List of Contributors: - Ukurova University, Adana, Turkey

LIST OF CONTRIBUTORS
Catherine N. Axinn Marketing Department, College of Business, Ohio University, Athens, OH, USA School of Marketing, University of Technology Sydney, Australia College of Business and Economics, Australian National University, Canberra, Australia Nottingham University Business School, The University of Nottingham, Nottingham, UK Marketing Department, College of Business, Ohio University, Athens, OH, USA Manchester Business School, The University of Manchester, Manchester, UK Groupe Sup de Co Montpellier, Montpellier, France Department of Marketing, University of Dortmund, Dortmund, Germany Nottingham University Business School, The University of Nottingham, Nottingham, UK Department of Business Administration, C ukurova University, Adana, Turkey ix
Nigel J. Barrett Stephen Chen
Thomas Chesney
Dawn R. DeeterSchmelz Pervez Ghauri
Ca Gura lin u Hartmut H. Holzmu ller Kathryn Houghton
Bahattin Karademir
Jorma Larimo Kannika Leelapanyalert
Department of Marketing, University of Vaasa, Vaasa, Finland Department of Business and Service Sector Management, London Metropolitan University, London, UK Department of Marketing, University of Dortmund, Dortmund, Germany Department of Marketing & Management, University of Southern Denmark, Odense M, Denmark College of Business, Florida Atlantic University, FL, USA University of Economics, Ho Chi Minh City, Vietnam; and School of Marketing, University of Technology, Sydney, Australia Vietnam National University, Ho Chi Minh City, Vietnam Department of Business Economics and Management, University of Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain School of Business Administration, Wayne State University, Detroit, MI, USA Department of Business Research, University of Pavia, Pavia, Italy Department of Management, University of Valencia, Valencia, Spain
Patrick Lentz Tage Koed Madsen
Michael R. Mullen Tho D. Nguyen
Trang T. M. Nguyen Aristides Olivares-Mesa
Richard N. Osborn
Giada Palamara Jose Pla Barber
List of Contributors
xi
Ashok Ranchhod
Southampton Business School, Southampton Solent University, Southampton, UK Department of Marketing & Management, University of Southern Denmark, Odense M, Denmark Department of Business Economics, College of Economics and Business Sciences, Autonomous University of Barcelona, Barcelona, Spain Department of Business Economics, College of Economics and Business Sciences, Autonomous University of Barcelona, Barcelona, Spain Department of Management, University of Valencia, Valencia, Spain University of Applied Sciences (FHM), Department of Marketing, Bielefeld, Germany Department of Marketing & Management, University of Southern Denmark, Odense M, Denmark College of Business, Florida Atlantic University, FL, USA Brand New Era, Caneld, OH, USA Department of Business Economics and Management, University of Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain
Erik S. Rasmussen
Alex Rialp Criado
Josep Rialp Criado
nchez Peinado Esther Sa Eric Schirrmann
Per Servais
Shirley Ye Sheng Brian T. Straley Sonia Suarez-Ortega
xii
Heidi Winklhofer
Nottingham University Business School, The University of Nottingham, Nottingham, UK School of Business Administration, Wayne State University, Detroit, MI, USA Brand New Era, Caneld, OH, USA Department of Business Research, University of Pavia, Pavia, Italy
Attila Yaprak
Ernest J. Zavoral, Jr. Antonella Zucchella
LIST OF REVIEWERS
Arild Aspelund Norwegian University of Science and Technology, Trondheim, Norway George Balabanis Cass Business School, City University, London, UK Colleen Collins-Dodd Faculty of Business, Simon Fraser University, Burnaby, BC, Canada Michael Czinkota McDonough School of Business, Georgetown University, Washington, DC, USA Richard Fletcher University of Western Sydney, Penrith South, NSW, Australia Esra Gencturk College of Administrative Sciences and Economics, Koc University, Istanbul, Turkey David Grifth The Eli Broad Graduate School of Management, Michigan State University, East Lansing, MI, USA Nukhet Harmancioglu Sawyer School of Management, Suffolk University, Boston, MA, USA
xiii
Marian V. Jones Department of Business and Management, University of Glasgow, Glasgow, Scotland, UK Gary A. Knight Department of Marketing, College of Business, Florida State University, Tallahassee, FL, USA Luis Filipe Lages Universidade Nova de Lisboa, Lisbon, Portugal Peter W. Liesch UQ Business School, The University of Queensland, St. Lucia, Queensland, Australia Paul Matthyssens Department of Marketing, University of Antwerp, Antwerpen, Belgium Oystein Moen Norwegian University of Science and Technology, Trondheim, Norway Ben Oviatt Department of Managerial Sciences, Robinson College of Business, Georgia State, Atlanta, GA, USA Piet Pauwels University of Maastricht, Maastricht, the Netherlands
xiv
LIST OF REVIEWERS
Marisa Ram rez Aleson Department of Business Economics and Management, College of Economics and Business Sciences, University of Zaragoza, Zaragoza, Spain Bernard Simonin The Fletcher School, Tufts University, Medford, MA, USA
Carl A. Solberg Norwegian School of Management, Sandvika, Norway Sharon V. Thach College of Business, Tennessee State University, Nashville, TN, USA Chris White The Eli Broad Graduate School of Management, Michigan State University, East Lansing, MI, USA
PREFACE
This special volume of Advances in International Marketing originated from many interesting papers that were presented at the 2005 Annual Meeting of our CIMaR (Consortium for Internatinal Marketing Research) network. The hosts of this meeting, Professors Alex Rialp and Josep Rialp served as guest co-editors. We are delighted to feature the latest research ndings and insights contributed by many authoritative colleagues from around the world. Our thanks go to Professors Alex Rialp and Josep Rialp for their efforts in creating this volume. They issued a call for papers, which then attracted a variety of submissions of high quality. We owe gratitude to them for screening and evaluating these submissions, and for preparing the nal set of chapters. We are also indebted to many colleagues who assisted in the review process. The resulting selections draw from a variety of perspectives and offer rich insights. At Michigan State University, I would like to recognize the professional assistance of Ms. Kathy Waldie, editorial assistant for the Advances in International Marketing series. Kathy carries the responsibility of corresponding with the authors, guest editors as well as the staff of Elsevier Science at various phases of the publication process. Finally, the co-authors and I express appreciation to Dr. Helen Collins, Ms. Joanna Scott, Ms. Julie Walker, Mr. Philip Tite and the other staff at JAI/Elsevier Science who saw the volume through the production process. S. Tamer Cavusgil Series Editor
xv
INTERNATIONAL MARKETING RESEARCH: OPPORTUNITIES AND CHALLENGES IN THE 21ST CENTURY

Alex Rialp and Josep Rialp
INTRODUCTION: STATE-OF-THE-ART AND OPPORTUNITIES AND CHALLENGES FOR INTERNATIONAL MARKETING RESEARCH IN THE 21ST CENTURY
According to a recent and interesting revision of advances in international marketing theory and practice, the international marketing literature has grown exponentially in recent years in order to offer sufcient support to corporate and public policy makers confronting todays hostile global business conditions (Katsikeas, 2003a). In fact, some of the most relevant academic journals in this eld (Journal of International Business Studies, Journal of International Marketing, International Marketing Review, International Business Review, Advances in International Marketing, among others) can be considered highly stable and mature publications, with research articles covering a wide range of topics within the international marketing domain and usually authorized by leading contributors to other
International Marketing Research: Opportunities and Challenges in the 21st Century Advances in International Marketing, Volume 17, 113 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1474-7979/doi:10.1016/S1474-7979(06)17019-2
ALEX RIALP AND JOSEP RIALP
high-ranking marketing journals (DuBois & Reeb, 2000; Malhotra, Wu, & Whitelock, 2005). However, according to some of the most outstanding critical assessments carried recently on the conceptual foundations, research traditions or earlier development, and future research agenda regarding the discipline of international marketing as a eld of study (Cavusgil, 1998; Czinkota & Ronkainen, 2003; Katsikeas, 2003a, 2003b; Balabanis, Theodosiou, & Katsikea, 2004; Cavusgil, Deligonul & Yaprak, 2005; Douglas & Craig, 2006) some promising research avenues are still open to further academic research in this discipline. For instance, Cavusgil (1998) proposes the following research agenda for international marketing: (1) mainstream marketing management issues in the international context; (2) special challenges in international marketing; (3) marketing engineering: performance in international marketing; (4) dynamic analysis of rm expansion in international markets; (5) interrm partnering in international markets; (6) internationalization process of rms; (7) government promotion of international business activity; (8) marketings interface with other functions; (9) comparative studies of marketing executive behavior; and (10) research methods in international marketing. As this same author brilliantly stands out, it is essential that we begin an open and constructive dialogue about what is important, how to go about adding to knowledge, and how to enhance best practices in international marketing (Cavusgil, 1998, p. 112). In their well-known international marketing manifesto, Czinkota and Ronkainen (2003) postulate that the eld of international marketing has already and can continue to make major contributions to the improvement of society. Seven propositions in support of a lively debate for the sake of a renaissance of this eld are provided and illuminated by these authors: (1) remember the roots and purpose of the eld, (2) resist the temptations of overspecialization, (3) work with a new paradigm and new methods, (4) look to the World, (5) maintain the dialogue with all possible constituents, (6) work also with those who place or show, and (7) profess expertise. Also, in an effort to isolate still remaining problems and issues underlying Czinkota and Ronkainens manifesto and suggest ways of pushing their propositions deeper, Katsikeas (2003b) stands out the lack of attention afforded to examining outcomes of rms international marketing activities and the need of incorporating specic company performance issues in this eld. Similarly, Cavusgil et al. (2005) take into account the fundamental changes currently taking place in the business global environment and in the business enterprise itself which compel international marketing scholars to continuously re-examine the progress being made by the elds researchers
International Marketing Research
in developing knowledge. By critically evaluating progress in international marketing as a eld of study through ontological, thematic, and methodological perspectives, these authors come up with a portfolio of research topics worthy of further scholarly attention. In this way, further theory development and reframing in the eld becomes particularly a must (Rialp & Rialp, 2001). According to Axinn and Matthyssens (2001a, 2001b), existing theory in internationalization is insufcient to explain the currently observed behaviors of rms in the international business marketplace (i.e., the increased speed of internationalization, the limits of psychic distance, the range of foreign entry modes accommodated by the rm, and so on). Accordingly, some new and challenging issues such as the increasing impact of the global economy, the service economy, the new knowledge-based economy, the high technology and connected knowledge/network economy, and the customer value-based economy are not only changing the shape of international business behavior, but also casting doubt on the applicability of traditional internationalization theories. Also, Douglas and Craig (2006) are of the opinion that international marketing research plays a vital role as rms expand globally. Yet limited attention, according to these authors, has been paid to the conceptual underpinnings of research needed to guide such a foreign expansion. Accordingly, these authors develop rened conceptual frameworks that are indeed capable to guide further research in the eld, also dealing with more appropriate unit of analysis selection and constructs measurement. In a previous international marketers-oriented work, these same authors (Craig & Douglas, 2001) had already identied four key areas where progress at conducting international marketing research had to be made. First, international marketing research efforts need to be more closely aligned with market growth opportunities outside the industrialized nations. Second, international marketing researchers must develop the capability to conduct and coordinate research that spans diverse research environments. Third, they need to develop new and more creative approaches to probe the cultural underpinnings of behavior. Finally, technological advances, such as the Internet, need to be incorporated into the research process in order to facilitate and expedite research conducted across the globe. Another traditional research sub-area in the eld, clearly demanding some more renement and re-elaboration, both from the theoretical and empirical standpoint, is, according to Balabanis et al. (2004), export marketing research. For these authors, rapid technological, institutional, legislative, economic, and attitudinal changes across the globe pose critical challenges
(but also, new opportunities) for the future development of export marketing research. This research should focus on the identication of the right export marketing capabilities that rms should develop or acquire, the ability to leverage or transfer them across international markets, and the ability of constantly upgrading them. Also, of critical importance are the processes currently used to develop such capability-based international (export) marketing strategies and to manage relationships with international customers and partners. In addition to this, entrepreneurship is becoming a growing phenomenon in World markets. Therefore, a new and highly related eld of inquiry in which further contributions from (international) marketing scholars as well as those provided by traditional entrepreneurship researchers are expected to widely increase in the years to come is referred to the emerging discipline of international entrepreneurship (Etemad & Right, 2003; Styles & Seymour, 2006). In Styles and Seymours (2006, p. 126) own words: there is considerable scope for marketing academics to contribute to the nascent eld of international entrepreneurship which would, in turn, advance marketing theory. In particular, some relevant analytic perspectives adopted in this international entrepreneurship eld (basically the result of mixing entrepreneurship and international business/marketing disciplines), are currently referred to the emergence and consolidation, both in traditional and hightech sectors, of the so-called born-global rms, global start-ups, and/or international new ventures (Knight & Cavusgil, 1996; Madsen & Servais, 1997; Rialp, Rialp, & Knight, 2005; Rialp, Rialp, Urbano, & Vaillant, 2005) as well as the interface of the Internet and rm entrepreneurial behavior in international markets (Sinkovics & Bell, 2006). With this general assessment of the current status of the international marketing discipline in mind and, more in particular, taking into account the diverse challenges and opportunities that can be associated with further research in this still very promising eld of scientic inquiry, we describe hereon the origin, structure, and contents of this Special Issue of Advances in International Marketing.
THE SPECIAL ISSUE: ORIGIN, STRUCTURE AND CONTENTS

