms will reduce the tendency to calculate the “profit” as though it was a percentage used to commence calculations rather than being the outcome of the accounting calculations. Following the implementation of GAAP and/or SAP-like applications, Japanese firms might be seen to incline more to the probity, transparency, and honesty demanded by Western bankers and investors. This is not to say that Occidental practice is completely without doubt, ambiguity, and needing interpretation–it isn’t. Yet there are observably many problems to be solved in the financial governance of Japanese MNEs that SAP-like applications may attenuate.

CONCLUSION We have seen from the data derived from our sample of the Japanese production subsidiaries in the UK that these firms were not dependant on inwards transfer of IT systems from Japan. Being managerially independent of their Japanese HQ, they have developed or

202 Kidd & Yau

acquired over the years of their operation their own systems, though perhaps in a rather piecemeal fashion. Now they were seen to be revising their systems, incorporating panEuropean data collection, and looking towards aggregation and comparability. This allows them to be, in a sense, stronger than their Japanese HQ, where traditionally it was only the HQ staff who would aggregate data. The transition to enterprise resource planning (ERP) applications offered by firms like SAP, Baan, PeopleSoft, or Oracle has indicated both to the US subsidiaries of our sample firms, and more especially to the Japanese HQ staff, that they may have to absorb reverse technology transfer (as it were). This would be to incorporate SAP-like programs into their US subsidiaries and into the Japanese HQ. Subsequently they would be fully integrated with their Japanese production subsidiaries in the UK, and thus all would become globally integrated. This will cause some angst in Japan and the US. However the Japanese production subsidiaries’ managers in the UK, perhaps through the advent of the common market of Europe, have had some years to acclimatize to the need for pan-European IT systems, and so they may have spear-headed a global integration of the Japanese firms. Happily, our data suggests conflicts due to cross-language and cross-culture issues which were apparent in the 1970’s and 1980’s were now largely dissipated in the Japanese production subsidiaries in the UK. These firms can progress towards the use of more open yet strongly integrated data systems.

POSTSCRIPT: IN MID-2001 Much has occurred round the world since this research was initiated, analyzed, and reported: the Asian financial meltdown, for instance. However, only three aspects will be considered in any detail–the continuing role of the Euro currency; the fall and rise of the fortunes of the ERP vendors; and the “hollowing out” of the workforce in Japan and now, generally speaking, globally.

The Euro (•) Currency Many of our respondents quoted their need to absorb the Euro currency into their IT systems–to modify their internal accounting systems obviously, but also to align their systems linked to their suppliers and to their customers. This was because they traded heavily with firms in continental Europe, and many of the Japanese manufacturers in the UK had subsidiaries scattered across this region. They were also well aware the European Union had agreed that the ‘electronic’ • would be traded by 1 January 1999, with paper-notes and coinage coming into circulation by 1 January 2002. The old •-zone currencies will cease to be legal tender by 1 March, 2002. Furthermore many firms, including our Japanese targets, were worried about the effects of the “Year 2000 (Y2K) bug” and brought forward their systems modifications to ameliorate the effects of both these problems: reasonably, they also considered implementing ERPs. There are frequent discussions about the •-changeover problems, and not least amongst these are the issues facing SMEs. Once more they have to face up to their lack of absorptive capacity and their lack of financial resources to make the change. The Association of German Banks says it will cost each of their branches US$ 63,000 in direct costs for the transportation of new coinage, insurance, customer information, and staff overtime. Crédit Lyonnaise says the French banks will have to absorb US$ 15 billion to make all the changes–on top of the US$ 3 billion they have spent on software and hardware

Management Integration Through Software Applications 203

(Business Week, 2 July 2001, pp. 42-47). The same article notes that SMEs are in a mess, and as we noted above, some 99% of the business world is composed of SMEs. It is not surprising that our target firms were in 1998 “seriously considering implementing ERPs” to help themselves and ostensibly to help their SME suppliers.

The ERP Marketplace Perhaps we might state without too much fear that most firms have legacy hardware and legacy software (COST, 1998). It is therefore not surprising to find that major firms were targeted by ERP vendors to initiate change and allow the ERP systems to fully integrate their legacy databases. At the same time, these firms were advised that they would “suffer” some form of business process re-engineering program to create synergies and efficiencies otherwise unobtainable. That optimism was tarnished not long ago when the technological stock markets tumbled, taking with them the ERP firms themselves. Even so, we see that “a tidal wave of enterprise resource planning implementations is sweeping US companies despite costs of $2 million and as much as fours years’ time to completion” (Krumwiede & Jordan, 2001, p. 49). Notwithstanding stock market “glitches,” many firms have continued to invest in ERPs whose vendors repositioned themselves to accept firms with lower annual turnovers in order to maintain their own business impetus. The ERPs capture organizational data, some say organizational knowledge also, by coherently mapping the firm’s business processes. Stijn & Wensley (2001) warn against the cognitive dissonance (Festinger, 1958) that may arise between the outputs from the new database structures and the memories from other media, such as human memory. Which is the more correct? The clarity of properly aggregated data from once disparate sources, or the fractured tacit models of a person who may believe in the ‘truth’ of legacy memories? As Wheatley (2001, abstract) says in referring to the need for better training in the use of ERP systems, “it is emerging that the [ERP] training that matters is not technical. What is called for, it seems, is an ability to figure out the underlying flow of information through the business itself.” This raises the issues of knowledge management (KM) and organizational learning (OL).

Hollowing Out of the Workforce Our original research coincided with the UK rising out of a mini-recession and continental Europe moving into a mild form of recession (note: the business cycles of each economy still remain out of synchronization). Thus, for the Japanese CEOs in the UK looking to make their businesses more efficient, the reduction in workforce hinted at by the implementation of ERPs was indeed enticing. And in Japan, even to this day, we find that more and more businesses have to greatly reduce their workforce to achieve economies that may just keep them viable. Notwithstanding this pressure to “downsize,” there are many Japanese firms that still employ “window workers” (madogiwa zoku), and the in-company unemployed remain on their payroll: or they create “leanness” by transferring their excess staff to subsidiaries (shukko). There is, however, a price to pay. One severs company with those staff who possible know most, being those who have worked in the firm for a long period. It is these people, in the main, who carry the important tacit knowledge of many aspects of the business practices (Kidd & Richter, 2001).

Finally… We wish to suggest from the dateline of mid-2001 that the major goal of an enterprise should be to utilize its data effectively. This is a broad issue encompassing ERPs as well as

204 Kidd & Yau

people issues. Inevitably, the latter become embroiled in knowledge management (KM) and organizational learning (OL) initiatives. Performed well, the ERP should be in accord with a KM initiative promoted top-down by a CEO; and the human resources management (HRM) function should be able to integrate the ERP into its promotion of OL in the workforce. Even better, the OL issues and the potential to extend the influence of the ERP to the SMEs of the supply chain could be more easily promoted by the HRM team if all have better access to pertinent data. To procure peer-to-peer OL is not too difficult within one’s own firm, but to cross organizational or cultural boundaries is more problematic. We suggest the integration of ones SMEs in the value and product chain should now be a dominant Boardroom topic, both from an economic and performance aspect and from the environmental aspect. In pursing this aim we have to pursue an additional program to enhance learning and trust in and between organizations (Kidd, Stumm & Richter, 2002).

REFERENCES Ashby, W. R. (1958). Requisite variety and its implications for the control of complex systems. Cybernetica, 1, 83-96. Bartlett, C. A. and Ghoshal, S. (1989). Managing Across Borders: The Transnational Solution. Boston, MA: The Harvard Business School Press. Cohen, W. M. and Leventhal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35, 128-152. COST 330 Action. (1998). Trends in Information Systems and Communications in European Ports. Luxembourg: Office for Official Publications of the European Communities. Dunning, J. H. (1986). Japanese Participation in British Industry. Kent: Croome Helm Ltd. Festinger, L. (1958). A Theory of Cognitive Dissonance. Evanson,Il, Row, Peterson. Galbraith, J. R. (1973). Designing Complex Organizations. Reading, MA: Addison-Wesley. Gray, S. and Radebaugh, L. H. (1993). International Accounting and Multinational Enterprises. New York: John Wiley & Sons. Gray, S. (1988). Towards a theory of cultural influence on the development of accounting systems internationally. Abacus, 24(11), 1-15. Gray, S. J. and Vint, H. M. (1995). The impact of culture on accounting disclosures: Some international evidence. Asia-Pacific Journal of Accounting, 2, 33-43. Gray, S. J. (1996). International comparisons of business performance: Measurement and disclosure issues. International Review of Business, 1(1), 1-15. Gupta, A. K. and Govindarajan, V. (1991). Knowledge flows and the structure of control within multinational corporations. Academy of Management Review, 16, 768-792. Hampden-Turner, C. and Trompenaars, F. (1997). Mastering the Infinite Game: How EastAsian Values Are Transforming Business Practices. Oxford: Capstone. Hedlund, G. (1986). The hypermodern MNC: a heterarchy? Human Resource Management, 25(1), 9-35. Hofstede, G. (1980). Culture’s Consequences: International Differences in Work-Related Values. London: Sage. Hofstede, G. (1991). Cultures and Organisations: Software of the Mind. London: McGraw-Hill. Hunt, B. and Targett, D. (1995). The Japanese Advantage? Competitive IT Strategies Past, Present, and Future. Oxford, Butterworth-Heinemann.