The origin of this Special Issue of Advances in International Marketing is found in an international conference of the Consortium for International Marketing and Research (CIMaR) that took place in Barcelona, Spain on
May 2831, 2005. Sponsored and co-organized by the Center for International Business Education and Research at Michigan State University (E. Lansing, Michigan) and the Department of Business Economics at the Autonomous University of Barcelona (Spain), the conference featured about 30 accepted papers and works in progress. In addition to this meeting research outcomes, a subsequent call for papers for a Special Issue of Advances in International Marketing with the title International Marketing Challenges in the 21st Century was launched in October, 2005 with the main objective of providing a new setting for scholars and academics around the globe to exchange and/or share their most current research interests and ideas related to this evolving and challenging scientic discipline of international marketing. In particular, given some recent and important trends of change being present today in the global business sphere (for example, consolidation of emerging sectors, more rapid pace of technological change and business internationalization processes, greater integration and interconnectedness of international economies and rms, converging buying behavior, recent but critical advances in telecommunications and transportation facilities, among others), a critical assessment and further explanatory research of the expected inuences, in the form of both new research opportunities and challenges in the eld, associated with these phenomena in the years to come had to be newly approached. Both conceptual as well as empirical papers were highly welcome for publication consideration for this volume. Empirical papers might employ quantitative and/or qualitative (e.g., case study) methodologies, but in any case submissions for this special issue had to be of high and robust academic rigor. For some CIMaR participants in the last meeting in Barcelona, this meant an excellent opportunity to re-submit a revised version of their earlier submissions (papers, research proposals, or works in progress), and/or any other paper tting the publications philosophy and guidelines. Nevertheless, all other researchers in this eld were also kindly invited to submit their papers for this special issue. As a result of this call for papers, a considerable number of submissions were received which were then submitted to a rigorous, double-blind review process. Although there were several other papers that could have been included in this volume, reviewers evaluations (and corresponding contributors revisions), space limitations, and other conditions, such as achieving enough geographical dispersion of authors and appropriate balance among topics covered by the different studies, led to our nal selection of the 15 papers presented in this volume which were delivered by more than
30 contributors in the discipline of international marketing from all over the World. In addition to the work of scholars from the United States, other papers developed by researchers in many other countries in this eld are also included. Most of these papers report on individual, rm, or institutional behavior in a wide array of World economies ranging from Australia, Denmark, Germany, or UK to Italy, Finland or Spain, and from Turkey to Vietnam, among others. Accordingly, we believe that this collection represents a comprehensive treatment of the contemporary issues and problems being currently faced mostly by small- and medium-sized rms in different type of sectors (including manufacturing industries and services) in both developed and emerging, developing economies. Thus, as regards to its structure and contents, apart from this introductory chapter, the special issue consists of 15, double peer-reviewed manuscripts covering very relevant topics in contemporary international marketing research, that we believe will become of very great interest not only for both academics and practitioners, but also for policy makers in this eld. The issue is structured in four different, but at the same time related, sections which proceed from the more general issues of the export and internationalization processes toward some more specic issues and applications both at the environmental and technological levels: (1) export behavior, development process, and performance (four chapters); (2) strategic internationalization process in different sectoral settings (three chapters); (3) environmental inuences and emerging markets for international marketers (three chapters); and nally (4) business internationalization and information technologies (ve chapters). Hereon, we briey introduce the 15 selected chapters for this volume of Advances in International Marketing according to the four different parts or sections in which they have been assigned.
PART I: EXPORT BEHAVIOR, DEVELOPMENT PROCESS, AND PERFORMANCE

In the opening piece of this volume, Jorma Larimo explores past and current research related to rm export performance a topic in which many studies have been conducted to date with mixed results. This new empirical paper is aimed at analyzing (1) the impact of the selected rm, management, and the export strategy-related variables on export performance; (2) the possible variation existing in the results depending on the different types of measuring this export performance dimension; and (3) the similarities and
differences in the results depending on the type of SME traditional exporters vs. born international companies. The study is based on a survey conducted among Finnish SMEs in early 2002. While none of the hypotheses were fully supported by all of the six measures of export performance being employed, this dimension was positively impacted by rm size, product/service quality, international orientation, and market diversication along ve measures. Additionally, the study indicates some similarities and differences depending not only on the measure selected for export performance, but also on the type of the exporting SME, and the operationalization being used for the born international companies. The second paper in this section, by Antonella Zucchella and Giada Palamara, conceives that small rms can approach foreign markets and reach high levels of export intensity combined with a broad geographic scope in spite of their limited resources just by adopting a niche strategy. Such a global niche approach also help explain, among other factors, why and how new rms can become international or even global since their inception. By means of applying case study analysis, this paper shows a positive relation between niche strategy and high international performance, in terms of export intensity, precocity, speed, and market scope. The international expansion of such niche-oriented rms is based on a horizontal microsegmentation of the global market: they move internationally following global customers, regardless the psychic/geographical distance in play, and compete mostly on a non-price basis. rez investigate, in the following contriArstides Olivares and Sonia Sua bution of this section, the issue of entry timing in the export development process of, in this case, Spanish manufacturing rms. This process is conceived as a sequential path along the following export stages: (1) the preengagement phase, where rms do not export; (2) the initial phase, where rms export by means of an agent; and (3) the advanced phase, where rms export via a sales subsidiary. This study is then focused on the type of factors which can accelerate or decelerate the decision to entry in and/or change across these phases. Event history analysis is applied to a data set comprised by 1,478 rms in 2002. Results indicate that the development of product or process innovations becomes the most signicant motivation for an early entry in the initial and advanced phases of the export development process. In addition, network ties, a broader scope of products, rm size, and foreign ownership participation are also key factors in accelerating entries in this process. Finally, the paper ending this rst section focused on export behavior and performance, authored by Trang T.M. Nguyen, Nigel Barrett and Tho
D. Nguyen, examines the roles played by market and learning orientations in relationship quality between exporters in transition economies and their foreign importers and subsequently, export performance. A random sample of 283 export rms in Vietnam provides enough evidence to support the hypothesized main effects. The results further indicate that learning orientation plays a role in building high-quality relationships for both new and mature relationships. However, the impact of market orientation on relationship quality is found only in the new relationship. In addition, rm ownership structure does not seem to moderate the relationships between learning orientation, market orientation, relationship quality, and export performance.
PART II: STRATEGIC INTERNATIONALIZATION PROCESS IN DIFFERENT SECTORAL SETTINGS

The following contributors, Ca Gura and Ashok Ranchhod, examine how lin u the accelerated globalization of World markets in the last three decades has dramatically increased the importance of internationalization models even more, both from an academic and a practitioner perspective. Actually, such internationalization process shows major implications for the strategic orientation of small rms. However, these authors contextualize this phenomenon in relation with the specic characteristics for various market environments and industrial sectors, as for instance, high-tech ones. Accordingly, by means of a comparative analysis of the internationalization processes of UK and US biotech SMEs, this study shows the impact of the domestic market prole on this process, outlining also the similarities and the differences that can be observed between these two countries. nchez and Jose Focusing on non-traditional economic activities, Esther Sa Pla stand out that, despite the increasing importance of the service sector in developed economies and the growth of foreign investments in this sector during the last decade, few studies have undertaken to empirically analyze the factors inuencing foreign entry mode choice in this context. For these authors, the special characteristics of services increase the complexity of the analysis and, thus, traditional explanations of international entry mode choice in manufacturing sectors may need to be complemented by other moderating inuences. Based on 174 entry decisions of service rms, their results suggest the importance of including strategic variables and the specic nature of services to understand such a complex phenomenon, which is not always associated just with efciency and value-based considerations but also with strategic issues and industry characteristics.
Also in the context of international market entry strategies for service providers, Kannika Leelapanyalert and Pervez Ghauri acknowledge that numerous studies have focused on retailing rms and their activities abroad; however, these have not been able to fully identify the factors that inuence the process of retail internationalization. Thus, their paper examines the factors that inuence the foreign market entry process in retailing rms and develops a conceptual model, which is then used to analyze two case studies: the entry strategies followed by IKEA in China, and by Marks & Spencer (M&S) in Hong Kong. By fully examining these two business experiences, the authors provide insights into the type of factors inuencing the foreign market entry process and how other rms can manage it.
PART III: ENVIRONMENTAL INFLUENCES AND EMERGING MARKETS FOR INTERNATIONAL MARKETERS
In the paper that introduces the third section of this volume, Michael R. Mullen and Shirley Ye Sheng complement and extend a growing body of work developing and using overall market opportunity indexes (OMOIs) to rank the attractiveness of potential foreign markets. Assuming that rms in most industries must look to expand into international markets to survive and thrive, the index developed in this paper assesses countries market potential beyond the traditional measures of market size and economic development also by including political risk, economic freedom, telecommunications as well as physical infrastructure and geographic distance. Accordingly, the authors provide a current and detailed analysis of market attractiveness and opportunity for the largest set of countries indexed and ranked to date. The validity of the index is also assessed by comparing the ranking of market opportunity to actual subsequent trade ows from the US. The modied OMOI is shown to be a exible, valid, and fairly stable tool for preliminary analysis of foreign market opportunity. Acknowledging the popularity of country-of-origin research in international marketing, but transferring it to new and unexplored context, Patrick Lentz, Hartmut H. Holzmu ller, and Eric Shrirrmann focus on the lack of attention usually been paid to effects which stem from the declaration of a products local origin. In this study, insights from country-of-origin research as well as exploratory qualitative studies are used to model determinants of preference for local products. Conjoint analysis and structural equations results based on a sample of consumers from three neighboring cities in
10
Germany conrm the importance of local origin for product preference as well as of the mechanism of such city-of-origin effects. Finally, Attila Yaprak, Bahattin Karademir, and Richard N. Osborn approach the issue of how do business groups function and evolve in emerging markets by analyzing the case of Turkish business groups. According to these authors, business groups have not received sufcient attention in the international marketing literature, though they have become a signicant phenomenon in the evolution and functioning of emerging markets. They also provide important local partnership opportunities to international marketers. Accordingly, this paper provides a general overview of the theories that explain how business groups function and evolve in these markets with the aim of generating subsequent propositions from that theory. Evidence on such business groups evolution in Turkey, which is taken as an illustration of one emerging country market, is presented and discussed in detail by these authors.
PART IV: BUSINESS INTERNATIONALIZATION AND INFORMATION TECHNOLOGIES

The last section of this volume is more specically devoted to the high degree of impact that the emergence and consolidation of the Internet, among other information technologies, is currently having on small rm internationalization and international marketing strategies, such as exporting. In the paper introducing this last section, Per Servais, Tage Koed Madsen, and Erik S. Rasmussen conceive e-business as a very important and revolutionary business tool, also relevant for small- and medium-sized rms (SMEs) aimed at internationalizing. With e-business and the Internet solutions, borders between countries are becoming less relevant, and more direct interaction between separate businesses becomes possible. In this chapter, the authors unravel the use of the Internet by different types of rms. First, a categorization of different local and international rms is presented, and then the level of Internet usage by the so-called born global rms as compared to the usage of this tool by other types of rms is analyzed. According to the results of this study, born globals make use of the Internet to convey their market presence, but only to a limited extent they sell their products electronically abroad. Instead, Internet help them support already existing actions by also describing their products on web pages, offering services related to their products, facilitating product development, and building and maintaining relations to foreign customers. Thus, this
11
article sheds some prior light on the key question of how small rms in general, and born global rms in particular, will continue to adapt the Internet technology in practice, though much more research on this issue is expected by these authors. In a similar vein, Stephen Chens chapter examines to what extent Internet-based rms have indeed globalized and the key factors that have enabled some rms to globalize more than others. Contrary to arguments that Internet-based rms automatically benet from a global market, this study shows that most Internet rms serve regional markets. However, the author nds a few notable exceptions. Interestingly, in these cases, a combination of early-mover advantages, unique product, technology standards, and complementary products and services seem to have created what this author calls a winner-takes-all market in which a few rms dominate markets worldwide. The following paper is jointly authorized by Catherine N. Axinn, Dawn R. Deeter-Schmelz, Brian T. Straley, and Ernest J. Zavoral Jr. Drawing from seminal research on organizational buying behavior, these authors make use of the exploratory case study approach to explore the impact of the Internet and internationalization on todays industrial procurement processes. In particular, by means of conducting several interviews with senior managers of an industrial distributor, a number of key insights and implications for future research regarding the impact of the Internet on buyersupplier interactions and the importance of global sourcing are revealed in this chapter. The following selected paper, co-authorized again by Tho D. Nguyen and Nigel Barrett, starts from the assumption that the Internet is a crucial source of information that can be transformed into knowledge. The authors of this study develop an Internet-based knowledge internalization process in which internationalizing rms in transition markets utilize the Internet to search for information about foreign markets, to assess its relevance, and then to internalize it for their internationalization purposes. It is also found that such a process underlies international orientation and foreign sales intensity which in turn, has a reciprocal effect on it. Further, learning orientation also facilitates the Internet-based knowledge internalization process. According to the authors, these ndings suggest that internationalizing rms should promote and value this process in order to mitigate their common lack of foreign market knowledge. Finally, the research paper that closes this section and the volume, coauthorized by Heidi Winklhofer, Kathryn Houghton, and Thomas Chesney, is focused on the drivers and inhibitors determining how advanced websites of SME exporters are. According to these authors opinion, despite the much
12
publicized advantages of a website for SME exporters, the level of website sophistication, as well as the factors which inhibit or stimulate exporting SMEs to develop their website beyond a basic level of sophistication, are still unknown. The literature is prone to discuss website establishment and development simultaneously, splitting rms into adopters and non-adopters, yet websites may be established and then neglected, or be continually developed. Accordingly, their paper introduces an instrument for measuring website sophistication within an export marketing context, and proposes and empirically tests a model that depicts factors impacting on perceived advantages of a website and its sophistication levels. The results of this empirical study identify export diversity and environmental pressure as the key determinants of perceived advantage of a website which in turn is a good predictor of website sophistication. However, the rms internal resources, i.e., information and communication technology (ICT) knowledge and time, in conjunction with entrepreneurship orientation, also determine small- and medium-sized exporting rms website sophistication level. Once we have briey introduced the different research articles selected for this Special Issue of Advances in International Marketing as a way of identifying and discussing emerging opportunities and challenges in this current century for academics and practitioners in the international marketing eld, we sincerely hope that you will enjoy reading and reecting on the work presented in this volume, at least as much as we enjoyed composing it!
ACKNOWLEDGMENTS
As the guest co-editors of this volume, we would like to especially acknowledge the series editors of Advances in International Marketing, S. Tamer Cavusgil and Kathy Waldie, from Michigan State University, for the inspiration, continued guidance, and leadership they have always provided to us as well as all of the authors and reviewers involved in this volume for having played an excellent role contributing not only with their precious time, but also with their great effort and talents.
REFERENCES
Axinn, C. N., & Matthyssens, P. (2001a). Reframing internationalization theory: An introduction. In: C. N. Axinn & P. Matthyssens (Eds), Reassessing the internationalization of the rm. Advances in international marketing, (Vol. 11, pp. 311). Amsterdam: JAI/ Elsevier.
13
Axinn, C. N., & Matthyssens, P. (2001b). Limits of internationalization theories in an unlimited World. International Marketing Review, 19(5), 436449. Balabanis, G., Theodosiou, M., & Katsikea, E. S. (2004). Guest editorial. Export marketing: Developments and a research agenda. International Marketing Review, 21(4/5), 353377. Cavusgil, S. T. (1998). Perspectives. Knowledge development in international marketing. Journal of International Marketing, 6(2), 103112. Cavusgil, S. T., Deligonul, S., & Yaprak, A. (2005). International marketing as a eld of study: A critical assessment of earlier development and a look forward. Journal of International Marketing, 13(4), 127. Craig, C. S., & Douglas, S. P. (2001). Conducting international marketing research in the twenty-rst century. International Marketing Review, 18(1), 8090. Czinkota, M. R., & Ronkainen, I. A. (2003). An international marketing manifesto. Journal of International Marketing, 11(1), 1327. Douglas, S. P., & Craig, C. S. (2006). On improving the conceptual foundations of international marketing research. Journal of International Marketing, 14(1), 122. DuBois, F. L., & Reeb, D. (2000). Ranking the international business journals. Journal of International Business Studies, 31(4), 689704. Etemad, H., & Right, R. W. (2003). Introduction. Internationalization of SMEs: Toward a new paradigm. Small Business Economics, 20, 14. Katsikeas, C. S. (2003a). Advances in international marketing theory and practice. International Business Review, 12(2), 135140. Katsikeas, C. S. (2003b). Reections on Czinkota and Ronkainens international marketing manifesto: A perspective from Europe. Journal of International Marketing, 11(1), 2834. Knight, G. A., & Cavusgil, S. T. (1996). The born global rm: A challenge to traditional internationalization theory. In: S. T. Cavusgil & T. K. Madsen (Eds), Export internationalization research enrichment and challenges. Advances in international marketing, (Vol. 8, pp. 1126). New York: JAI Press. Madsen, T. K., & Servais, P. (1997). The internationalization of born globals: An evolutionary process? International Business Review, 6(6), 561583. Malhotra, N. K., Wu, L., & Whitelock, J. (2005). An overview of the rst 21 years of research in the international marketing review, 19832003. International Marketing Review, 22(4), 391398. Rialp, A., & Rialp, J. (2001). Conceptual frameworks on SMEs internationalization: Past, present, and future trends of research. In: C. N. Axinn & P. Matthyssens (Eds), Reassessing the internationalization of the rm. Advances in international marketing, (Vol. 11, pp. 4978). Amsterdam: JAI/Elsevier. Rialp, A., Rialp, J., & Knight, G. A. (2005). The phenomenon of early internationalizing rms. What do we know after a decade (19932003) of scientic inquiry? International Business Review, 14(2), 147166. Rialp, A., Rialp, J., Urbano, D., & Vaillant, Y. (2005). The born-global phenomenon: A comparative case study research. Journal of International Entrepreneurship, 3(2), 133171. Sinkovics, R. R., & Bell, J. D. (2006). Current perspectives on international entrepreneurship and the internet. Journal of International Entrepreneurship, 3(4), 247249. Styles, C., & Seymour, R. G. (2006). Opportunities for marketing researchers in international entrepreneurship. International Marketing Review, 23(2), 126145.
DIFFERENT TYPES OF EXPORTING SMEs: SIMILARITIES AND DIFFERENCES IN EXPORT PERFORMANCE

Jorma Larimo
ABSTRACT
Research related to rm export performance dates back to the early 1960s, ever since many studies have been conducted with mixed results. The three main goals of the present study were to analyze (1) the impact of the selected rm, management, and the export strategy-related variables on the export performance; (2) the possible variation in the results depending on the measure of export performance; and (3) the similarities and differences in the results depending on the type of SME traditional exporters vs. born international companies. Based on a literature review, 14 hypotheses were developed to be tested. Consequently, the empirical part of the study is based on a survey conducted among Finnish SMEs in early 2002. The export performance was analyzed using six different types of performance measures. None of the 14 hypotheses were fully supported by all employed measures of performance. However, the export performance was positively impacted by rm size, product/service quality, international orientation, and market diversication along ve measures.
17
18
JORMA LARIMO
Additionally, the study indicated some similarities, but also some differences depending on the measure of export performance, type of the exporting SME, and the operationalizations used for the born international companies. Based on the results, management implications and proposals for future research are presented.
1. INTRODUCTION
The intensication of the global-scale competition has led an increasing number of rms to look for opportunities in the international markets in order to achieve their objectives as well as to safeguard their market positions and survival. The most common mode of foreign operation of the small- and medium-sized companies (SMEs) has been export. Compared to other types of foreign operations, the export usually requires less nancial, human, and other resources, demands less investments, involves less nancial risks, and allows for greater structural and strategic exibility in the foreign markets than most other forms of operation (see e.g., Luostarinen & Welch, 1990; Young, Hamill, Wheeler, & Davies, 1989). To be successful in the foreign markets is, however, not so easy because of e.g., cultural differences, erce competition, and increasing dynamics in the market. The results concerning the links between various rm, management, and export strategy-related variables and export performance are of signicant importance from the point of view of both business managers and public policy makers. Furthermore, the results are also of great interest to marketing researchers for the development of theory building in international marketing. Research into rms export performance dates back to the early 1960s, with the pioneering study of Tookey (1964) as the rst one attempting to analyze the factors associated with successful exporting. Subsequently, numerous studies have been conducted trying to analyze the determinants of export performance. The empirical results related to variables of superior export performance have been relatively mixed. The differences are apparently caused at least partly by the differences in the measures of performance, samples, time periods, and operationalizations used for various rm, management, and export strategy-related variables. However, more research related to export performance needs to be undertaken. There are three main goals with the present paper. To analyze: (1) the impact of the selected rm, management, and the export strategyrelated variables to the export performance;
Different Types of Exporting SMEs
19
(2) the possible variation in the results depending on the measure of export performance; and (3) the similarities and differences in the results depending on the type of SME traditional exporters vs. born international companies. The contribution of the paper to the present stock of knowledge related to SME export performance is mainly in the following issues. The impact on export performance of some of the variables at work in the present study has been only limitedly assessed in previous analyses. Furthermore, there has been mixed results in earlier studies. For the measurement of performance, six different variables representing both objective and subjective measures have been employed in this study. Although there is an increasing number of studies including analysis of results based on several performance measures, our analysis attempts to provide a comprehensive view of the export performance of the reviewed companies. Furthermore, the previous studies have mainly analyzed the differences in SME export performance based on the size of the reviewed companies, the speed of internationalization has been of greater interest in only very few studies so far. Related to the speed of internationalization, the classication is usually made by classifying rms into the traditional exporters or born global companies but the conditions regarding e.g., to the time frame of starting exports/foreign operations, the level of exports within that time frame and later on have been very different. Two different bases are used to classify the companies into traditional exporters and born globals in order to analyze the impact of both looser and tighter denitions for born global companies on the results. Finally, although there have been numerous studies focusing on various aspects of export and internationalization behavior in the Nordic countries, there are very few larger-scale empirical studies focusing on the SME export performance. This concerns especially the analysis of export performance using several different performance measures. The structure of the paper is as follows: In Section 2, a review of the types of exporting SMEs focusing on traditional exporters and the so-called born global rms is presented. In Section 3, rst a review of the measures of export performance is made followed by an analysis of the relation between the rm, management, and export strategy-related variables and the export performance, and nally, based on the reviews the hypotheses are developed for the empirical part of the study. Section 4 in the paper includes the methodology, operationalizations of variables, and sample description. Section 5 presents the main results of the study, and Section 6 summarizes the main ndings and conclusions, and suggestions based on the study.
20
JORMA LARIMO
2. TYPES OF EXPORTING SMEs: TRADITIONAL EXPORTERS VS. BORN INTERNATIONALS