Management Integration Through Software Applications 205

JETRO. (1998). The 14th Survey of European Operations of Japanese Companies in the Manufacturing Sector, JETRO, Tokyo, December. JETRO. (2000). The 16th Survey on Japanese Manufacturing Affiliates in Europe, the Middle East and Africa, JETRO, Tokyo, November. Karimi, J. and Konsynski, B. R. (1991). Globalization and information management strategies. In Deans, C. and Jurison, J. (Eds.), Information Technology in a Global Business Environment, 169-189. Danvers, MA: Boyd & Fraser. Kay, J. (1993). Foundations of Corporate Success. Oxford: Oxford University Press. Khalil, T., El-Gammal, Lefebre, L., Hosni, Y. and El-Laithy, H. (Eds.). (1999). Civilization, modern technology and sustainable development. In Proceedings of the 8th International Conference on Management of Technology, IAMOT. Cairo, March 15-17. Kidd, J. B. and Richter, F. J. (2001). The hollowing out of the workforce: What potential for organizational learning? Human Systems Management, 20(1), 7-18. Kidd, J. B. and Teramoto, Y. (1995a). The learning organization: the case of the Japanese RHQs in Europe. Management International Review, 35(2), 39-56. Kidd, J. B. and Teramoto, Y. (1995b). Can the Japanese localize? A study of Japanese production subsidiaries in the UK. In Park, S. J. and Jovanovic, M. (Eds.), What is Behind the Japanese Miracle? 136-152. London, Megatrends I.E.C. Kidd, J. B. (1994). Gobalization through localization: Reflections on the Japanese production subsidiaries in the UK. In Schütte, H. (Ed.), The Global Competitiveness of the Asian Firm, 39-56. London: Macmillan. Kidd, J. B., Stumm, M. and Richter, F. J. (Forthcoming). Learning and trust in supply chain management. In Logistics Information Management Special Issue: Enterprise Systems and Business Integration. Krumwiede, K. R. and Jordan, W. G. (2000). Reaping the promise of enterprise resource systems. Strategic Finance, October, 49-52. Lacity, M. C. and Hirshheim, R. (1995) Beyond the Information Systems Outsourcing Bandwagon: The Insourcing Response. Chichester: John Wiley & Sons. McCormick, K. (1997). Managing R&D overseas: Japanese companies in the UK. Japan Forum, 9(1), 53-74. Merry, P. (2001). Cultural literacy–its link to business success in Asia-Pacific. In Kidd, J. B., Li, X. and Richter, F. J. (Eds), Maximizing Human Intelligence Deployment in Asian Business: The Sixth Generation Project. London & New York: Palgrave Press. Milne, I. (2001). Global Briefing Note 9: Foreign Direct Investment. Accessed July, 2001 from the World Wide Web at: http://www.globalbritain.org/Docs/MW/1144_GB-B9.htm. Nohria, N. and Ghoshal, S. (1997). The Differentiated Network. San Francisco, CA: Jossey-Bass. Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organizational Science, 5(1), 16-35. Oliver, N. and Wilkinson, B. (1992). The Japanization of British Industry: New Developments in the 1990s. Oxford: Blackwell. Peters, T. J., Peters, T. and LeBaron, D. (1997). The Circle of Innovation. San Francisco, CA: Knopf. Porter, M. E. (1986). Competition in global industries: A conceptual framework. In Porter, M. E. (Ed.), Competition in Global Industries. Boston, MA: Harvard Business School Press

206 Kidd & Yau

Salter, S. B. and Niswander, F. (1995). Cultural influences on the development of accounting systems internationally. Journal of International Business Studies, 26(2), 379-398. Stijn, E. V. and Wensley, A. (2001). Organizational Memory and the completeness of process modeling in ERP systems: Some concerns, methods and directions for future research. Business Process Management Journal, 7(3), 181-194. Trompenaars, F. (1993). Riding the Waves of Culture: Understanding Cultural Diversity in Business. London, The Economist Book. UNCTAD. (2000) World Investment Report. United Nations Conference on Trade and Development. Geneva. Wheatley, M. (2001). ERP training stinks. CIO, 13(16), 86-96. Zarzeski, M. T. (1996). Spontaneous harmonisation effects of culture and market forces on accounting disclosure practices. Accounting Horizons, 10, March, 18-37. American Accounting Association.

Can National Information Infrastructures Enhance Development? 207

Chapter XIV

Can National Information Infrastructures Enhance Social Development in the Least Developed Countries? Peter Nelson Meso and Nancy Bogucki Duncan Kent State University, USA

The need for national information infrastructures (NII) in the world’s least developed countries (LDCs) tends to be overshadowed by the nation’s severe deficiencies in physical infrastructure. Consequently, NII may be inadequately addressed by governments and supporting agencies in their plans for stimulating social growth. The example of Singapore’s TradeNet and other less-developed countries developing national, electronic information infrastructures suggests that information technology infrastructure may enable an LDC to develop at a particularly advanced rate. This paper studies the relationship of information infrastructure and social development. It establishes a clear correlation between 1) levels of information infrastructure and social development, and 2) growth rates of information infrastructure and social development. The findings suggest that governments of LDCs may enhance their countries’ growth by developing strategic plans for NII development.

INTRODUCTION The value of information technology infrastructure in business has been well documented over the past few years (e.g., Brancheau et al., 1996; Broadbent et al., 1996; Duncan 1995). It is understood to affect the firm’s process efficiencies (Keen, 1991; Weill, 1994) and to have strategic potential for the firm’s comparative performance in its industry (Keen, 1991; Brancheau et al., 1996; Duncan, 1995). It enables firms to economize on transactions through “virtual integration” (Clemons and Row, 1991; Miller et al., 1993) and to compete in markets that would otherwise be inaccessible. Appeared in Journal of Global Information Management, vol. 8, no. 4, 2000. Reprinted by permission.

208 Meso & Duncan

The concept and perceived value of a national information infrastructure (NII) arises from similar needs within and across national boundaries. As information technology expands a nation’s interconnectivity and capacity for information integration, hitherto independent sectors such as education, health, social policy, commerce and trade, government, agriculture, communications, and science and technology can be integrated. An NII that allows members of the various sectors to share information and related resources may increase process efficiencies and intellectual activity that lead to economic productivity. Sharing resources reduces the cost of affected projects or services, which in turn increases feasibility of new endeavors. Consequently, the span and scope of its information infrastructure can affect a nation’s delivery of social services and national productivity, and may ultimately stimulate economic growth. This synchronous development of a vision for integrated national services makes evident the importance and value of an integrated national information infrastructure. The value of greater information integration across independent sectors has been explored and exploited by nations with highly developed economies such as the U.S. In economically less developed countries, the need for NII is less obvious. It may be obscured by needs for more basic infrastructure (such as roads, electricity, and water treatment networks), or it may simply not be in demand because the need for information occurring in service-based economies is not yet pressing (Odedra et al., 1993). Yet if the NII can offer strategic economic benefits to a developed country, it may likewise offer means for more efficient economic development in the least developed countries (LDCs). Indeed, it may offer new and more efficient means to both social and economic growth. Since the concept of an NII is very recent, hardly any empirical research has been conducted to study the correlation or causal relationships existing between the development of an NII and the social and economic development of particular countries, least of all the LDCs. LDCs face immense disadvantages in social and economic development. Their precarious position is exacerbated by the current global economic order centered on international open-market trade. However, it is generally accepted that the increasing polarity between the developed nations and the LDCs is not beneficial for the sustainable growth of the international economy. Therefore sustainable growth of LDC economies is rapidly becoming an international concern. This paper examines the potential relationship between national information infrastructure and social development in LDCs. The next section of the paper examines the hypothesized role of information infrastructure in the social development of developing countries as well as in the international economy. It posits the unique capabilities of information infrastructure in providing access to markets and basic services unattainable without an adequately developed physical infrastructure. Such capabilities may offer an efficient alternative to physical infrastructure through which infrastructure-dependent services can be dispensed within LDCs. Part three presents the research design and method. Data collection and analysis are described, and the assumptions of the study elaborated. Part four documents the findings of the study and the inferences drawn therefrom. Our conclusions focus on the implications of this study for future research.

INFORMATION INFRASTRUCTURE AND SOCIAL DEVELOPMENT Previous research establishes that the quality and levels of services provided by sectors including education, health, commerce, government, agriculture, communications, and

Can National Information Infrastructures Enhance Development? 209

science and technology impact positively the rate of economic and social development in LDCs (Sadowsky, 1996; Antle, 1983; Romero, 1995; Saunders, 1996; Leff, 1984; Stone, 1993; Wingrove, 1993). To the extent that service quality is dependent on quality and quantity of information, and to the extent that delivery of service depends on efficient access to and delivery of information, we may expect economic and social development to be affected by the breadth and capabilities of the NII. In this section, we discuss the research problem, describe the research model, and present the hypotheses of the study.

The Problem: Can NII Development Catalyze Social Development in LDCs? Economic development data shows that developed countries exhibit higher information infrastructure indices—as measured by the proportions of the population having direct access to telephones, television sets, radio sets, cable television, newspapers and periodicals, postal services, and the Internet—than do developing countries. They also exhibit more extensive communication networks—as measured by total length of telephone, television, Internet, and Electronic Data Interchange cable per geographical area—than do developing countries. Further, these countries have a larger total number of broadcasting stations, television channels, newspaper and periodical titles, libraries, and post offices per geographical area than do the developing countries. A comparison of the developing countries to LDCs reveals the same type of relationships—developing countries have higher information infrastructure indices than the less developed countries (World Factbook, 1994; World Bank, 1994; Danowitz, 1995). This data suggests that as a given country achieves a higher level of economic development, it requires a more advanced information infrastructure to support the increased volume and complexity of the transactions taking place within its economy. Therefore the country’s national information infrastructure develops in the same direction as the overall economic development of that nation (Danowitz, 1995). These data suggest increased economic development pushes NII development. As a given country achieves a higher level of economic development, it will require a more advanced information infrastructure to support the increased volume and complexity of the transactions taking place within its economy. Therefore the country’s NII develops in the same direction as the overall economic development of that nation (Danowitz, 1995). In order to bring about rapid and widespread economic development, LDCs might develop concurrently all infrastructure components to such levels as will capacitate citizens to exploit fully the production resources within these countries. However, these countries lack the capital to develop these infrastructures (Odedra, 1993). The costs involved in developing the infrastructure have proved to be prohibitively high for most LDC governments (Antle, 1983). A feasible and quicker solution may be to develop an NII that will deliver intangible services such as education, health care, commercial services, and technological know-how to a wider population of citizens, regardless of their physical location or economic status in the LDC. The knowledge acquired by these citizens may then capacitate them to derive their own solutions to the tangible infrastructure limitations, such as poor transport networks and inadequate machinery. With adequate information access, citizens may obtain information on where to source cheap and reliable machinery and how to obtain financing for such machinery. They may also obtain information on how to use locally available materials to develop or build cheap and reliable machinery, hence expanding their productive capacity. They may identify methods

210 Meso & Duncan

of collaboration with government and private organizations to develop transport networks, access to markets, access to energy, and access to material inputs. In so doing the citizens will be able to improve their productivity and increase their output far more quickly than they would should they wait on the government to develop all the needed physical infrastructures—schools, hospitals, commercial services organizations, roads, railroads, airports, and machinery—for them. How the resources for NII development should be acquired and managed is a further problem. Development literature identifies government as the key player in stimulating development and in using it for purposes of social welfare. This literature suggests government actions 1) create an environment conducive to investment in information infrastructure; 2) create the initiative for research and development of the technologies that constitute the information infrastructure; and 3) promote this infrastructure’s applications in the provision of both economic and social services (Sadowsky, 1993; King & Konsynski, 1995). The private sector is commonly the key source of the investments in information infrastructure and plays a pivotal role in enhancing the efficient, effective, and innovative provision of information services. Private sector participation enhances the efficient distribution of the information infrastructure resources to demand areas within a country. The private sector is also credited with providing the fastest approach to expanding the information infrastructure in a country owing to its profit-centric objectives (Jackson, 1995; Swaroop 1994; Stone, 1993; Stewart-Smith, 1995; Ingram, 1993). The problem with private sector development of NII, particularly in LDCs, is that national economic growth requires that development occur throughout the country. Where demand or private enterprise do not result in information service provision, NII is particularly needed to enable inexpensive delivery. Governmental influence is needed to ensure that information infrastructure services are extended to the areas and sectors inadequately serviced by the market-type economy (Jackson, 1995; Noam, 1995; Odedra, 1993; Danowitz, 1995). If NII can indeed affect the social development (and subsequently the economic growth) of an LDC, early government intervention and investment may be justified. This paper limits its scope to examining the potential contributions of an NII to the social development of LDCs.

The Model The role of NII in the development of LDCs is understood as a complex function, including social development, economic development, investment, and governance. Figure 1 depicts a model of the relationships among these variables and NII. The dashed line isolates the variables of the present study: the relationship between NII and social development. We are currently conducting further research on the role of governance and the other variables. According to economic theory, an abundance of quality information yields more competitive economic markets (Berliant et al.,1998; Mokerjee et al., 1997; Pietra et al., 1998). Such markets optimize both the efficiencies and productivity of economic activities evidenced in commerce, trade, agriculture, manufacturing, and other value-creating sectors of the country’s economy. The greater the productivity and efficiencies of a nation’s economy, the larger the rate of economic growth. Abundant and widely accessible information reveals wealth creation opportunities, the risks inherent in such opportunities, the valuation of existing and emerging investment opportunities, the sources of existing and emerging investment opportunities, the sources of funding for entrepreneurial undertakings, and the markets available

Can National Information Infrastructures Enhance Development? 211

Figure 1: Model of NII in social, economic, and governance environment Governance Investment

Economic Growth

NII

Social Development

for investment transactions. In so doing, it impacts the quality and magnitude of investments in the nation’s economy. Easy and plentiful access to quality information enhances the quality and levels of educational, health, and habitat services in a nation. It enhances the awareness and knowledge of the nation’s citizens, resulting in higher nutritional, hygiene, habitat, economic engagement, and primary health care standards among the citizens. The net effect is an improvement in the social development of the nation owing to a better-educated, healthier, more economically productive citizenry and better standards and conditions of living throughout the nation. Access to information, and to the channels for information distribution, are therefore linked to the rate of national development. However, access to quality information depends largely on the nature of the nation’s NII. Therefore the reach and scope of the NII may be expected to affect the rate of development of the nation. But the level of economic growth and investment may reciprocally influence the rate of NII development. Low levels of gross investment result in less investment in the NII, which results in a slower rate of NII development. Slow economic growth results in less demand for quality information by the key agents and sectors of the economy, which further slows the rate of NII development. Hence we posit a bidirectional relationship between the NII and the economic development, and between NII and gross investment of a nation.

Hypotheses We may well expect that different countries will place development emphasis in different sectors and related infrastructure components. In this regard, some LDCs are expected to have significantly higher information infrastructure capabilities compared to fellow member countries. Likewise, the level of social development attained by each country is expected to differ, with some countries recording much higher levels of social develop-

212 Meso & Duncan

ment than others. We posit that those countries that exhibit high levels of information infrastructure development also exhibit higher levels of social development. H1: A significant positive correlation is expected between the level of social development and the level of NII development. It is possible that without causal interaction, infrastructure and social development may both increase as a country develops. Identifying three possible reasons for economic development as measured by the net increase in aggregate output, McCallum (1989) emphasizes that true and sustainable economic development seems to stem from technological progress, rather than in the increase in population or capital stock. Hayami and Ogasawara (1999) augment this perception. They write: Widely recognized as a stylized fact in the growth of advanced market economies is that the growth of national product depends predominantly on improved efficiency (or technological progress broadly defined) rather than capital accumulation…. Such results underlie the basic characterization of “modern economic growth” (MEG) by Simon Kuznets (1966, p. 9) as predominantly dependent upon sustained improvements in technology rather than capital accumulation due to the “extended application of science to the problems of economic production.” (1999, p. 2) In another study, David Mowery and Joanne Oxley (1995) acknowledge that the influence of international technology transfer on national economic development has rapidly increased, especially in the post-World War II era. They identify the importation of technology as the chief cause of the meteoric economic development evidenced in most of the newly industrialized countries. This is in direct contrast to the success stories of the nineteenth century—such as the United States—whose development was largely from the domestic generation of new technologies based on the exploitation of local natural resources and increasing physical capital accumulation. Mowery and Oxley (1995) further claim that those nations that have benefited the most from international technology transfer are those that have advanced “national absorptive capacities.” They define national absorptive capacity as the institutions, policies, and infrastructure that enables a country to readily adapt new technologies and internalize them into their local economic productivity activities to the extent that they are able to advance these new technologies without external assistance. National absorptive capacity in turn relies on investments in social infrastructure such as scientific and technical training, research and development institutions, and a healthy and a well-educated pool of human resources. In addition, economic policies that enforce competition among domestic firms enhance the national absorptive capacity of a given economy. Therefore, it may be that the higher the level of social development in a country, especially the level of education and technical human resources skills, the greater its national absorptive capacity. The higher the level of a nation’s national absorptive capacity, the greater is its rate of technological progress. A country with a higher rate of technology progress certainly experiences faster economic and social development. This suggests that the rate of technology development, of which the national information infrastructure is part, may be strongly correlated to that of social development within a given country or group of countries. To test whether a causal relationship exists between information infrastructure and social development, we further posit: H2: The rate of NII development is positively correlated with the rate of social development. Regardless of the trade method used to implement international trade—whether direct foreign investment, licensing, technical agreements and cooperation, joint ventures, turnkey

Can National Information Infrastructures Enhance Development? 213

projects, or the purchase of capital goods—three core flows distinguish how technology transfer content flows from the exporting to the importing nation: engineering services, managerial services, and capital goods; skills and know-how for operation and maintenance; and knowledge, expertise, and experience for generating and managing technical change. Each succeeding level of content flow entails a greater intensity of knowledge than its preceding flow (Wei, 1995). All are information intensive. In fact, the nature of these flows suggests that information infrastructure may enhance the rate and intensity of technology transfer between nations by providing the conduits though which such knowledge can more readily flow from the exporting to the importing nations. Hence the suggestion made earlier, that the rate of technology development may be strongly correlated to that of social development within a given country or group of countries, can be made even more specific: There may be a strong correlation between the rate of national information infrastructure development and social development within a given country or group of countries. This gives further credence to the second hypothesis (H2). Madden and Savage (1998) suggest that “investment in telecommunications infrastructure has the potential to improve national productivity and economic growth” (p. 174). Studying a group of 11 transition economies of Central and Eastern Europe, they found that the level of development in telecommunications infrastructure, as measured by the investment in the infrastructure, was indeed related to economic growth. They built on the work by Aschauer (1989), who also argued that public infrastructure investments have a strong association with productivity and growth. Fisher (1995) goes further to suggest that the rate of investment can determine the rate of development in a particular country. So can policies geared at promoting capital formation. Applying their suggestion to infrastructure, the rate of investment in economic infrastructure—of which the national information infrastructure is part—and the adaptation of policies that enhance the development of these infrastructures, are expected to influence the rate of a nation’s social and economic development.

RESEARCH DESIGN To study the impact of the NII on social development in the LDCs requires an operationalization of NII that is relevant to LDCs. The Organization for Economic Cooperation and Development (OECD) envisions NII as a cohesive network, able to send and transmit text, graphics, audio, and video messages at any time, from any source to any destination on the network. The concept of the NII emerged in the early 1990’s—an offshoot of the commercialization of the Internet and the World Wide Web, and the globalization of the world economy (OECD, 1996). The OECD recognizes that pure national information infrastructures are nonexistent. Instead, a country will have a limited collection of integrated or integrating networks that provide for the exchange and dissemination of information. Thus information infrastructures can be seen to include: broadband communication technologies which, through the process of digitalization of communication infrastructures, the convergence of these technologies with broadcasting technologies, and recent technological developments for switching and transmission allow rapid transmission of large quantities of information at low cost (OECD, 1996, p. 12). These infrastructures, regardless of their configurations or topologies, are capable of handling multimedia services and represent the information in digital form (OECD, 1996). It is important to recognize that NII does not necessarily mean the Internet. Until this decade, NII throughout the world meant radio, newspapers, postal services, etc. These are

214 Meso & Duncan

still vital components of information infrastructure throughout the world, and they are still the basis of information infrastructure in LDCs. While our ultimate goal is to establish the value Internet resources offer LDCs as an inexpensive means for rapidly expanding NII, the relationship of NII to social and economic development can only be empirically tested using the “old technologies” that comprise current LDC information infrastructure. Interest in national information infrastructure as an avenue for social and economic development is relatively new (Firestone, 1995). The empirical testing and validation of the emergent theories in this topical area is just beginning to take shape (Palvia, 1997; OECD, 1996). This study intends to provide initial or foundational insights on the association between social development and national information infrastructure—hence its use of correlation analysis. This approach has been used in the past in similar studies. Notable among these is the work by Aschauer (1989), who studied the relationship between public investment and productivity growth. McGuire (1987), Merz (1999), Verbrugge (1999), and Malliaropulos (1999) also used correlation in their studies on economic relationships. The study employs longitudinal data to test the hypotheses. Data on social development is collected for each LDC in the study for the years 1984 and 1994. Likewise, data on information infrastructure is collected for each of these countries over the same period of time. The information infrastructure data is then analyzed to compute the Information Infrastructure Index (II-INDEX) for each country. These measures are then used to classify the LDCs into two categories: High-II-Countries and Low-II-Countries. A social development factor is derived for each country by analyzing the data measuring the social development parameters—education, health, and economic performance. The study includes only those LDCs that have been listed as such over the 10-year period 1984-1994. This period constitutes an “NII-Emergent” period—i.e., during this time the concept of the NII received global appeal as a potential pivot for economic and social development. Increasing government awareness of the potential value of NII to LDCs appears to have led to stepped-up development of state information resources. At present there are 48 countries classified as LDCs. Of these, 33 are African countries, three are Asian countries, and one is a Latin American country. The rest are small island states of the Pacific and Indian Oceans. Only 45 of these countries have adequate quantities of consistent data over the period of the study. These 45 countries make up the sample for this study (UNDP, 1998). A list of these countries can be found in Appendix A. NIIs of LDCs cannot be expected to include widespread sophisticated computer networks. Consequently, the reach of information media is a reasonable initial measure of information infrastructure. Radio, TV, and phone are essential and basic forms of information transfer. We will expect radio density always to be highest because radios can function without corresponding physical infrastructure (such as electrical wiring), while TV and phone—at least during the period of study—would not. The Information Infrastructure Index (II-Index) is calculated from the following: 1. Radio density (Number of radios per 1,000 population) 2. Television density (Number of televisions per 1,000 population) 3. Telephone density (Number of telephones per 1,000 population) 4. Newspaper circulation (Number of newspapers per 1,000 population) The first two measures assess the broadcasting infrastructure, the third measures the telecommunications infrastructure, while the fourth assesses the newsprint infrastructure. We calculate the “Information Infrastructure Iindex (II-Index)” by obtaining the mean of the radio density, television density, telephone density, and newspaper circulation as a percentage.