Until mid-1990s, the reviewed companies in export marketing and performance studies were often divided into exporting and non-exporting, or to highly vs. limitedly exporting companies. The past 10 years witnessed an increasing interest toward rms which do not try rst to grow and get strong market positions in their home countries and after that start foreign operations but rather are involved in foreign operations in their rst years of operations or even straightaway start their foreign sales. These rms have been labeled quite differently in various studies: born globals (Knight & Cavusgil, 1996; Madsen & Servais, 1997; Aspelund & Moen, 2001), international new ventures (INVs) (Oviatt & McDougall, 1994, 1995, 1997), global start-ups (Oviatt & McDougall, 1994), or high-technology start-ups (Jolly, Alahuhta, & Jeanet, 1992). More recently, the term born-again global rm has also been proposed, to describe long-established rms that used to focus on their domestic markets, but suddenly opt for rapid internationalization (see e.g., Bell, McNaughton, & Young, 2001). The main driving forces for the existence of these new types of companies are (see Knight & Cavusgil, 1996; Madsen & Servais, 1997; Rialp, Rialp, & Knight, 2005): (1) the increasing role of niche markets, (2) the advances in process technology, (3) the advances in communication technology, (4) the inherent advances of small companies, (5) the means of internationalization that have become more accessible to all rms and increasing support activities for the greater international contacts and cooperation, and (6) the increasing amounts and use of international networks. Looking at the operational denitions used in various studies focusing on the born globals, several reviews indicate clearly that there has been a lot of variation. Based on an analysis of 55 studies dealing with born globals, Dominguinhos and Simoes (2004) found that two criteria were more commonly used in those denitions: (1) the time frame from establishment to start of exports/foreign operations and (2) the share of exports within that time frame. Less used criteria were the timing of founding of the company, the number of export target countries, and even more limitedly the type of target countries. Concerning the time frame, some relatively few studies used a time frame of one to two years from the establishment to the start of export. Some studies have used a time frame of three years, but in several studies also a longer time frame of six years has been used, apparently inuenced signicantly by the pioneering studies by Oviatt and McDougall (1994, 1995, 1997). In some studies only the reference in the rst years is
21
at work. Concerning the role of exports, some studies do not impose any specic share, in some studies 5, 15, or 25%, respectively has been used as the minimum share for exports. Regarding foundation, some studies have focused on companies founded in 1976 or more recently, or in some studies in 1989 or more recently, but several studies do not include any limitations, because there is evidence that there has been born globals also before mid1970s. In some studies, the demand of at least three target countries of exports has been used; in some studies, even more demanding condition like exports/operations in two continents, but only after a rather long time since establishment (Gabrielsson, Sasi, & Darling, 2004, use a period of 15 years). Furthermore, a key feature of the operation should be that the foreign sales should be continuous and signicant for the company (mainly increasing role). (For the reviews, see Dominguinhos & Simoes, 2004; Rialp et al., 2005). As the earlier discussion indicates, there has been a great variation in the denitions used for born globals. The differences may be expected to have an impact also on the results related to the behavior, strategies, and export performance. Based on the denitions and operationalizations presented above, two different classications were agreed upon for the present study: (1) a looser classication and (2) a tighter classication. In the looser classication, the companies are divided into two sub-groups only based on the time frame and the existence of exports: no exports vs. exports within three years from the establishment, without any additional conditions. The subgroups are labeled as truly traditional exporters and born internationals. In the second classication, three conditions were taken into consideration: (1) the time frame and the existence of exports: no exports vs. exports within three years from the establishment, (2) the share of exports within three years: whether the share of exports had reached the 25% limit of the total sales within the agreed time frame, and (3) the continuous and important/ extensive role of exports: if the share of foreign sales had reached the 50% limit of the total sales. The sub-groups are labeled as traditional exporters and truly born internationals. The cutoff rate of 25% was based on the earlier studies and on the fact that if the foreign sales reach that level, it means that the company has to take the international operations seriously and that internationalization is not sporadic anymore. Furthermore, the requirement of at least 50% share of foreign sales was set to indicate the continuous and increasing/signicant role of foreign sales. Without this type of additional conditions, e.g., a rm which has started its foreign sales within three years from the establishment (share from total sales may even exceed 25%), possibly based on an unsolicited inquiry, but which
22
JORMA LARIMO
stops mainly or totally the exports after those three years, would be classied as a born global. The term born global indicates operations in several different countries which are located on various continents. Because these conditions were not set as it has also been the case in most earlier studies although the term is used it was decided that the term born international will be employed in this study to dene better the nature of these SMEs. As discussed above, there is clear evidence that there have been born international companies also before 1990s. In fact, e.g., in the study by Aspelund and Moen (2005) the oldest born international company in their sample was established already in 1874. However, their number has clearly increased in the 1990s (see Moen, 2002; Moen & Servais, 2002). In a study by Moen (2002), the majority of both French and Norwegian exporting rms established in the 1990s could be classied as born internationals. The results from Denmark indicate a similar situation (see Knight, Madsen, & Servais, 2004). Additionally there are born international companies outside hi-tech sectors, for instance, in the consumer goods and the service sectors (see e.g., Madsen & Servais, 1997). This seems to be the case especially in smaller developed countries with small domestic markets. Therefore, it was decided not to make any limitations for born international companies based on the time frame of establishment, or eld of industry. Several features are considered to dene the born international rms (see e.g., Knight & Cavusgil, 1996; Madsen & Servais, 1997; Moen & Servais, 2002; Rialp et al., 2005): (a) they focus on highly specialized global market niches, (b) they expand their export/foreign operations rapidly into several markets, (c) they enter their lead markets more quickly than the traditional rms, (d) they may proceed very rapidly along the steps of internationalization, (e) they may use leap-frogging in their operation strategy development, and (f) they use more networking and alliances than the traditional rms. Several authors argue that the existence of this kind of companies is focused on hi-tech sectors, but there is evidence that these type of companies are not limited only to hi-tech sectors (see e.g., Madsen & Servais, 1997). According to a recent study, the share of foreign sales in Danish born internationals was on average 71%, whereas the respective share in US-based born internationals was 47% (Knight et al., 2004). (For a recent study modeling the forces inuencing the speed of internationalization see Oviatt & McDougall, 2005.) The expected differences e.g., in export strategy between traditional exporters and born internationals and relation with export performance is discussed more in Section 3.2 of the study.
23
3. LITERATURE REVIEW OF EXPORT PERFORMANCE STUDIES AND DEVELOPMENT OF HYPOTHESES

3.1. The Measurement of Export Performance The concept of export performance is rather complicated and multidimensional. Several different measures can be used and have been utilized to measure export performance. Katsikeas, Leonidou, and Morgan (2000) identied in their extensive analysis of previous export performance studies over 100 articles of which they included 93 into their more detailed investigation of key issues related to export performance measurement. In these 93 studies, 42 different performance measures had been used. In a more recent review made by Sousa (2004), 43 export performance studies made in 19982004 were covered. The results indicated that 50 different measures for export performance had been applied in those studies. Most of the measures used were in both studies economic (23/27 measures), but there were also several non-economic measures (14/17), and some generic measures (5/6). Several measures were used in only one study (22/20 measures), but an increasing amount of measures had been employed at least in ve studies (11/13), some even in more than 10 studies (5/6). The review by Leonidou et al. (2000) indicated that in about one-third (in 33 cases) of the studies only one single measure of performance was used, whereas in almost two-thirds (60 cases) two or more measures of performance were utilized. However, the review by Sousa (2004) indicated that in only 3 of the 43 studies one single measure was considered and that in half of the studies four or more performance measures were in use. Thus the number of measures has clearly increased in more recent studies. The most frequent single measures were export sales ratio (in 57/16 studies), export sales growth (41/16 studies), export sales volume (20/17), export protability (20/20), and growth of the export sales ratio (12/9). Thus, all the most commonly used measures were economic measures and four of the ve were sales-related measures. The next most commonly utilized measures were export protability growth, overall export performance, perceived export success, export market share, export market share growth, achievement of export objectives, number of export countries/markets, and strategic export performance. The above results indicate that the export sales ratio has been used earlier in well over half and more recently in about half of the studies as the measure of export performance. The use of this measure at least as the
24
JORMA LARIMO
single measure of export performance has been criticized (see e.g., Kirpalani & Balcome, 1987) because it can be affected by several other issues than competitiveness/successful export operations. Also the other often-utilized sales-related export performance measure export sales growth has received criticism, because the measure may overstate performance based on price escalation and market growth, or understate performance because of the experience curve effects and deteriorating demand (see ibid and Katsikeas et al., 2000). Protability in exports and growth in export protability has been cited as the ultimate goals of the companies (Aaby & Slater, 1989). However, the companies seem to be less willing to reveal export protability information partly because of the fact that the companies do not necessarily know exactly their export-related prots, especially if the rms utilize marginal cost pricing (Samiee & Anckar 1998; Katsikeas et al., 2000). Among the non-economic measures, for instance, the number of export countries has been commonly employed as a measure of export performance. However, as Piercy (1982) has stated, the number of foreign markets is not an end in itself but is contingent on the specic company, product, market, and marketing factors. Thus, instead of using the number of export markets as the measure of performance it would be better to use this variable in the analysis of various export strategy decisions on export performance. As stated above, the generic measures were taken into consideration rather seldom, but increasingly in the export performance studies. From the measures in this category, perceived export success and achievement of export objectives were among the most commonly used measures. Their advantage is that the evaluation is based on the goals and measures of performance utilized by the company itself, although they can be regarded only as crude measures of performance since they cannot adequately capture the domain of the construct (Katsikeas et al., 2000). The results by Katsikeas et al. (2000) indicated that the analysis of export performance is mainly based on current export performance (in 82 studies); secondarily on historical performance (56), or on a combination of both (45 studies), and only in very few cases on anticipated future performance (the review by Sousa does not include this piece of information). The unit of analysis was usually at the corporate level (in 84/28 studies), only seldom on export venture (12/15) and rarely product/product line level (4/0). The scope of the analysis has usually covered all export markets by the rm (83/28 studies), secondarily one single country (17/15), and in very rare cases some region (1/0 study). Almost all studies have been based on primary data
25
(96/43) usually collected via mail questionnaire based on the views of a single key informant, mainly the manager directly responsible for exports (commonly the CEO or the export manager). The earlier evaluations were more often based on objective (80 studies) than on subjective (51) assessment, but the situation has changed to the opposite recently (54 vs. 151). Also a combination of both types of measures seems to have increased clearly in more recent studies (in the earlier review in ca. 20% of the studies, whereas about one-third in the latter review). Noteworthy is, however, that no study examined the relationship between objective and subjective export performance in more detail. The results by Katsikeas et al. (2000), Sousa (2004), and Diamantopoulos (1999), clearly indicate that the export performance is a complex phenomenon. Based on the reviews it can be concluded that: (1) any single measure of export performance is not enough, thus there should be at least two or more measures for export performance; (2) there should be both objective and subjective measures; (3) there should be both economic and noneconomic or/and generic measures; (4) the measurement should usually cover both all export markets and main product or/and main target marketrelated measures; and (5) variation in the time dimension should be considered current and historical.
3.2. The Relationships between the Firm, Management, and Export Strategy and the Export Performance A review of export performance studies indicates that special attention was given to the impact of over 50 different rm characteristics and competencies, management characteristics, management attitudes and perceptions, industry characteristics, domestic market characteristics, target markets and their characteristics, and export marketing strategy-related variables on export performance. Furthermore, several different measures of export performance have been used in those studies. Of interest in several studies has been the impact of the rm size, the key sales object issues (product/service strengths), the export marketing strategy, including market concentration vs. market spreading, the standardization of the marketing mix elements, and the international experience. In this study, the analysis will focus on the relationships between 11 variables and the export performance. The selected variables represent the rm characteristics, including basic rm characteristics and strategy (1, rm size; 2, product/service quality; 3, niche product/ service; and 4, export age), the management characteristics (5, international
26
JORMA LARIMO
orientation; 6, international commitment; and 7, international experience), and the export marketing strategy (8, product adaptation; 9, communication adaptation; 10, price adaptation; and 11, market diversication). Table 1 includes a summary of the results presented by ve review articles of various export performance studies. The reviews by Madsen (1987), Aaby and Slater (1989), and Gemunden (1991) included 17, 55, and 50 empirical export performance studies made between mid-1960s and late 1980s. Zou and Stan (1998) had in their review 50 export performance-related studies made between 1987 and 1997, and Manolova and Manev (2004) analyzed 21 articles published between 1996 and 2001. In all the aforementioned works, three eligibility criteria were at work for the selected studies: (1) the articles had to be empirical in nature, reporting data analysis and statistical tests; (2) the articles had to use some kind of export performance measure as a dependent variable; and (3) the articles had to be cross-sectional in nature (i.e., the case studies were excluded). As shown in Table 1 the results related to the impact of most variables on export performance have been very mixed. They will be discussed in more detail below, in connection to the single variables selected for this study. 3.2.1. Firm Characteristics and Basic Strategy-Related Variables Firm size. There are three fundamental factors supporting the expectation of a positive relationship between rm size and export performance: the organizational resources, the economies of scale, and the perception of risk in the international activity (see Bonaccorsi, 1992; Katsikeas, Piercy, & Ioannidis, 1996). Large exporting manufacturers are widely considered to possess more nancial and human resources, enjoy higher levels of scale economies, and perceive lower levels of risks about foreign markets. Although there are counter arguments, these size-related advantages seem not only to facilitate understanding of foreign market characteristics, but also to enhance a rms ability to respond effectively to the requirements of foreign customers thus potentially leading to better export performance. As the results in Table 1 indicate, the impact of the rms size has been the most frequently reviewed variable in export performance studies. Several studies seem to indicate no relation between rm size and export performance (Katsikeas et al., 1996). However, other studies seem to indicate a positive relation (e.g., Chetty & Hamilton, 1993; Chadee & Mattsson, 1998; Nakos, Brouthers, & Brouthers, 1998; Piercy et al., 1998; Preece, Miles, & Baetz, 1998; Baldauf, Cravens, & Wagner, 2000), and some studies conrm that the positive relation was found among both manufacturing and service
Table 1.
The Relationships between Reviewed Variables and Export Performance in Various Review Studies.
Madsen (1987) (N 17) Aaby and Slater (1989) Gemunden (1991) Zou and Stan (1998) (N 50) Manolova and Manev (2004) (N 25) POS. NEG. NS IMP. IMP. TOTAL
POS NEG. NS IMP. IMP
POS. NEG. NS IMP. IMP.
Firm size/rm resources Product/service quality/strength Niche product/service Export agea/rm age International orientation International commitment International experience Product adaptation Price adaptation Communication adaptation Market diversication
4 7 4 3
2 0 1 1
5 2 1 2
5 1 3 5 7 3 1 1 3
7 0 0 0 0 0 1 0
3 1 0 0 0 0 4 0
7 4 3 4 2 0
1 0 0 0 0 0
10 2 2 4 2 2
9 13 6 10 15 15 12 7 3
b
5 2 2 0 0 1 2 1 3
b
23 27 3 6 2 10 13 6 2
b
5 0 5 8 5 1 5
3 3 0 0 0 0 0
8 3 0 0 1 0 0
30 25 3 6 23 30 23 22 10 3 11
18 2 0 5 0 0 1 4 1 3 1
49 32 0 6 8 2 11 22 8 4 2
Note: In one study more than one export performance measures may have been used. POS. IMP., POSITIVE IMPACT; NEG IMP., NEGATIVE IMPACT; NS, NON-SIGNIFICANT. a Not included in any reviews. b Not specied in the review, included into the rm general export strategy results.
27
28
JORMA LARIMO
companies (e.g., Chadee & Mattsson, 1998; White, Grifth & Ryans, 1998). It seems that the positive relation has been identied more often when the size of the rm was measured based on the number of employees, and the performance measured using export ratio to total sales, than when other measures for the rm size and/or export performance has been used although e.g., Nakos et al. (1998) found a positive relation between rm size and export performance based both on export ratio and protability of exports. Recent support for the positive relation between rm size and export performance was given by Majocchi, Bacchiocchi, and Mayrhofer (2005) based on Italian SMEs, and by Voerman (2004) based on an European multicountry SME sample. Thus, for the empirical part of the study we expect that: Hypothesis 1. There is a positive relation between rm size and export performance. Product/service quality. Quality can be dened in several ways. However, often quality refers to how well the perceived fundamental characteristic of a product or service meet the expectations of the customers (Mohr-Jackson, 1998). A quality emphasis can be expensive and have a negative impact on product standardization. On the other hand, research has also shown that higher quality does not necessarily incur substantial extra costs. More often quality has been linked to improved competitiveness and to superior organizational performance (see Buzzell & Gale, 1987; Mohr-Jackson, 1998; Porter, 1990). In an increasingly global economy, buyers can access a greater volume and variety of product choices. As buyers have more alternatives to choose from, their expectations of product/service quality may grow, leading to pressure on companies to improve their product/service quality (Knight, 1997). High quality offers also an opportunity to differentiate the product/service from those of competitors (Calantone & Knight, 2000). Concerning this aspect, the empirical results are mixed (see Table 1). However, the non-signicant relation with export performance has usually been found, when the relation between product strength and export performance has been analyzed, whereas a positive relation has been identied mostly, when exactly the relation between product quality and export performance has been reviewed (see e.g., Thirkell & Dau, 1998; Knight et al., 2004). Both results by Calantone and Knight (2000) and Knight et al. (2004) indicated a very strong support for the positive impact of quality on export performance in US-based industrial companies. Furthermore, a recent study focused on British and Spanish exporters clearly indicated that product and/ or service quality was the key antecedent of export performance in both
29
samples (see Lages, Lages, & Lages, 2006). Thus, for the empirical part of the study we expect that: Hypothesis 2. There is a positive relation between product/service quality and export performance. Niche product/service. With heightened competition in many industries, there is a growing pressure on SMEs to concentrate on niche markets in order to serve market segments that are small enough to avoid the competition of large rivals. The niche may be even very small in one country, but via expanding the operations to several countries the company may reach economies of scale (see e.g., Knight, 1997). The technical development has facilitated greater specialization in SMEs and therefore increased the possibility to concentrate on niche markets. As the results in Table 1 indicate, the impact of niche product/service or product/service uniqueness has been analyzed only to a limited degree so far. However, the impact appears to be mainly positive. Table 2 includes the results of the marketing strategy export performance summary review by Leonidou, Katsikeas, and Samiee (2002). The results offer additional information about the relation between various variables and export performance adding the type of export performance measures used to the analysis. Their review was based on the analysis of 36 export performance studies, covering the period from early 1970s to the end of 1990s. The results by Leonidou et al. (2002) clearly indicate that export product uniqueness has in general had a signicant positive impact on performance, particularly in the case of export proportion of sales. The results in a recent study (Knight et al., 2004) indicated the positive impact of product differentiation on export performance in the US sub-sample, whereas in the Danish subsample a non-signicant relation was found. Because of the resource limitations, the operations of SMEs have to be more concentrated than those of their larger competitors in order to be competitive in the foreign markets. Based on the above, for the empirical part of the study we expect a positive relation between niche focus and export performance. Hypothesis 3. There is a positive relation between niche product/service focus and export performance. Export age. The argumentation for an assumption of a positive relationship between export age (exporting experience) and export performance lies in the issue of uncertainty and the way various rms cope with it (Erramilli, 1991). Less experienced exporters are likely to perceive considerable uncertainty, which in turn might adversely affect their perceptions of
30
Table 2.
Export Strategy Area
Summary of Marketing Strategy Export Performance Relationship (p-Values) in Various Empirical Studies.
Export Performance Indicator Time of Study 1980s Geographic Focus Product Type Both
Overall Export Export Export Prot Other Composite 1970s Export Sales Prots Contribution Performance Performance Performance Intensity Level Measure Measure Targeting Market concentration Market spreading Product Quality New/unique product(niche) Product adaptation Price Price adaptation Promotion Promotion adaptation
1990s America Europe Other Industrial
0.000 0.000
0.000 0.000
0.057 0.063
0.100 0.100
0.100 0.100
0.019
0.000 0.000
0.004 0.004
0.000 0.003
0.010
0.000 0.002
0.000 0.003
0.000 0.001
0.000 0.000 0.000
0.049 0.002 0.000
0.049 0.013 0.122
0.049 0.038 0.100 0.325
0.059
0.049
0.001 0.000 0.000
0.100 0.019 0.000
0.099 0.007 0.000
0.001 0.000 0.001
0.100
0.099 0.011 0.000
0.001 0.000 0.084
0.007
0.009
0.001
0.000
0.001
0.031
0.004
0.004
0.000
0.004
0.344
0.004
0.004
0.000
0.002
0.361
0.197
0.000
0.003
0.001
0.004
0.000
0.003
JORMA LARIMO
Source: Leonidou et al. (2002).
31
potential risks and returns about foreign sales. Higher export age/export experience is likely to determine the rms to be less uncertain, related to foreign sales based on increased market and customer knowledge and networks, leading to more effective export sales planning and strategies (Madsen, 1989; Katsikeas et al., 1996). There should also be a learning curve or experience effect that reduces the foreign operating and coordination costs. In this case as well the empirical results are very mixed (see Table 1). Some studies do not indicate any relation between international/export experience and performance (see Nakos et al., 1998) and some studies indicate even a negative relation (e.g., Kaynak & Kuan, 1993). Nevertheless, support for a positive relation has been found clearly more often than for a negative relation. This has been the case whether the experience has been operationalized using the age of the company in general (Contractor, Hsu, & Kundhu, 2005; Kundhu & Renko, 2005; Majocchi et al., 2005), the number of export ventures in which the rm has been involved, or the time frame of rms export involvement (Diamantopoulos & Inglis, 1988; Kaynak & Kuan, 1993; Cavusgil & Zou, 1994; Piercy et al., 1998), or export market knowledge (e.g., Leonidou & Kaleka, 1998). This has also been the case when using various measures for export performance, like export sales (e.g., Francis & Collins-Dodd, 2000), or export ratio/intensity (Baldauf et al., 2000; Francis & Collins-Dodd, 2000; Contractor et al., 2005; Kundhu & Renko, 2005), export growth (Contractor et al., 2005; Kundhu & Renko, 2005 in the Indian sample), or using a composite measure (Thirkell & Dau, 1998; Leonidou & Kaleka, 1998). Finally, in some studies the impact has been dependent on a sub-sample like in the study by Kundhu and Renko (2005), where a negative relation was found in the Finnish software exporter sub-sample when the performance was measured using export growth, whereas a positive relation was identied in the Indian sub-sample (and also in the Finnish sub-sample when performance was measured employing export intensity as the measure). In summary, the empirical evidence is mixed, but it slightly tends to support a positive relation between export experience and performance. Consequently, we expect that: Hypothesis 4. There is a positive relation between the length of export experience of the company and the export performance. 3.2.2. Management Characteristics Related Variables International orientation. An internationally oriented rm can better identify and benet from emerging international opportunities and avoid threats
32
JORMA LARIMO
(Zou & Stan, 1998). In highly internationally oriented rms the advantages of exports outperform the barriers of exporting, whereas in the highly domestically oriented rms the situation is expected to be the opposite. Therefore a positive relation between the international orientation and the export performance could be expected. The results in Table 1 indicate that there is a wide empirical support for the positive impact of international orientation on export performance. Among the studies indicating support for the positive relation can be mentioned those by Preece et al. (1998) and Calantone and Knight (2000). More recently the results in the studies by Contractor et al. (2005) and Kundhu and Renko (2005) indicated no relation between international orientation and export performance in the Indian software sub-sample, but results in the Taiwanese and Finnish sub-samples gave support for the expected positive relation. Thus, the latter results support the earlier ndings by White et al. (1998) based on the US origin business-to-business service rms. The empirical evidence also indicates that there has been a positive relation when both nancial measures and composite measures for export performance have been used (see Zou & Stan, 1998). Based on the above, for the empirical part of the study we expect that: Hypothesis 5. There is a positive relation between the international orientation of the management and the export performance. International commitment. The high commitment of the management toward exports allows a rm to aggressively go after the export market opportunities, and pursue effective export marketing strategies that improve export performance (Cavusgil & Zou, 1994). The market building in exports is usually long-term oriented requiring high commitment to these markets. As such, a positive relation between international commitment and export performance could be expected. The empirical support for the positive impact of international commitment on export performance is very wide (see Table 1). Support for the positive relation has been found both in manufacturing and service rms (although greater among the latter ones) (Chetty & Hamilton, 1993; Chadee & Mattsson, 1998) and using both export ratio and export protability as the measure of performance (e.g., Nakos et al., 1998). Therefore for the empirical part of the study we expect that: Hypothesis 6. There is a positive relation between the international commitment of the management and the export performance.
33
International experience. The international experience of the manager(s) helps a rm to identify and leverage on the international opportunities, while avoiding or at least having a more realistic view of the threats of exports to foreign countries. Thus, a positive relation between the international experience of the management and the export performance could be expected. The results by Das (1994) indicated that the managers of successful exporting rms had less experience in exporting and in foreign settings as compared to the managers of unsuccessful rms. One possible explanation for the unexpected nding may be that Das focused on rms operating in highly turbulent environments and the unique context explains the ndings. Recently also Contractor et al. (2005) and Kundhu and Renko (2005) found a negative relation between international experience and export performance, but only in their Indian sub-sample and only when performance was measured based on export growth. The results in the Finnish and Taiwanese sub-samples indicated no relation between international experience and export performance. The authors explain their ndings that given the nature of the industry of the sample software industry the entrepreneurs do not have to possess signicant international experience, as they can reach out to foreign buyers in the target countries through a combination of internet and formal/informal networks. However, in general, international experience of the management seems to have had either a non-signicant or in a majority of cases a positive impact on export performance whether the performance was measured based on export sales, export prots, export growth, or based on a composite measure (see results in Table 1 and Zou and Stan, 1998, p. 349). Therefore we expect that: Hypothesis 7. There is a positive relation between the international experience of the management and the export performance. 3.2.3. Export Strategy-Related Variables Product/service adaptation. The proponents of a positive relation between standardized product/service and export performance refer to the cost savings and to the similarities in buyer behavior in several markets (Keegan, 1995). The assumption of a positive relation between product/service adaptation and export performance is based on three key benets related to adaptation (see Leonidou et al., 2002). First, adaptation reects customerorientation based on a more or less systematic analysis of buyer behavior and target market characteristics (Douglas & Wind, 1987). Second, the better productmarket match can result in greater buyer satisfaction, which
34
JORMA LARIMO
can lead to greater sales volumes or to greater pricing freedom in comparison to competitors, which translates into better export performance. Third, pressures to meet specic host market requirements often demand creative and innovative thinking, which may result in additional products both for the domestic and foreign markets (Czinkota & Ronkainen, 1998). Some empirical studies indicate no relation at all (Diamantopoulos & Inglis, 1988), or even a negative relation (e.g., Fenwick & Amine, 1979) between product adaptation and export performance. However, the majority of studies seem to indicate support for a positive relation (e.g., Cavusgil & Zou, 1994; Nakos et al., 1998; Shoham, 1996, 1999, 2002; Baldauf et al., 2000). Also the results in the review by Leonidou et al. (2002) clearly indicated that product adaptation was positively correlated with superior export performance. The results seemed to be valid across all time frames and geographic contexts, but only in studies using export sales-based performance measures. Additionally, in a recent study (Calantone, Kim, Schmidt, & Cavusgil, 2006), the results indicate a positive relation between product adaptation and export performance in all three reviewed samples: US, Japanese, and South Korean. In an examination of exporters along the internationalization process, Cavusgil and Kirpalani (1993) found that high product adaptation was not so important in enhancing export performance at entry as it was in subsequent penetration. Based on the ndings in several earlier studies and on the expectation that the benets of customer-orientation and better product/service match outweigh the benets of costs savings in the standardized strategy, we expect that: Hypothesis 8. There is a positive relation between product adaptation and export performance. Price adaptation. The adaptation of pricing in export marketing is grounded on several reasons: economic, politicallegal, price control, differences in distribution and transportation arrangements and costs, market structures, tariffs, taxes, and other barriers, etc. This diversity of foreign market pricing determinants makes price adaptation more or less necessary for the rms to survive and remain competitive in export markets (see Leonidou et al., 2002). Whereas Shoham (1995) reported that price adaptation enhanced protability, he later reported rst a negative impact (1996), and later on a nonsignicant impact (Shoham, 2002). Similarly, Koh and Robicheaux (1988) reported inconclusive ndings; price adaptation enhanced performance, but only when it was higher than domestic prices. However, the results of the
35
review by Leonidou et al. (2002) indicated support for a strong positive link between price adaptation and both overall and individual export performance measures with the exception of export sales volume (see Table 2). Thus, the diversity of foreign market pricing determinants seems to make price adaptation necessary for rms to survive and remain competitive in the host markets. Therefore, we expect that: Hypothesis 9. There is a positive relation between price adaptation and export performance. Communication adaptation. Proponents of the positive relationship between standardized communication refer to the similarities in buyers consumption patterns, and to the existence of international market segments as in the case of product standardization and to the cost savings based on the use of standardized communication strategy. In contrast, those referring to the positive relationship between adapted communication and export performance cite the differences in government restrictions, competitive practices, communication infrastructures, etc. (see Keegan, 1995; Leonidou et al., 2002). Cavusgil and Zou (1994) discovered a negative relation between promotion adaptation and export performance. They found that the relation between those variables is apparently more complex. The identied negative relation may be caused by the universal appeal of some products, poor judgment in altering of the positioning, or promotion mix. The results in various studies by Shoham have also been mixed. Whereas in one study (Shoham, 2002) no relation between adaptation of advertising and export performance was found, the results in his two other studies (Shoham, 1996, 1999) indicated support for a positive impact of promotion adaptation on export performance along various parameters: export sales, export prots, or growth herein. The results in the review by Leonidou et al. (2002) indicated, however, that adaptation of advertising/promotion had a strong positive impact on overall export performance, irrespective of the time, place, and products focused on in the studies reviewed (see Table 2). Noteworthy is, that his more detailed analysis uncovered a strong positive effect only on export sales growth and export intensity, while its impact on export prot contribution was limited. Although the empirical results have been mixed it is expected that the benets related to adaptation of communication outweigh the benets of standardization and therefore we expect that: Hypothesis 10. Here is a positive relation between communication adaptation and export performance.
36
JORMA LARIMO
Market diversication. One key question in exports is to decide whether to concentrate, or to diversify the sales in several target countries. Those favoring a positive relation between concentration and export performance, refer to the better possibilities for bigger market shares based on the more targeted and possibly bigger marketing efforts. In contrast, those proposing a positive relation between market spreading and export performance refer to risk spreading and market coverage (see Dean et al., 2000). As the results in Table 2 indicate, there has been support for both the positive impact of market concentration and for market diversication on export performance. Support for the positive relation between market diversication and export performance is indicated in Diamantopoulos and Inglis (1988), and Cavusgil and Zou (1994). More recently, Contractor et al. (2005) and Kundhu and Renko (2005) found no relation, in the Taiwanese sub-sample even a negative relation based on export intensity, but support for a positive relation in the Finnish sub-sample. The unexpected negative result was explained by the fact that the focus in those studies was on software industry where to concentrate solely on the US market seemed to be more appealing to some Taiwanese software exporters, rather than to scatter the sales across several target countries. The results in Table 1 indicate, however, that there has been much more support for the positive impact of market diversication on export performance than the opposite. Based on these ndings and considering factors like the risk spreading advantages, the developments in information technology, and the market turbulence we expect that: Hypothesis 11. There is a positive relation between market diversication and export performance. Adaptation of the marketing mix elements to the target country conditions could be expected to increase competitiveness of the company in those markets and therefore increase sales, as the results in several earlier studies indicate (see above). However, the different types of adaptation product/ service, price, and/or communication adaptation usually increase the costs, when compared to standardization. Therefore the impact on export performance could be expected to be more dependent of the measure on export performance in the case of adaptations than related to the other variables of interest in this study. Hence, we expect that: Hypothesis 12. The positive relation between adaptation of marketing mix variables and export performance is more signicant when the export performance is measured using subjective export performance measures
37
and export sales intensity, than when the performance is measured using protability-based export performance measure. In Section 2 of the paper, traditional exporters vs. born international were discussed. In a study by Aspelund and Moen (2005) focusing on Norwegian born internationals vs. traditional type of exporters the results showed that based on perceived international performance, the born internationals had performed better than the other sub-groups, but taking into account the objective nancial measures (ROI and ROE) no signicant differences in export performance between sub-groups was found. Based on the discussion in Section 2 of the paper, the key features of these born international companies seem to be rapid expansion into several markets soon after the establishment of the company. Therefore, we may expect that key aspects of the successful performance in foreign markets are a niche focus, a large number of target markets, and a higher degree of standardization of the marketing mix elements like product, price, and communication than in traditional exporting rms, in order to be able to expand rapidly into several target countries (see Jolly et al., 1992). The rapid expansion also demands great international orientation and commitment. Therefore, for the empirical part of the study is expected that: Hypothesis 13. The positive relation between niche focus, international orientation, and international commitment with export performance is greater in born international companies than in traditional exporting companies. Hypothesis 14. The positive relation between product, price, and communication adaptation and export performance is lower in born international companies than in traditional exporting companies.
4. SAMPLE, METHODOLOGY, AND OPERATIONALIZATIONS OF THE MEASURES