Can National Information Infrastructures Enhance Development? 215

The Social Development Index (SDI) is determined from three types of social development indices: i) Education: measured by the illiteracy rate (as a percentage of population above 14 years of age) and primary school enrollment rate (as a percentage of school-age population). ii) Health: measured by life expectancy at birth (years), infant mortality rate (as a ratio per 1,000 live births), access to safe water (as a ratio to 1,000 population), and access to health care (as a ratio to 1,000 population). iii) Economic: measured by the Per Capita Gross National Product (GNP). In order to determine the SDI, we convert all the above indices into percentage scores and ensure that all use the same interpretation scale (the larger the value the better). To do this, we convert: i) Life expectancy into a percentage score by setting an ideal score of 100 years to be equal to 100%, ii) GNP per capita into a percentage score by setting an ideal score of US $1,000 to be equal to 100%, and iii) Infant mortality rate into a percentage score by first setting 1,000 deaths per 1,000 population to be equal to 100%, then obtaining the “infant survival” rate by subtracting the score so derived from 100. Therefore the Social Development Index is the mean of the literacy rate, primary school enrollment rate, life expectancy, access to health care, access to safe water, “infant survival,” and gross national product as a percentage. Three other indices are used in testing Hypothesis 2: i) Increase in telephone lines (net increase in the number of telephone lines over the decade); ii) Rate of growth in telephone density (compounded annual increase in telephone density); and iii) Rate of growth in telephone lines (percentage change in telephone lines over the decade). We use these indices to corroborate the results obtained from the analysis of the first hypothesis by correlating the rate of development of social development to each of these indices. All three measure the rate of development of the telecommunications infrastructure. Data on the other infrastructure was insufficient to determine their rate of development over the decade 1984 to 1994. However, the telecommunications infrastructure is the most extensive type of information infrastructure in the countries we studied. Therefore we feel that the correlation measures obtained between the rate of development of the telecommunications infrastructures and social development rate are indicative of the general relationship between information infrastructures’ development and social development. The main sources of data used in this study are development databases and social development publications. Social development data is derived from the World Development Indicators, The World Development Report, and the Social Development Indicators for the years 1984-1997. Data on information infrastructure is obtained from the World Telecommunications Handbook, the World Fact Book, Lexis-Nexus Database, and also from the social development data sources above.

Limitations and Assumptions of the Study Data collection on information technology in LDCs is constrained by numerous factors. In our own effort, we found that:

216 Meso & Duncan

i)

No source provides complete data on the range and reach of computers and computer networks in the least developed countries. This is not surprising when we consider that the infrastructure limitations themselves would impede accurate collection of the data. Thus, we excluded computer data from this study and relied on more basic information media. ii) For every country, at least one piece of data in the years studied was missing. No country has a complete data set. iii) 1984 data on NII resources in LDCs is quite sketchy. Apparently in 1984, NII was not yet viewed as a major concern to development, and consequently, data was not collected systematically. In order to optimize the accuracy of our study and our interpretation of the data, we make the following assumptions: i) All data obtained from development databases and publications on social development are accurate, correct, and truly representative of the state of affairs in LDCs. ii) These secondary sources of data provide rational and unbiased data about the LDCs. iii) The results of this finding are representative of LDCs but may not be generalized to the rest of the global family of nations. iv) Because, on the whole, information infrastructure in LDCs is still grossly underdeveloped compared to that of developed nations, none can be said to own what can truly be defined as a national information infrastructure. However, since an NII evolves from a collection of integrated information infrastructures, we assume that the relationships observed between the common information infrastructures in LDCs and their social development are truly representative of the relationship between an NII (including computers) and social development. v) The mean of a given variable is a better indicator of the measure of that variable for a country where no data on that variable exists, than is the null value. Thus, we use the means-plugging method of managing missing data. vi) The overall level of development of the information infrastructure in a particular country is a stronger determinant of the rate of social development than is the rate at which that infrastructure was (or is currently being) developed.

RESULTS Findings with Respect to Hypothesis 1 Correlation analysis of the data collected in the study indicate that Hypothesis 1 is fully supported. The analysis revealed a significantly strong positive correlation between the level of information infrastructure development (as measured by the II-INDEX) and the level of social development (as measured by the SDI) of 0.35. Table 1 outlines the correlation measures of the information infrastructure indices and all primary social development indices for the year 1994. As can be observed from these measures: i) The correlation between the SDI and each of the information infrastructure indices except radio density is significantly positive. This is a clear indication that information infrastructure directly relates the level of social development in given country. Therefore developing the information infrastructures, especially those which strongly influence social development in LDCs, may provide an avenue through which the development of these countries can be accelerated.

Can National Information Infrastructures Enhance Development? 217

Table 1: Correlation of information infrastrucuture indices and social development indices for 1994 CORRELATION OF: Social Development Index Literacy Rate Rate of Primary School Enrollment Infant Mortality Rate (I-Infant Mortality Rate) Access to Health Care Access to Safe Water Life Expectancy at Birth GNP per Capita

ii)

iii)

iv)

v)

INFORMATION INFRAST RUCT URE INDEX (II-INDEX) 0.35

TELEPHONE DENSIT Y (1994) 0.63

TELEVISION DENSIT Y (1994) 0.15

RADIO DENSIT Y (1994) 0.02

NEWSPAPER CIRCULATION (1994) 0.43

(0.08)

(0.01)

(0.03)

(0.32)

(0.42)

0.33 (0.18) 0.18

0.29 (0.54) 0.54

(0.04) 0.01 (0.01)

0.20 0.02 (0.02)

0.52 (0.23) 0.23

0.20 0.08 0.28

0.42 0.45 0.53

0.10 0.09 (0.04)

0.18 (0.28) 0.03

0.28 0.06 0.28

0.37

0.61

0.36

0.13

0.45

Save for television density, there is a strong positive correlation between the rate of primary school enrollment and each information infrastructure index. Once more, this provides the evidence that information infrastructure may directly impact the levels of educational attainment–one of the pillars of social development–hence providing a feasible avenue by which the social development of LDCs can be accelerated. Information infrastructure may be equally contributing to the enhancement of health services in these countries. The correlation measures of Access to Health Care and all information infrastructure indices are significantly positive. So are the measures of the “infant survival” rate (1–infant mortality rate) and the life expectancy at birth to all information infrastructure indices, except television and radio density. Health is the second pillar to social development. Therefore the direct relationship observed between the level of information infrastructure development and the level of health care may be indicative of the primary contribution of information infrastructure to the acceleration of health care and medical services development in LDCs. This reinforces the importance of information infrastructure to the social development of each country. There is a significant positive correlation between each information infrastructure index and the per capita GNP of the LDCs. This confirms that information infrastructures directly relate to the enhancement of the economic sector. Developing adequate information infrastructures for the LDC countries may thus provide a concrete avenue by which these countries can overcome the economic constraints they currently face and hence alleviate the poverty problem they face. A close look at the telephone density and how it correlates to each social development index provides revelations of immense potential. Except for the literacy rate, the correlations between telephone density and each social development index are strongly positive. Given that the telecommunications infrastructure is the foundation to the present-day electronic information superhighway (as exemplified in the Internet, the World Wide Web, cable media networks, and high speed backbone networks), these correlations reveal the immense potential that the information superhighway has in impacting the rate and the level of social development in the LDCs. It is this one observation that convincingly suggests the true worth of information infrastructure to the enhancement of developmental activity in these countries. Therefore there is good

218 Meso & Duncan

reason to support the direct development of information infrastructure in the LDCs as an effective, efficient, and reliable strategy bringing about meaningful and sustainable development in these countries. In summary, the strong correlations between the SDF and the II-INDEX, and between the individual information infrastructures and the individual social development infrastructures, suggest that information infrastructure may be directly contributing to the state of the economic, education, and health sectors in a country. The strong correlation between this infrastructure and education is a clear indication of the potential this resource has to the enhancement of education in the least developed countries. The combined strong correlations between these infrastructures and all aspects of social development make them strategically valuable as assets for sustainable development. The fact that they seem to positively impact each sector without having to be fine-tuned to the needs of each sector make them a viable and feasible avenue for fostering sustainable development in the LDCs.

Findings with Respect to Hypothesis 2 To test Hypothesis 2, we correlated the infrastructure growth rate variables to the social development indices—the volume of telephone lines, the rate of growth in telephone density, and the percentage increase in telephones over the study period. Correlation results for this test are presented in Table 2. The correlation of the net change in SDI over the period 1984 to 1994 and each of the II measures was significantly positive. This provides evidence of a direct relationship between the rate of information infrastructure development and the rate of social development. Therefore, a development strategy pegged on enhancing the rate of development of information infrastructure may accelerate the rate of social development. This finding is indicative of the central position of information infrastructure in the development of a country and its potential contributions to a sustainable level of development. Given that the key development problem in almost all LDCs is their inability to achieve sustainable development, a strategy centered around enhancing the information infrastructures within an LDC may provide the means by which that LDC can attain a sustainable level of development. This dynamic positive relationship may provide a means by which current obstacles to development can be skipped. Bennet (in Odedra, 1993) asserts that “IT is not, and will never be, a way to leapfrog development, an immensely complex process having its roots in educational and infrastructural building which cannot be bypassed” (p. 28). The strong positive correlation values observed between education and all information infrastructure indices may be in contradiction with this assertion. By providing the means through which a larger number of citizens can be educated concurrently, information infrastructure may enhance capacity building. Since social and economic development result from a better educated and more productive population, the beneficial ripple effects of adapting an information-infrastructure-based education policy may be immense.

Further Analysis of the Results To obtain a better understanding of the results obtained in Hypothesis 2, we used cluster analysis to classify the LDCs in the study. Figure 2 identifies the ensuing clustering of all LDCs studied. To corroborate the results in Figure 2, we clustered the countries on the basis of the volume of telephones, the percentage increase in telephone lines over the study period, and the compounded annual growth rate of telephone density, vis-a-vis the social development factor. Figures 3, 4, and 5 pictorially outline these results.