The data for the study was collected as a part of a larger survey analyzing the export behavior, strategies, and performance of the Finnish SMEs. As discussed earlier the survey method has been the clearly dominating method of data collection in export performance studies. The questionnaire was ve pages long. The target group was manufacturing (NACE 15-36) and service rms (NACE 72.2 software, 74.2 engineering, and 74.3 advertising) having 10500 employees, and performing exports according to the Yritys Suomi
38
JORMA LARIMO
2000 database. On the basis of these sources, the total target group consisted of 2,856 companies. The survey took place between November 2001 and March 2002. In companies having less than 50 employees the questionnaire was directed to the managing directors, whereas in larger companies it was directed to the export/international business manager. In 48 cases the companies were in bankruptcy or the address was wrong and the new address could not be identied. Moreover, in 154 cases the companies were not performing any exports, had more than 500 employees or were operating in elds not included in this study. Excluding these companies the nal target group was totaled 2,654. In total, 489 answers were received (18.4%), from which 386 were usable in this study resulting in a response rate of 14.5%. One explanation for the rather low response rate was very evidently the length of the used questionnaire. The response rates in the review by Sousa (2004) in studies focusing on SME export performance varied from 9.8 to 51.8% and the median sample size was 181 and the mean 232 companies. Also the reviews by Zou and Stan (1998) and Manolova and Manev (2004) reveal that in about half of the studies analyzed the sample size was 100200 rms, and in very few cases more than 300 companies. In general, response rates in largescale surveys targeted to the top management ending to response rates between 14 and 20% seem to be rather common (see e.g., Menon, Bharadwaj, & Howell, 1996). As such, even though the response rate was not very high, the sample size in this study was clearly over the average in export performance studies made during the years. Based on the number of employees, annual turnover, and eld of industry, there were no signicant differences between responding and non-responding companies. Furthermore, no greater differences were found between early- and late-responding companies. The respondents were dominantly managing directors, export or nancial managers of the companies. Regarding the participating companies, 84.7% were manufacturing, mainly industrial goods producers (NACE 29 machinery and equipment, 27 basic metals and fabricated metal products, and 26 non-metallic mineral products as the main elds of industries), and 15.3% service companies. About half of the companies were family-owned companies, almost onefourth was a part of a larger group, and the remaining were independent limited liability companies. The mean year of establishment was 1974, and the mode value 1992. The rst year of export was on average 1985; and in 2001, the participating companies had an average of 64 employees. The mean annual turnover was in 2001 h8.67 million, and the mode was h1.7 million. Only extremely few companies had more than 250 employees. The
39
mean share of exports in 2001 was 39.2%, and on average the companies had exports to 9.4 foreign markets. The most common and most important target countries for exports were Sweden, Germany, and Russia. In almost two-thirds of the 25 empirical export performance made in 19962001 reviewed by Manolova and Manev (2004) regression analysis was employed as the statistical tool. Therefore also in this study cross-sectional OLS-regression analysis was decided to be employed in the data analysis. Appendix A includes the operationalizations for the dependent variable six different types of measures for export performance used in the study and for the 11 independent variables. Because the sample included both manufacturing rms and service companies, an industry dummy is included in the analysis. Almost 60% of the sample was industrial goods manufacturers. Therefore, this group was selected as the base to which consumer goods manufacturers (consumer) and service companies (service) were compared. The descriptive statistics of independent variables are presented in Appendix B. The descriptive statistics of the sample did not indicate any statistically signicant differences between truly traditional exporters and born internationals. Instead, among the traditional exporters and truly born international companies the export age was higher in the rst group, whereas international commitment and international experience of management were higher in the latter group. The correlations between various variables were usually quite low (see Appendix C), except between international orientation, international commitment, and international experience of the management. The variance ination factor (VIF) was analyzed to study the potential multicollinearity problems. A VIF value of less than 10 is considered indicative of the data having no such problems (see e.g., Grifths, Hill, & Judge, 1993). All the VIF values were below 3. Thus, no multicollinearity problems existed in the data.
5. RESULTS OF THE STUDY

5.1. Export Performance along Different Measures The results related to the performance along the ve measures of export performance used in this study are presented in Table 3. The results in the whole sample indicate that using a 15 scale for the measures, the performance was in the middle class (scale value 3) except in the case of using
40
JORMA LARIMO
Table 3.
Performance in SME Exports.

Truly Born Traditional Truly Born Traditional Internationals Exporters Internationals Exporters (N 194) (N 287) (N 77) (N 164) 3.23 3.24 3.34 32.57 2.04 3.33 3.33 3.46 47.35d 1.96 3.25 3.24 3.36 33.13 2.10d 3.39 3.47b 3.58c 67.78d 1.62
Total Sample (N 358)
Performance related to main goals (scale: 15) General foreign operation performance (scale: 15) Performance of the main product (scale: 15) Foreign sales ratio (0100) Objective nancial total performance during the last 3 years (scale: 5 alternatives; NEG. to X20%) Composite performance
3.28 3.29 3.41 40.58 2.00
3.09
3.02
3.17
3.02
3.40
Note: 1 Exports within 3 years, no share limits; 2 Exports within 3 years, shareX25%, FSALES in 2001X50%. Level of statistical signicance: a 0.1; b 0.05; c 0.01; d 0.001.
objective nancial total performance during the last three years (scale value 2). The more detailed mean performances based on the three subjective measures were very close to each other with the variation from only 3.28 to 3.41. The highest performance evaluation ratio was related to the performance of the main product. The mean foreign sales ratio was 40.58 indicating clearly a higher mean ratio than in several other studies focusing on SME exports. The mean objective nancial performance indicated that in general the operational prot of the companies has been between 0 and 4% in 1999 2001. The composite performance measure yielded a mean value of 3.09. A more detailed analysis of the performance results indicated that those based on the net results were clearly different, and that the variable correlated negatively with the share of foreign sales variable. The two classications used indicated clear differences in the sub-sample sizes. Using the looser denition for born international companies almost 55% of the sample was classied as born internationals, whereas using the tighter denition some 21% of the sample companies were born internationals. As discussed earlier, the latter denition indicates which rms can be regarded as really born internationals and therefore this sub-group was also labeled as truly born internationals. The performance results based on classication 1 between truly traditional exporters and born international companies indicated to rather
41
small differences between the two groups when the looser denition of born international companies was used. When considering the three subjective measures evaluations and composite measure mean, the values were somewhat higher in the born international group; whereas based on the objective nancial performance, the performance was somewhat poorer than in traditional exporter companies. The only case where a statistically signicant difference was found was based on foreign sales ratio, which was clearly higher in the born international company group. Classication 2 where a tighter operational denition of born international companies was employed indicated more differences in the performance between the traditional exporters and the truly born internationals. As in the case of the looser denition, the foreign sales ratio was clearly higher in the born international group (based on the operational denition this could also be expected). However, also the results based on two subjective performance measures, namely on general foreign operation performance and performance of the main product, the results indicated clearly better performance in the born international companies than in the traditional exporter group. Only in the case of performance related to the main goals no statistically signicant difference could be found; although even in this case the performance was better in the born international group. The born international companies have performed more poorly in the case of the objective performance. Apparently, the faster and more intensive international expansion had demanded a signicantly higher level of nancial resources that the results based on this will be, hopefully, better in the born international group at least after ve years when the foreign operations should be on a more stable basis (the truly born international companies were younger than the companies in the other subgroups). In the tight classication category, also the composite performance indicated statistically higher performance in the born international subgroup.
5.2. Relationships between Reviewed Variables and Various Export Performance Measures 5.2.1. Results in the Total Sample The results of the cross-sectional OLS regression in the total sample are presented in Table 4. All models, except for the objective nancial performance, were signicant with quite good explanatory power (with R2 in several models higher than 0.25).
42
Table 4.
Performance in SME Exports (Total Sample).