Can National Information Infrastructures Enhance Development? 219

Table 2: Correlation Of Control Information Infrastructure Indices and Net Change In Social Development Indices (1984 To 1994) CORRELATION OF:

Increase in Telephones over the Decade (1984 to 1994)

Percentage Increase in Telephones over the Decade (1984 to 1994)

Co mpounded Annual Growth Rate of Telephones over the Decade (1984 to 1994)

0.17 0.20 0.20 (0.13) 0.13 (0.15) 0.08 0.20 0.04

0.40 0.21 0.14 (0.08) 0.08 (0.10) 0.25 (0.04) 0.41

0.33 0.25 0.17 (0.11) 0.11 (0.08) 0.23 0.00 0.25

Net Changes in: Social Development Inde x Literacy Data Rate of Primary School Enro llment Infant Morality Rate (1-Infant Mortality Rate) Access to Health Care Access to Safe Water Life Expectancy at Birth GNP per Cap ita

(Net change is the measure of the magnitude of change in each social development index over the period 1984 to 1994) Figure 2: Clustering Of LDC’s By Information Infrastructure Index And Social Development Index (1994) high SOCIAL DEVELOPMENT INDEX (SDI) 1994

LESOTHO BENIN ANGOLA COMOROS ZAMBIA

GUINEA BISSAU BANGLADESH BURKINA FASO YEMEN RWANDA

LAO TANZANIA BURUNDI HAITI NIGER CAMBODIA NEPAL UGANDA ZAIRE

DJIBOUTI BHUTAN GUINEA CENTRAL AFRICAN MALI AFGANISTAN SOMALIA MOZAMBIQUE

low

CAPE VERDE VANUATU KIRIBATI SOLOMON ISLANDS SAO TOME & PRINCIPE EQUITORIAL GUINEA MYANMAR MAURITANIA LIBERIA MADAGASCAR CHAD SIERRA LEON ETHIOPIA

INFORMATION INFRASTRUCTURE INDEX (II-INDEX) 1994

GAMBIA MALDIVES SUDAN MALAWI TOGO

high

Figure 3: Clustering Of LDC’s By Increase In Number Of Telephone Lines and Increase in Social Development Index (1984-1994) high

INCREASE IN SOCIAL DEVELOPMENT INDEX (1984-1994)

low

MALDIVES

BURUNDI

BANGLADESH

SOLOMON ISLANDS CAPE VERDE TOGO LESOTHO VANUATU MALI MAURITANIA BHUTAN UGANDA KIRIBATI COMOROS AFGANISTAN BENIN SUDAN SAO TOME ANGOLA ZAIRE DJIBOUTI

GUINEA MALAWI GAMBIA GUINEA BISSAU SOMALIA CHAD NIGER

MYANMAR BURKINA FASO NEPAL

CENTRAL AFRICAN LIBERIA CAMBODIA SIERRA LEON LAO HAITI RWANDA MADAGASCAR EQUITORIAL GUINEA

YEMEN ZAMBIA ETHIOPIA TANZANIA MOZAMBIQUE

INCREASE IN NUMBER OF TELEPHONE LINES (1984-1994)

high

220 Meso & Duncan

Figure 4: Clustering Of LDCs By Percentage Increase in Telephone Lines and Increase in Social Development Index Over The Decade 1984-1994 high

INCREASE IN SOCIAL DEVELOPMENT INDEX (1984-1994)

low

TOGO LESOTHO BANGLADESH VANUATU MALI MAURITANIA UGANDA AFGANISTAN BENIN ZAMBIA SUDAN SAO TOME ANGOLA ZAIRE ETHIOPIA

GUINEA MALAWI GUINEA BISSAU SOMALIA CHAD NIGER DJIBOUTI LIBERIA CAMBODIA SIERRA LEON HAITI MADAGASCAR TANZANIA MOZAMBIQUE

EQUITORIAL GUINEA KIRIBATI MALDIVES COMOROS BURKINA FASO KIRIBATI COMOROS YEMEN CENTRAL AFRICAN LAO RWANDA EQUITORIAL GUINEA

PERCENTAGE INCREASE IN TELEPHONE LINES (1984-1994)

high

Figure 5: Clustering Of LDC’s By Compounded Annual Growth Rate In Telephone Density and SDF high

INCREASE IN SOCIAL DEVELOPMENT INDEX (1984-1994)

low

i)

ii)

iii)

LESOTHO MALI MAURITANIA GUINEA MALAWI GAMBIA

SOMALIA NIGER

UGANDA AFGANISTAN BENIN ZAMBIA SUDAN SAO TOME ANGOLA ZAIRE ETHIOPIA

DJIBOUTI LIBERIA CAMBODIA SIERRA LEON HAITI MADAGASCAR TANZANIA MOZAMBIQUE

MALDIVES SOLOMON ISLANDS CAPE VERDE TOGO BANGLADESH VANUATU BHUTAN KIRIBATI COMOROS YEMEN CENTRAN AFRICAN LAO RWANDA EQUITORIAL GUINEA

BURUNDI MYANMAR GUINEA CHAD BURKINA FASO NEPAL

COMPOUNDED ANNUAL GROWTH IN TELEPHONE DENSITY (1984-1994)

high

A comparison of the results in these tables indicates that Majority of the LDCs exhibited a dismal level of development in information infrastructure. Sixty percent (27 out of 45 countries) of the LDCs registered a II-INDEX below the group mean (Figures 2 and 6). Further, all the LDCs in the study registered a II-INDEX of less than 25%. This means that when assessed on a 100% scale, they all fall in the lowest quartile of the assessment scale. This observation is consistent with the findings of Danowitz et al. (1995), who assessed the scope of computing in North Africa. The assessment of Odedra et al. (1993) of computing in Sub-Saharan Africa reveals the same results. 53.4% of all the LDCs in the study registered an SDI less than the group mean. Thus, about half of the LDCs performed under par with respect to social development. When the SDI is assessed on the 100% point scale, 4 LDCs fall in the highest quartile, 34 in the second highest quartile, and 9 in the third highest quartile (Figure 2). 53.4% (24 out of 45 countries) of the LDCs registered a net change in SDI (depicted by the magnitude of increase in SDI from 1984 to 1994) below the group mean of 2.51% (Figures 3, 4, 5, and 6). This result is consistent with the findings in Figure 2 where the same percentage of countries registered a 1994 SDI Figure below the group mean.

Can National Information Infrastructures Enhance Development? 221

iv)

v)

vi)

vii)

viii)

ix)

x)

80% (36 out of 45 countries) of the LDCs in this study registered an increase in number of telephone lines Figure that was below the group mean of 17.79% (Figure 3). This result indicates that there was dismal improvement in the telephone and telecommunications infrastructure of most LDCs during the period 1984 to 1994. 64% (29 out of 45 countries) of the LDCs in the study registered a percentage increase in telephone lines that was below the group mean of 2.47% (Figure 4). This result confirms that there was dismal improvement in the telephone and telecommunications infrastructure of most LDCs during the period 1984 to 1994. 55% (25 out of 45 countries) of the LDCs registered a compounded annual growth rate in telephone density that was below the group mean (Figure 5). This result confirms that there was dismal improvement in the telephone and telecommunications infrastructure of most LDCs during the period 1984 to 1994. The proportion of countries that cluster into the HIGH/HIGH segment of Figure 2 (11 out of 45 countries) is very close to that of the same segment in Figure 6 (10 out of 45 countries). The same is true for the HIGH SDI, LOW II-INDEX segment in Figure 1 (10 out of 45 countries) and the HIGH SDF, LOW II-INDEX segment of Figure 6 (11 out of 45 countries). Those countries that clustered into the HIGH/HIGH segment in Figure 1 tended to cluster into the same segment in Figure 6. This indicates that the social development policies in these countries may have been closely tied to the information infrastructure. Those countries that clustered into the LOW/LOW segment in Figure one tended to cluster into the same segment in Figure 6 (10 out of 17 countries). This result may point to the relationship between information infrastructure and the rate of social development. A high level of information infrastructure development does not guarantee a high level of social development. This is evidenced in the significant proportion of countries that registered a high II-INDEX that occupies the LOW SDF, HIGH II-INDEX segment (8 out of 24 countries) in Figure 6. A similar proportion of countries appears in the LOW SDI, HIGH II-INDEX (7 out of 24 countries) in Figure 2. The same proportion of countries is maintained in the LOW SDF, HIGH PERCENTAGE INCREASE IN TELEPHONE LINES segment of Figure 4 (7 out of the 24 countries). Figure 4 reveals the same result (7 out of the 24 countries registered high compounded annual growth rate in telephone density). These results indicate that for the information infrastructure to contribute directly to enhancing social development, there may be need for the existence a clear national policy that directly links the country’s information infrastructure to the social

Figure 6: Clustering of LDCs by II-Index 1994 and increase in Social Development Index (SDI) 1984 -1994 high

INCREASE IN SOCIAL DEVELOPMENT INDEX (SDF) (1984 - 1994)

low

LESOTHO GUINEA BISSAU BANGLADESH BURKINA FASO BURUNDI NIGER BENIN ANGOLA COMOROS ZAMBIA YEMEN RWANDA LAO TANZANIA

NEPAL BHUTAN GUINEA MALI SOMALIA

CAPE VERDE VANUATU SOLOMON ISLANDS TOGO GAMBIA

HAITI CAMBODIA UGANDA ZAIRE DJIBOUTI CENTRAL AFRICA AFGANISTAN MOZAMBIQUE

KIRIBATI SAO TOME & PRINCIPE EQUITORIAL GUINEA SUDAN LIBERIA MADAGASCAR SIERRA LEON ETHIOPIA

INFORMATION INFRASTRUCTURE INDEX (II-INDEX) 1994

MALDIVES MALAWI MYANMAR MAURITANIA CHAD

high

222 Meso & Duncan

xi)

development activities. Therefore we confirm Odedra’s (1993) assertion that: “LDCs should use IT (and information infrastructure at large) for selected and discriminated applications to bring about substantial benefits to their economies and people. The choice of applications must match the priorities set by government and have a high developmental impact.” (p. 27) Though a high II-INDEX does not guarantee a high rate of social development, a low rate of development in information infrastructure almost certainly curtails the rate of social development. This is evidenced by the fact that the majority of LDCs that registered below mean II-INDEX in Figure 2 (17 out of 27 countries) also registered below mean SDI in 1994 (17 out of 24 countries). Figure 6 confirms the results observed in Figure 2 since a similar proportion of countries (16 out of 27 countries) that recorded below mean II-INDEX also registered below mean SDF (16 out of 24) over the study period. Similar proportions of the countries studied cluster into the LOW/ LOW segment of Figure 4 (17 out of 24), hence further confirming the result.

CONCLUSION If the quality of NII substantially affects social development of a country, then we have a strong case for encouraging governments, both internal and international, to subsidize the development of NII. If NII affects social development in such a way that the quality and size of the labor pool improves, then business and industry, both national and international, will benefit likewise from contributing to the development of the NII. Our study addresses the relationship between NII and social development. It clearly establishes a link between the sophistication of information infrastructure and the rate of social development in LDCs. This relationship is significant where the infrastructure is strategically developed to support the core development mission of a country. However, accurate interpretation of the direction of causality is critical to effective investment strategies. While the analysis does not indicate direction, a logical interpretation based on previous research, infrastructure theory, and our own findings suggest that NII enables increased rate of growth of social development. To the extent that this is true, governments may justifiably link the planning of their countries’ development strategies to the development of their information infrastructure. It is reasonable for them to expect their nations to benefit significantly socially, as well as economically, from this infrastructure. The positive correlation found in this study strongly supports the unique and pivotal role that national information infrastructure may play in the future advancement of any nation. In so doing it provides a feasible and high-impact alternative through which the LDCs can leapfrog development, despite the limitations they suffer owing to their inadequate or nonexistent traditional infrastructure. The model presented in this paper proposes bidirectional relationships between economic growth, investment, and NII. The characteristics of these relationships require a great deal further research. Similarly, the causality in the relationship established between social development and NII must be studied further. We suggest that it is likely that eventually social development may affect NII. For instance, as a population becomes better educated, it may begin to demand continuous improvements in access to information for better education, more flexible ways of working, and the new avenues for economic enterprise

Can National Information Infrastructures Enhance Development? 223

available through the World Wide Web. As entrepreneurial efforts increase in a nation, business owners may begin to add to the NII because it is economically advantageous for them to do so. NII research has only begun to identify and “flesh out” the mediating and moderating variables that affect the relationships illustrated in our model. Further research on these variables and their place in the model will have profound implications for NII planning for governments and for NII theory development.

APPENDIX A: LIST OF THE LEAST DEVELOPED COUNTRIES Countries included in the study: AFGANISTAN ANGOLA BANGLADESH BENIN BHUTAN BURKINA FASO BURUNDI CAMBODIA CAPE VERDE CENTRAL AFRICAN REPUBLIC

CHAD COMOROS DJIBOUTI EQUITORIAL GUINEA ETHIOPIA GAMBIA GUINEA GUINEA BISSAU HAITI KIRIBATI

LAO PEOPLE’S DEMOCRATIC REPUBLIC LESOTHO LIBERIA MADAGASCAR MALAWI MALDIVES MALI MAURITANIA MOZAMBIQUE MYANMAR NEPAL

NIGER RWANDA SAO TOME AND PRINCIPE SIERRA LEON SOLOMON ISLANDS SOMALIA

Countries excluded from the study due to insufficient data: ERITREA SAMOA TUVALU Reproduced from UNDP .(1998). “List of the Least Developed Countries,” gopher://gopher.un.org/00/sec/dpcsd/ oscal/idc1997.txt. June 19th.

REFERENCES Antle, J. M. (1983). Infrastructure and aggregate agricultural productivity: International evidence. Economic Development and Cultural Change, 31(3), 609-619. Aschauer D. A. (1989). Public investment and productivity growth The Group Of S. Economic Perspectives, 13(5), 7-25 Berliant, M. and De, S. (1998). On the revelation of private information in stock market economies. Journal Of Mathematical Economics, 30(2), 241-256 Brancheau, J.C., Janz, B. D. and Wetherbe, J. C. (1996). Key issues in information systems management: 1994-95 SIM Delphi results. MIS Quarterly, 20(2), 225-241. Broadbent, M., Weill, P., O’Brien, T. and Neo, B. S. (1996). Firm context and patterns of IT infrastructure capability. Proceedings of the 17th International Conference on Information Systems, 174-194. Clemons, E. K. and Row, M. C. (1991). Sustaining IT advantage: The role of structural differences. MIS Quarterly. September, 274-292. Danowitz, A. K., Nassef, Y. and Goodman, S. E. (1995). Cyberspace across the Sahara: Computing in North Africa. Communications of the ACM (38:12), 23-29. Duncan, N. B. (1995). Capturing IT infrastructure flexibility: A study of resource characteristics and their measure. Journal of Management Information Systems, 12(2). Firestone, C. M. (1995). The search for the holy paradigm: Regulating the information infrastructure in the 21st Century. The Changing Nature of Telecommunications/

224 Meso & Duncan

Information Infrastructure, 34-62. Fisher, E. O. (1995). Growth, trade, and international transfers. Journal Of International Economics, 39(1-2), 143-158. Gatica, L. (1994). Liberation and tariff legislation in Chile. IEEE Communications Magazine, November, 34-35. Hayami, Y. and Ogasawara, J. (1999). Changes in the sources of modern economic growth: Japan compared to the United States. Journal of the Japanese and International Economics. 13, 1-21. Jackson, C. L. (1995). Telecommunications infrastructure from the carrier’s point of view. The Changing Nature of Telecommunications/ Information infrastructure, 199-203. Keen, P. (1991). Shaping the Future: Business Design through Information Technology. Cambridge, MA: Harvard Business School Press. King, J. and Konsynski, B. (1995). Singapore TradeNet: A Tale of One City, September 30, 487-504. Madden, G. and Savage, S. J. (1998). CEE telecommunications investment and economic growth. Information Economics and Policy, 10(2), 173-195. Malliaropulos, D. (1999). Identifying the Effects of Nominal and Real Shocks on the S&P 500 Stock Price Index, 67(3), 304-324. The Manchester School. McCallum, B. T. (1989). Monetary Economics Theory and Policy. New York: Macmillan Publishing Company. McGuire, M. C. (1987). Foreign assistance, investment, and defense: A methodological study with an application to Israel, 1960-1979. Economic Development and Cultural Change, 35(4), 847-873. McKay D. T. and Brockway, D. W. (1989). Building I/T infrastructure for the 1990s. Stage by Stage, 9(3), 1-11. Merz, M. (1999). Heterogeneous job-matches and the cyclical behavior of labor turnover. Journal of Monetary Economics, 43(1), 91-124. Miller, D. B., Clemons, E. K. and Row, M. C. (1993). Information technology and the global virtual corporation. In Bradley, S. P., Hausman, J. A. and Nolan, R. L. (Eds.), Globalization, Technology, and Competition: The Fusion of Computers and Telecommunications in the 1990s, 283-308. Boston, MA: Harvard Business School Press. Mookerjee, R. and Qiao, Y. (1997). Macroeconomic variables and stock prices in a small open economy: The case of Singapore. Pacific-basin Finance Journal, 5(3), 377-388. Mowery, D. C. and Oxley, J. E. (1995). Inward technology transfer and competitiveness: The role of national innovation systems. Cambridge Journal of Economics, 19(1), 67-93. Nidumolu, S. R., Goodman, S. E., Vogel, D. R. and Danowitz, A. K. (1996). Information technology for local administration support: The governorates project in Egypt. MIS Quarterly, June, 197-221. Niederman, F., Brancheau, J. C. and Wetherbe, J. C. (1991). Information systems management issues for the 1990s. MIS Quarterly, 15(4), 474-500. Noam, E. (1995). Economic ramifications of the need for universal telecommunications service. The Changing Nature of Telecommunications/ Information infrastructure, 161-164. Odedra, M., Lawrie, M. and Goodman S. (1993). Sub-Saharan Africa: A technological desert. Communications of the ACM, 36(2), 25-29. OECD. (1996). Information Infrastructures: Their Impact and Regulatory Requirements. Paris: OECD. Philip, G. (1988). Regulation and deregulation of telecommunications: The economic and

Can National Information Infrastructures Enhance Development? 225

political realities. Part III: Japan, The Pacific Basin and the less developed countries. Journal of Information Science Principles & Practice, 14(6), 329-334. Pietra, T.and Siconolfi, P. (1998). Fully revealing equilibria in sequential economies with asset markets. Journal Of Mathematical Economics, 29(2), 211-223 Sadowsky, G. (1996). The Internet society and developing countries. On The Internet: An International Publication of the Internet Society, November-December, 23-29. Sadowsky, G. (1993). Network connectivity for the developing countries. Communications of the ACM, 36(8) 42-47. Stewart-Smith, M. (1995). Private financing and infrastructure provision in emerging markets. Law & Policy in International Business, 26(4), 987-1011. Stone, P. B. (1993). Public-private alliances for telecommunications development: Intracorporate baby bells in the developing countries. Telecomminications Policy, 17(6), 459-469. Swaroop, V. (1994). The public finance of infrastructure: Issues and options. World Development, 22(12), 1909-1919. The Economist Intelligence Unit. (1997). Country Report 3rd Quarter 1997 Singapore. London: The Economist Intelligence Unit. The Economist Intelligence Unit. (1995). Country Report 4th Quarter 1995 Singapore, London: The Economist Intelligence Unit. The Economist Intelligence Unit. (1991). Singapore, Country Profile 1990-91: Annual Survey of Political and Economic Background. London: The Economist Intelligence Unit. The New ITU Association of Japan. (1997). World Telecommunications Visual Data Book, 1997: Telecommunications & Broadcasting in the World and ODA by Graphical Representation, Tokyo: The New ITU Association of Japan. The World Bank (1994) Social Indicators of Development 1994. Baltimore: The John Hopkins University Press. The World Bank. (1988) Social Indicators of Development 1988. Baltimore: The John Hopkins University Press. Verbrugge, Randal J. (1999). Cross-sectional inflation asymmetries and core inflation: A comment on Bryan and Cecchetti. The Review of Economics and Statistics, 81(2), 199-202. Wei, L. (1995). International technology transfer and development of technological capabilities: A theoretical framework. Technology in Society, 17(1), 103-120. Weill, P. (1993). The role and value of information technology infrastructure: Some empirical observations. In Banker, R. D., Kauffman, R. J. and Mahmood, M. A. (Eds.), Strategic Information Technology Management: Perspectives on Organizational Growth and Competitive Advantage, 547-572. Hershey, PA: Idea Group Publishing. Economic Commission for Africa. (1998). Paris Declaration for the Least Developed Countries for the 1990s : A Review Of the Progress Achieved by African Least Developed Countries in the Implementation of the Paris Declaration and Programme of Action for the Least Developed Countries for the 1990s (1990-1994), http:// www.un.org/plweb-cgi/idoc2.pl?919, February 18, 1998. Government of India. (1998). Information Infrastructure Initiatives in India. Available on the World Wide Web at: http://www.ncb.gov.sg/nii/96scan4/india2.html#p1. Palvia, P. (1997). Research, theory, and practice (Lunch keynote speech delivered at the IFIP Working Group 8.7: Information in Multinational Enterprises in Washington D. C.):

226 Meso & Duncan

http://www.people.memphis.edu/~ppalvia/speech.htm, October 1997, pp.1-11. Singapore Internet Next Generation Advanced Research & Education Network. (1998). Infrastructure Developments in Singapore. Available on the World Wide Web at: http://www.sigaren.net.sg/html/infrastructure.html. Singapore Trade Development Board. (1997). Press Release and I&E Notice. Available on the World Wide Web at: http://www.tdb.gov.sg/index.html. The World Factbook. (1994). Available on the World Wide Web at: http// elo.www.media.mit.edu/people/elo/cia/index.html. UNDP. (1998). List of the Least Developed Countries. Available on the World Wide Web at: gopher://gopher.un.org/00/sec/dpcsd/oscal/idc1997.txt. Accessed June, 19, 1998.

Contract, Control, and “Presentation” in IT Outsourcing 227

Chapter XV

Contract, Control, and “Presentiation” in IT Outsourcing: Research in Thirteen UK Organizations Thomas Kern Erasmus University Rotterdam, The Netherlands Leslie Willcocks University of Oxford, UK

Information technology (IT) outsourcing continues to experience phenomenal growth, with an estimated market size of over $100 billion in 2000. Its adoption by some of the largest international corporations has seen outsourcing become considered a key component of the information management agenda. Critical to this agenda is the formulation of comprehensive contracts. For this, legal experts and/or advisors can be consulted, but enforcement depends very much on client and vendor account managers. A theoretical analysis of the contract contrasted with empirical data from client and vendor post-contract management practice revealed that the contract has a number of purposes beyond its sole legal nature, outlining a number of control dimensions both parties aim to enforce. This paper presents findings from 13 UK-based organizations on the role of the outsourcing contract and its purpose for ensuring control over the client’s outsourcing destiny.

INTRODUCTION Information technology (IT) outsourcing continues to experience a phenomenal growth rate, especially in North America, Europe, and more recently Australia. Ever since the Kodak watershed deal in 1989, IT outsourcing has continued to mature to a status where nowadays it is considered a viable management option, making it an integral component of the information management agenda (Feeny & Willcocks, 1997; Rockart et al., 1996; Rockart & Appeared in Journal of Global Information Management, vol. 8, no. 4, 2000. Reprinted by permission.

228 Kern & Willcocks

Ross, 1995). This agenda in many circumstances deems it prudent to compare the performance of the in-house IT department with the services available externally (Willcocks et al., 1996). IT outsourcing is broadly defined as a decision taken by an organization to contract out or sell the organization’s IT assets, people and/or activities to a third-party vendor, who in exchange provides and manages assets and services for monetary returns over an agreed time period (Loh & Venkatraman, 1992; Lacity & Hirschheim, 1993). Research so far has shown that the client-vendor relationship is indeed more complex than a mere contractual transaction-based relationship (Kern, 1997; Klepper, 1994, 1995; McFarlan and Nolan, 1995; Willcocks and Choi, 1995; Willcocks and Kern, 1998). A major complexity is the near impossibility of presentiation2 of future requirements in long-term deals such as outsourcing, due to the volatility of information technology and the likely changes in user and company requirements. Suggestions have thus been made that the client-vendor relationship has to include relational contract and/or partnering dimensions (Kern, 1997; Willcocks and Kern, 1997). However, research and industry best practice has clearly shown that a central focus has to remain on the contract and hence its enforcement in the postcontract management stage (Lacity & Hirschheim, 1993; 1995). So far, little to no research has focused on explaining which contractual dimensions are eventually operationalized. Such information would allow practitioners to better understand and prescribe the contractual dimensions of the client-vendor relationship. Moreover, we conjecture that these dimensions essentially define the client company’s, and to some extent the vendor’s, control agenda over whether the major objectives of the outsourcing arrangement are being achieved. Pilot analysis of two IT outsourcing contracts revealed a number of dimensions that pervaded the post-contract management agenda. In each case the client attempted to maintain control through two detailed contractual clauses/schedules, acting essentially as a third-party judicial entity. By control we mean the process by which the client company initiates activities to assure contractually agreed terms are by the vendor’s company delivered in full and according to expectations and objectives. “Control, in other words, is aimed at ensuring that a predictable level and type of performance is attained and maintained” (Child, 1984, p. 136). In this paper we discuss and analyze the role of the contract in IT outsourcing to elucidate the post-contract management agenda. This agenda essentially prescribes the operationalization of the contract and the control dimension in IT outsourcing. Drawing upon two precedent contracts, we highlight some of the clauses being effectuated. Next, we present findings from an exploratory research study into the client-vendor relationship that reveals how organizations attempt to enforce the contract. The ensuing discussion identifies five different purposes of the contract in the client-vendor relationship, which allow us to infer that a number of contractual dimensions also define the control agenda for the client in the post-contract management stage.

IT OUTSOURCING CONTRACT The contract in outsourcing has been described as a mechanism that establishes the balance of power between the client-vendor (Lacity & Hirschheim, 1993). Contracts essentially have to be as “airtight” as possible (Lacity & Hirschheim, 1993; Fitzgerald & Willcocks, 1994abc; Saunders et al., 1997), because research has shown that vendors tend to refer to it as their chief source of obligation. Vendors, however, would prefer to see the contract as a working document (cf. EDS lawyers Hartstang & Forster, 1995), giving them flexibility to suggest improvements and new services. Clearly, this is in the interest of most vendor companies, for their goal is one of profit margins.

Contract, Control, and “Presentation” in IT Outsourcing 229

An IT outsourcing contract tends to be more complicated than other business contracts, resembling as it does a “hybrid between an asset purchase and sale agreement, and a sale/leaseback agreement, in that there is a sale of assets or transfer of operations, transfer of employees, and a lease back to the customer of the information technology services that were divested” (Halvey & Melby, 1996, p. 43). This legal complexity is evident in the detail and in the time typically invested in negotiating agreement. Third-party legal experts have for quite some time emphasized the need for a comprehensive contract, not only because it is their livelihood, but also because it basically becomes a reference point specifying how the client and vendor relate (Fitzgerald & Willcocks, 1994). The table in the appendix summarizes the main clauses of an outsourcing agreement as specified by legal experts (Burnett, 1998; Clifford Chance, 1997; Halvey & Melby, 1996; Klinger & Burnett, 1994; Mayer et al., 1996).

The Nature of the Outsourcing Contract As previously discussed, the outsourcing contract is unlike other service contracts because of the nature of what is being contracted for and the length of the contracts. This makes it extremely difficult to presentiate service provision or any other exchanges that may be needed in the future. Outsourcing contracts, and indeed most long-term contracts, have a tendency to be incomplete, which raises the possibility of opportunistic behavior by the vendor (Williamson, 1975; Hart, 1995). Macneil (1974b; 1980) in turn proposes to alleviate the incompleteness and presentiation situation through a relational, as opposed to a transactional, contract. However, there is evidence that, despite Macneil’s theorem, most effective outsourcing contracts are essentially neither completely transactional nor relational but mainly transactional intertwined with relational aspects (Currie and Willcocks, 1997; Lacity and Hirschheim, 1993; Willcocks and Kern, 1998). Actual operationalization of the written letter of any contract requires procedures that Macneil (1974ab) prescribes to the relational contract (e.g., extensive cooperation). In this paper we will focus solely on the transactional aspect of how the contract is enforced, which we identify as determining the control agenda. In the following sections we first discuss control, before looking at the post-contract management agenda in outsourcing. The analysis of two proforma precedent contracts elicits the dimensions which pervade post-contract management and further defines the ‘transactional’ level of the client-vendor relationship (or “contractual” as we have termed it elsewhere—see Kern, 1997).

THE CONCEPT OF CONTROL Control is a complex issue that has received considerable attention in the literature. Table 1 lists some important contributions. Anthony’s (1965) contribution in particular is often referred to and well known for its distinction between strategic planning, management control, and operational control. Strategic planning is defined as the process of deciding on the objectives of the organization, on changing these objectives, on the resources used to attain these objectives, and on the policies that are to govern the acquisition, use, and disposition of these resources. Anthony distinguishes strategic planning from the more management and operational control issues like Child (1984) does. Management control is defined as the process by which management assures that resources are obtained and used effectively and efficiently in the accomplishment of organizational objectives, whereas operational control focuses more on the actual efficient and effective performance of

230 Kern & Willcocks

activities. Child (1984), Eisenhardt (1985), Hofstede (1981), and Ouchi (1979) determine the context of control in organizations and determine a number of characteristics that lend themselves to define a typology. On the other hand, Boland (1979) and Orlikowski (1991) reveal how control is applied in the context of information management. Looking at the existing theories, three common dimensions can be identified according to Fischer (1993) that describe a useful typology for analyzing control in IS: focus of control (directed at whom or what), measures of control (degree of control), and process of control (means of enforcing control). Using this typology as an underlying guide, the next section presents the post-contract management agenda as the focus of control, before looking at the process of control in the empirical findings. The subsequent discussion will then allude to the degree of control in outsourcing.

POST-CONTRACT MANAGEMENT IN IT OUTSOURCING The greatest challenge that client companies face following the signing of the contract is the achievement and the enforcement of agreed terms. A number of stipulated terms will always be integral to driving the outsourcing venture forward for both sides. For the majority of client organizations we can assume they are service levels and costs (i.e., payments), whereas for the vendor it is clearly payment (i.e., profits). This control agenda defined by the contract is an integral part of post-contract management. By 1997, based on over 10 years of drawing up outsourcing contracts, several law firms had developed pro-forma precedent contracts. Here we analyze these and related sources to establish templates against which to assess our case histories and also briefly look at the often elusive concept of control. To enforce the contract, the appointed client managers and/or residual IT group are held accountable for the management and delivery of all products and services related to the outsourcing venture (Fitzgerald and Willcocks, 1994a,c). Related to this task will be the Table 1: Post-contract management agenda (adapted from Halvey & Murphy Melby, 1996). 1. Setting and/or approving IT strategy, architectural directions, and business improvements; 2. Insuring user service objectives and customer satisfaction targets are achieved; 3. Insuring quality and continuous improvements; 4. Setting and changing priorities to insure objectives of the business are met; 5. Being the focal point for determination and translation of all new business requirements necessitating vendor action; 6. Resolving disputes that arise; 7. Overseeing the vendor’s performance as specified in the agreement; 8. Monitoring overall service quality and continuous improvement initiatives; 9. Assuring proper charges and billing for the services rendered; 10. General contract administration and amendment control; and 11. Involvement in allocating new managers in vendor company to handle the account.

Contract, Control, and “Presentation” in IT Outsourcing 231

Table 2: Overview of important control literature Author & Year Anthony (1965) Boland (1979) Child (1984) Eisenhardt (1985) Hofstede (1981) Orlikowski (1991) Ouchi (1979)

Control context Management control process Control versus control over Control use in organisations Control as part of organisational design Typology of management control Impact of IT on control in organisations Typology of control

conjoint rationalization and implementation of the vendor’s processes for service changes, charges, and general financial and operational management of the contract, as well as approval of any special projects that may arise. Appointing duties and responsibility in the post-contract stage is critical, as effective management necessitates continuous communication between the responsible persons to ensure services, payments, and extra requirements are met and conflicts resolved. In our sources we found the post-contract management agenda for the client to cover the issues detailed in Table 2. For the client to accomplish this management agenda, an effective communication and operations structure has to be established in each organization and between both parties. McFarlan and Nolan (1995, p. 22) suggest interfaces have to occur at multiple levels: “At the most senior levels, there must be links to deal with major issues of policy and relationship restructuring, while at lower levels, there must be mechanisms for identifying and handling more operational and tactical issues. Both the customer and outsourcer need regular, full-time relationship managers and coordinating groups lower in the organization to deal with narrow operational issues and potential difficulties.” To ensure the various interfacing points on the vendor’s side matches the client’s operational and management style, they should ensure mutual involvement in selecting key vendor personnel.

Contract Operationalization and Control We undertook a closer analysis of some of the key clauses and schedules in a UK (Clifford Chance, 1997) and a North American (Halvey and Melby, 1996) precedent contract. The following contractual issues, for analytical purposes classified under each type of interaction, were found to demarcate the transactional focus of the post-contract management agenda.

A. Service Exchanges The most important part of every outsourcing agreement is the description of the services the vendor has to supply (Klinger and Burnett, 1994). Therefore, the service requirements have to be as detailed as possible to avoid disputes over service scope. Services are typically described in a series of exhibits, i.e., schedules. The following give an indication of what exhibits might cover: • architecture and product standards (services delivered according to these standards); • company critical services (non-achievement may result in liquidated damages); • new technology environment (new products/services and migration plan); • security services (data and/or physical security services); and • value-added services (e.g. risk/reward sharing, marketing, and others).

232 Kern & Willcocks

B. Service Enforcement and Monitoring To enforce services, performance standards and service levels have to be established and agreed for each service the vendor has to provide. Service levels typically include response times, processing priorities, systems availability used to provide services, and percentages for processing errors (Halvey and Melby, 1996). In some cases clients want to streamline the service levels through the joint efforts of the parties after the contract has been signed. In these circumstances, but also when all service levels have been specified, it is essential that the client monitors the service levels and possibly benchmarks them against previous and/or best-in-class service levels. The vendors’ non-achievement of service levels is often directly tied to liquidated damage provisions, giving clients a powerful leverage and control mechanism. Typical clauses and exhibits for service enforcement and monitoring are: • application development measure procedures (allows client to monitor development); • benchmarking (the client can compare at any time services with other services of third parties); • customer satisfaction survey (to be carried out by the vendor at certain intervals and reported to the client); and • performance-reporting requirements (reports on service performance to be provided by the vendor in intervals).

C. Financial Exchanges For the vendor to provide the services according to the agreement, the client will be charged a base fee with respect to the base services provided. Fees allocated to these base services need to be as exact as possible to avoid any confusion and dispute. Any additional services outside the previously agreed service scope (e.g., overtime, additional volume of services) will be charged according to agreed price rates. Similarly, details of fees payable to the vendor for services above baseline (e.g., hourly, daily, weekly, monthly) should also be covered to give the client flexibility. The timing of the payments has to be agreed, but they depend largely on the client’s approval. The following are typical clauses and schedules: • base fees (fees payable for base services); • expenses (previous agreement on expense budget); • incremental fees (fees payable for additional services above baseline); and • new service fees (fees payable for new services out of services scope).

D. Financial Control and Monitoring To ensure the client has some level of control over the costs, appropriate means to monitor the fees charged by the vendor need to be agreed. Klinger and Burnett (1994) suggest the client must have the right to audit and request verification of the charges put forward by the vendor at any time. Additionally, the client can enforce liquidated damages in circumstances of nonperformance of service levels. Other suggested means included: • audit (access to financial information and charges); • liquidated damages (for failing to meet service levels); • open book accounting (vendor keeps an open book of costs and charges accessible by the client); and • price lists (list or catalogue of prices of different services, hardware, software and technology that the client can request from the vendor).

Contract, Control, and “Presentation” in IT Outsourcing 233

E. Key Vendor Personnel Of fundamental importance to the client will be to know who is responsible for certain services, areas, or technology in the vendor company, and vice versa. The precedent contracts suggest that key vendor employees will be explicitly listed in the contract as remaining with the account for specified time periods, or equivalent staff made available (Clifford Chance, 1997). The client should have approval rights over vendor employees and their length of involvement with the account. Furthermore, the vendor has to disclose to the client how it will structure its organization to be able to respond to the client’s needs. Both the key contact points and the structure are fundamental to ensure communication. The contract will address these under the following terms or schedules: • supplier personnel and key employees; and • vendor account management team structure.

F. Dispute Resolution Typically the contract will give guidance on an informal resolution mechanism to deal with disputes. If disputes cannot be resolved in the first instance, the contract describes an escalation procedure and the appropriate persons, groups, or committees to contact. As a final resort, a neutral third-party advisor or even a judgement is suggested in the contract to resolve a dispute: • dispute resolution (details of procedures for resolution).

G. Change Control and Management Finally, to ensure the contract allows for flexibility to cater for changes in user and company requirements, the contract specifies change procedures the client can initiate formally. For this a change request has to be issued detailing the changes required from the vendor. This will most likely affect pricing arrangements and possibly take effect with a time delay. The agreed procedures will be specified in a schedule referred to something like: • change control procedures. The analysis revealed that service exchanges, service enforcement and monitoring, financial exchanges, financial control and monitoring, and change control and management circumscribe the nature of the ongoing transactional processes between both parties. Secondly, it identifies in the vendor company the key contact points and provides an overview of the account management team’s organizational structure describing some of the structural factors of the client-vendor relationship, whereas the dispute resolution method describes some of the skills the managers handling the relationship must have. Combined, these arrangements will come to bear upon the client-vendor relationship and describe the focus of the control agenda in post-contract management.

FIELD STUDIES Findings from the research portray in detail those issues that play a crucial part in client organizations’ attempts to control their outsourcing arrangements. Our analysis suggests that maintaining control over the outsourcing venture through contractual means was seen as particularly important. The general concern following hand over was loss of control and

234 Kern & Willcocks

over-dependency on the vendor. Clients were thus keen to affirm their control in the venture, through a number of different contractual dimensions and considerations. Although we recognize the study focuses on a small and geographically distinct sample where a particular set of contracting practices apply, there are a number of learning pointers applicable to outsourcing contract management in general.

Research Approach In early 1997 we undertook research into client and vendor relationship practice in 13 organizations based in the United Kingdom. Using a semi-structured interview protocol a series of interviews were undertaken with a range of participants, including lawyers, IT managers, contract managers, account executives, general managers, and support managers in both customer and vendor organizations. Questions addressing the contract, post-contract management, relationship management, the nature of the working relationship, and the evolution of a relationship were posed, with a strong emphasis on what characteristics influenced the operationalization of the contract. These are some of the key questions asked: • What role does the contract play in the relationship? Have you had to refer, enforce it, or use it in anyway in the relationship so far? • Could you describe the state of the relationship? What operational difficulties have you encountered? • What were some of the major milestones, achievements and/or developments in the relationship? Examples? • Are you achieving your expectations and outsourcing intentions? Why not? • What are the upcoming challenges for the relationship? While we made no judgement as to whether the 13 contracts were successful or otherwise, we did need to be in a position to assess whether specific contractual practices were experienced as essential, helpful, or otherwise. The interviews were scheduled for one hour but in many cases lasted anywhere up to three hours. All of the interviews were tape-recorded and transcribed, after which the responses from the client and the vendor companies were grouped together into subject categories by applying a “data display” method (Miles & Huberman, 1994). The resulting checklist matrices of subject categories were then classified into areas of agreement and commonality, and into sets of disagreement and problems. The areas of agreement that illustrated a within-group similarity (Eisenhardt, 1989) identified those variables which underpinned the outsourcing relationship and also provided the means for further subjective cross-case analysis. In some cases it was possible to cross-case analyze a client company’s response with its respective vendor company’s response. The interviews formed the basis for a number of case studies, which were corroborated by the collection and the ensuing analysis of relevant documentation, including internal memos, company information, minutes of meetings, and outsourcing contracts. Tables 3, 4, and 5 present an overview of the companies interviewed, clients’ perceptions of their relationship focus, and subjective third-party ratings of the partnering capabilities of the vendor companies interviewed. Tables 3, 4, 5 are constructed from interview responses, as indicated.

Contract, Control, and “Presentation” in IT Outsourcing 235

FINDINGS The Contract Contracts were found to stipulate at length the terms and obligations the vendor and the client had to fulfill. Their ultimate function was explained by their legal nature, allowing the client and/or vendor to produce the contract in court, in cases of, for example, dispute or termination (Solicitor, Service A). This legal nature of the contract3—in the United Kingdom—makes it near impossible to include non-legalistic terms such as trust, cooperation and so on in the contract (Partner, Service A). The vendor could not commit to these terms contractually because they did not know what the client’s requirements are. In essence then, the contract outlines the bare bones of the deal “which is the formal structure of the relationship on which you might actually rely if it all collapsed and you had to seek legal redress” (Partner, Service A). The contract was found to build the foundation on which the ensuing client-vendor relationship is then based (Business Director, Vendor E; Managing Director, Vendor A). It “defines how you are going to work more than anything else, but you then still have to make it work. This is just paper, it’s people that make things work. This gives them the guidelines, the stepping stones, the structure” (Management Service Manager, Client C). Getting the contract, i.e., the foundations right (Group IS Manager, Client B), is vital, because a badly formulated contract can lead to frustration and in some circumstances—as Client A experienced—lead to costly renegotiations, and eventuate in possible failure.

Table 3: Client companies Client Company & Interviewee

Industry

Annual Turnover

Origin

Outsourced

Client A Business Support Manager Client B Group IS Manager

Retailing & Stores

£780mn1 (1995)

British

Total

Chemicals Manufacturer

£10bn 2 (1996)

British

Client C Management Services Manager

Property Investment & Development

£472mn 1 (1995)

British

Client D Economic Analyst

Aerospace Manufacturer

£3.5bn 2 (1995)

British

Client E MIS Executive

Motor Car Manufacturer

£397mn 2 (1995)

British

Client F IT Coordinator

Electronics Manufacturer

£270mn1 (1995)

Japanese

Client G Corporate IT Adviser

Oil, Gas, & Nuclear Fuels

£453mn 2 (1995)

Dutch/ British

Selective (Europe) 1. telecomms network 2. data center, software support & legacy systems 3. desktop systems & phones Selective 1. hardware & software maintenance 2. legacy system, & software development Total & major business process reengineering programs Selective 1. software development & IT operations 2. systems integration 3. global networking Selective legacy & operating systems Selective & Total (Global) 1. desktop computing, networking, & others 2. Total

1

Start of Deal

Length of Deal

Size of Deal

No. of People Transferred 120

Relationship 3 Focus (1997)

1993

10 years

£1bn

1. 1994 2. 1995 3. 1997

1. 3 years 2. 5 years 3. 4 years

1.