2. General Foreign Performance +d +d NS NS +d NS NS NS NS NS +b NS NS 0.310 0.279 10.142 3. Performance of the Main Product +b +c NS NS +d NS NS +a NS NS +b NS NS 0.279 0.247 8.750 4. Foreign Sales Ratio 5. Objective Financial Performance +a +c NS NS NS NS b NS NS NS NS NS NS 0.030 0.013 0.705 6. Composite
1. Performance Related to Main Goals Firm size Product/service quality Niche product/service Export age International orientation International commitment International experience Product adaptation Price adaptation Communication adaptation Market diversication Consumer Service R2 Adjusted R2 F-ratio +d +d NS +b +c NS NS NS NS NS +a NS NS 0.220 0.185 6.370
NS NS NS NS +c NS NS +a NS NS +c NS NS 0.221 0.186 6.400
+b +d NS NS +d NS NS NS NS NS +d NS NS 0.325 0.295 11.000
JORMA LARIMO
Note: NS Non-signicant. Level of statistical signicance: a 0.1; b 0.05; c 0.01; d 0.001.
43
The results in the total sample indicated that no variable had signicantly impacted on the export performance along all six measures (see Table 4). However, seven variables had a signicant relationship with export performance at least once. In the case of rm size, product/service quality, international orientation, and market diversication, a signicant positive relation was found using ve different performance measures. In the case of product/service quality and international orientation there were strong impacts independent of the measure of performance, whereas related to rm size and market diversication the level of the relation depended much more on the measure of performance. All four variables were signicant using the three subjective measures of export performance and composite measure; in turn the fth measure which was objective was the objective nancial performance in the rst two, and foreign sales ratio in the latter two cases. Additionally, noteworthy is that both product/service quality and international orientation were along all ve measures signicant at least at the 0.01 level. Thus, the results give support to the earlier ndings by e.g., Knight and Calantone (2000), Knight et al. (2004), and Lages et al. (2006) about the key role of quality and those by, e.g., Cavusgil and Zou (1994) and Nakos et al. (1998) about the role of commitment on export performance. Product adaptation had signicantly impacted along two, and export age along one measure. Product adaptation was signicant, based on one subjective and one objective measure while the export age was signicant, but only mildly, on one subjective measure. In addition, international experience was once signicant, but against expectation a mild negative impact on export performance was found. In four cases niche product/service, international commitment, price adaptation, and communication adaptation the variables did not have any signicant impact along any of the six export performance measures. The greatest surprise was the non-signicance of international experience. As stated before (see Table 1), earlier studies have indicated a strong positive impact of international experience on export performance. The impact of price adaptation and international commitment of the management have also been reviewed rather intensively and the especially concerning the latter variable the results have in a majority of studies indicated a positive relation, whereas in the former case the results have been more mixed. The impact of niche focus and communication adaptation on export performance has been so far analyzed only to a limited extent. Analyzing the former case, the results have in all cases supported a positive impact on performance, but they have been more mixed when considering the communication adaptation. The relatively low importance of product adaptation and total
44
JORMA LARIMO
non-signicance of advertising adaptation may be explained by the fact that if a rm offers a high-quality product that meets customers needs, countryspecic adaptation is not necessary for success, as also concluded by Styles and Ambler (1994). In addition to the non-signicance of those four variables, the results indicated that there were no signicant differences in the results depending on the eld of industry of the sample rms. In conclusion, it can be argued that the hypotheses 1, 2, 5, and 11 were validated because those variables indicated statistically signicant support using ve measures of performance. In the case of hypotheses 4 and 8 the support is only partial, and the rest of the hypotheses (3, 6, 7, 9, and 10) did not receive any support in the total sample. In hypothesis 12, it was expected that the relationships between adaptation of marketing mix variables and export performance to be more signicant when the export performance is interpreted in terms of subjective performance measures and export sales intensity than when the performance is measured using objective prot measures. As discussed above, price and communication adaptation were insignicant using all six measures. Only product adaptation had the expected positive impact on performance although only mildly in two cases, of which one was a subjective measure and the other one the foreign sales ratio. Thus, the results give mild support to the hypothesis but only in the case of product adaptation. 5.2.2. Classication 1: Truly Traditional Exporters vs. Born Internationals The results based on the looser denition of born internationals, i.e., in the truly traditional and born international samples are presented in Table 5. As expected from the analysis of the total results, the clearly lowest explanatory powers were also found now when objective nancial performance was used as the measure of performance. In the other cases, however, the models explanatory powers ranged between satisfactory and good. In the two sub-groups, both international orientation and product/service quality had a statistically signicant inuence using ve and respectively four different performance measures. Interesting to note was that international orientation did not have any statistically signicant inuence when objective nancial performance was used, but in the born international sub-group the quality variable was signicant when this measure was employed. Also a third variable, rm size, was signicant in all seven cases. Thus, these three variables seemed to be the most important ones in both sub-groups. In addition, market diversication was in total signicant in four cases, but three of those situations concerned traditional exporter sub-group. Thus the market diversication variable seems to be more important only in this sub-group.
Table 5.
Performance in SME Exports Based on Classication 1: Truly Traditional Exporters vs. Born Internationals.
Truly Traditional Exporters (N 164) 1. 2. General 3. Performance Foreign Performance Related to Performance of the Main Main Goals Product 4. Foreign Sales Ratio b NS NS NS NS +b NS NS NS NS NS +b NS NS 0.252 0.185 3.767 Born Internationals (N 194) 4. Foreign Sales Ratio b NS NS NS NS +a NS NS NS NS NS NS NS NS 0.251 0.195 4.528 5. Objective 6. Financial Composite Performance Performance
5. Objective 6. 1. 2. General 3. Financial Composite Performance Foreign Performance Performance Performance Related to Performance of the Main Main Goals Product +c NS NS NS NS NS NS
a
Constant Firm size Product/service quality Niche product/ service Export age International orientation International commitment International experience Product adaptation Price adaptation Communication adaptation Market diversication Consumer Service R2 Adjusted R2 F-value
NS +b +d NS NS +b NS NS NS NS NS NS NS NS 0.238 0.170 3.491
NS +a +c NS NS +b NS NS NS NS NS +a NS NS 0.368 0.312 6.501
NS NS +b NS NS +b NS NS NS NS NS NS NS NS 0.262 0.196 3.966
+b +c +c NS NS +b NS NS NS NS NS +d NS NS 0.353 0.310 8.197
+c +d NS NS NS +a NS NS NS NS NS NS NS NS 0.206 0.148 3.521
NS +d +c NS NS +c NS NS NS NS NS NS NS NS 0.316 0.265 6.251
+a +b +b NS NS +c NS NS NS NS NS +a NS NS 0.312 0.261 6.146
NS NS +c NS NS NS NS
b
+a +a +c NS NS +a NS NS NS NS NS NS NS NS 0.323 0.223 3.230
NS NS a NS NS NS 0.101 0.021 1.258
NS NS NS NS NS NS 0.095 0.028 1.413
Note: NS non-signicant. Level of statistical signicance: a 0.1; b 0.05; c 0.01; d 0.001.
45
46
JORMA LARIMO
Product adaptation, niche focus, price adaptation, international commitment, and export age were all insignicant in both sub-groups using all six measures of performance. In the case of international experience, the variable was once statistically signicant in both sub-groups. This was the case when employing the foreign sales ratio as the measure of performance and in both cases the impact was opposite to the negative expectations (in the born international sub-group somewhat greater than among traditional exporters). Similarly, the results indicated that the industry variable did not have any greater impact using any of the performance measures. Finally, what concerns communication adaptation, the variable was all the time insignicant in the born international sub-group, whereas among traditional exporters the variable had once a signicant negative impact. Having in mind that the above ndings were identied when the objective nancial performance variable was at work, it can be argued that the real costs of the adaptation of communication have been greater than the increase in sales. 5.2.3. Classication 2: Traditional Exporters vs. Truly Born Internationals The results based on the tighter classication of the born international companies are presented in Table 6. The explanatory power of the objective nancial performance model is again low, but only in the traditional exporter sub-group. In the truly born international sub-group model, the explanatory power is much higher. Also in all other models the explanatory power is at least rather good. In the total results six of the eleven variables had at least once a statistically signicant impact, but using the tight classication altogether, nine of the variables had a statistically signicant impact at least once. In general, rm size, product/service quality, international orientation, and market diversication, most often had a positive impact on export performance. All of them were signicant in using ve different measures of performance in the traditional exporter sub-group. In the born international sub-groups, rm size and international orientation were signicant in three cases and the other two in fewer cases. Thus, especially in the traditional exporter group, the results gave additional support of the great impact of these four variables. Similarities in the results between the sub-groups were that both product adaptation and niche focus were against expectations insignicant using all of the six performance measures. Price adaptation had a signicant positive impact on the truly born international sub-group three times. Price adaptation was once important also in the traditional exporter sub-group, but the sign was against expectations negative. It is difcult to explain the nding, especially why such a result was
Table 6.
Performance in SME Exports Based on Classication 2: Traditional Exporters vs. Truly Born Internationals.
Traditional Exporters (N 281) 1. 2. General 3. Performance Foreign Performance Related to Performance of the Main Main Goals Product 4. Foreign Sales Ratio b NS NS NS +c +a NS NS NS NS NS +d NS NS 0.266 0.299 7.240 Truly Born Internationals (N 77) 5. Objective 6. 1. 2. General 3. Financial Composite Performance Foreign Performance Performance Performance Related to Performance of the Main Main Goals Product +d +b +b NS NS NS NS NS NS NS a NS NS +a 0.099 0.054 2.202 NS +b +c NS NS +b NS NS NS NS NS +b NS NS 0.373 0.310 5.864 +b +b NS NS NS NS NS NS NS +b NS NS NS NS 0.245 0.084 1.523 +b +b NS NS NS +b NS NS NS NS NS NS NS NS 0.340 0.199 2.417 +b NS NS NS NS NS NS NS NS +c NS NS NS NS 0.370 0.236 2.759 4. Foreign Sales Ratio NS NS NS NS NS +b NS NS NS NS +b a NS NS 0.466 0.352 4.099 5. Objective 6. Financial Composite Performance Performance
Constant Firm size Product/service quality Niche product/ service Export age International orientation International commitment International experience Product adaptation Price adaptation Communication adaptation Market diversication Consumer Service R2 Adjusted R2 F-value
+b +d +d NS b +b NS NS NS b NS +b NS NS 0.241 0.203 6.360
NS +c +d NS NS +b NS +a NS NS NS +c NS NS 0.358 0.326 11.138
+b +a +c NS NS +b NS NS NS NS NS +c NS NS 0.290 0.254 8.163
NS NS +b NS +b NS +c b NS +a NS NS NS NS 0.345 0.206 2.476
NS +b +c NS NS +b NS NS NS NS NS +b NS NS 0.373 0.310 5.864
Note: NS non-signicant. Level of statistical signicance: a 0.1; b 0.05; c 0.01; d 0.001.
47
48
JORMA LARIMO
obtained when performance was measured using the main goals of the rm. A second clear difference between the sub-groups was found in relation to the impact of market diversication. The variable had the expected positive impact four times in the traditional exporter sub-group, whereas only twice in the truly born international sub-group. From those two cases in the latter group, only one was according to the expectations positive, whereas one although mild was negative (using foreign sales ratio). Thus those truly international companies which had diversied very widely do not seem to have gained such strong positions; those having concentrated somewhat more their operations had succeeded better and therefore also the role of foreign sales among the total sales was higher. A third difference between the sub-groups was related to the international experience. The variable was signicant in both sub-groups once but using different measures and the sign differed from one group. In the truly born international sub-group, a negative sign was found using objective nancial performance. Thus, the result was similar to that found using the loose denition of born internationals. In the traditional exporter sub-group, the sign was positive using general foreign operation performance measure. As such, in this sub-group the sign and measure were different compared to the ndings using the loose denition. Furthermore, export age was signicant in three cases, two times in the traditional exporter and once in the truly born international sub-group, and every time using different measures. In two cases, the expected positive impact was found but in the traditional exporter group the results indicated against expectations once a negative impact. Therefore, in this case the traditional exporters having long export experience had denitely not reached the goals set for the export operations. As discussed earlier, the export age did not have any signicant impact in the other analyses. A difference from the earlier results was also a nding that international commitment was in the truly born international subgroup once (using objective nancial measure) signicant and the results indicated also the expected positive sign. Because the positive relation was found only in one of the twelve possibilities, the hypothesis receives only a mild support. Finally, when the performance was evaluated using objective nancial performance, the results indicated the only case when the industry of the sample rms had somewhat more inuence on the results. In this case service companies had performed better than industrial (and consumer) goods companies. In summary, the results based on the tight denition of born internationals do not give full support to any of the hypotheses 111. Hypotheses 1, 2, 5, and 11 mainly receive support in the traditional exporter sub-group; in
49
turn in the truly born international sub-group hypotheses 1, 5, and 7 receive partial support. Furthermore, hypotheses 2, 4, and 10 receive limited support. In the other cases either the support is so mild (once at the 0.1 level) and/or the results are mixed. Thus the other hypotheses are rejected. In hypothesis 13, it was expected that the positive relation between niche focus, international orientation, and international commitment with export performance is greater in born international companies than in traditional exporting companies. As discussed above, the niche focus variable was insignicant in both sub-groups using both classications and all six performance measures. International orientation was signicant in both sub-groups several times, but using the tight classication, it was more often signicant in the traditional exporter sub-group than in truly born internationals. International commitment had the expected positive impact only in one situation in the truly born international sub-group. Consequently, it may be said that the hypothesis 13 receives only extremely mild support and only in the case of international commitment. In hypothesis 14, it was expected that the positive impact of adaptation of various marketing mix elements on export performance would be lower among born internationals than among traditional exporters. As mentioned before, the results did not give support to this hypothesis. In fact, there was limited support for the opposite results. Thus hypothesis 14 is rejected.
6. SUMMARY AND CONCLUSIONS

Export performance has received a lot of attention since the pioneering studies of the early 1960s. However, there have been very mixed results and the general view is that more carefully planned research is needed in order to analyze the export performance. The present paper had three main goals. To analyze: (1) the impact of the selected rm, management, and the export strategy-related variables to the export performance; (2) the possible variation in the results depending on the measure of export performance; and (3) the similarities and differences in the results depending on the type of SME traditional exporters vs. born international companies. There are a lot of studies focusing on export performance. However, several of them have employed only one or two measures of export performance. Furthermore, especially from the late 1990s there have been several studies reviewing the export and internationalization behavior of the so-called INVs or born globals (in this study, the term born international was used because the term refers better to the nature and operations of these rms). However,
50
JORMA LARIMO
several of those studies have been case or very small-scale survey studies. Thus, the empirical evidence from export performance along several performance measures, based on large-scale samples, discriminating in the analysis between traditional exporter vs. other types companies referred to as born internationals in this study is much more scarce, especially based on empirical evidence from smaller Organization for Economic Cooperation and Development (OECD) countries. Departing from the presented literature review, 12 hypotheses were developed regarding the relationships between various rm, management, and export strategy-related variables, and export performance. Furthermore, two additional hypotheses were developed in relation to the expected differences in the impacts of various variables between traditional exporters and born internationals. The empirical part of the study was based on the answers of 386 Finnish SMEs in a survey study conducted in early 2002. The export performance was interpreted using six different measures for export performance and the sample companies were distributed into two subgroups rst using a looser denition and then a tighter denition born international companies. The OLS regression was employed for the analysis. The results of the study are summarized in Table 7. Among the hypotheses 111, none received support in all reviewed situations. However, hypotheses 1, 2, 5, and 11 received all support using ve measures of performance. In these conditions, the rm size, product/service quality, international orientation of the management, and market diversication seemed to be in general the main reviewed variables having positive impact on export performance. In addition to these, two variables had positive impact product/service adaptation and export age but more limited. The results indicated some variation depending on the measure of performance. Thus, it can be concluded that also the results of this study give support to the view that the performance should be analyzed using several measures of export performance. Results in the majority of earlier studies have also indicated positive impact of international orientation and market diversication on export performance. Thus the results of this study coincide in this respect well with earlier ndings. What concerns the impact of product/service quality and especially rm size, the results have been much more mixed. The results in this study coincide with the recent ndings, e.g., by Calantone et al. (2006) and Lages et al. (2006) of the positive impact of quality and those by Voerman (2004) and Majocchi et al. (2005) about the positive impact of rm size on export performance. Against expectations, niche focus, international commitment of the management, and price and communication adaptation were not found to have
Table 7.
Expected Sign Total
Summary of the Results.

Truly Traditional Exporters Partially supported Born Internationals Traditional Exporters Truly Born Internationals Partially supported
H1 H2 H3 H4 H5 H6 H7
Firm size Product/service quality Niche product/service Export age International orientation International commitment International experience
+ + + + + + +
Mainly supported Mainly supported Not supported Limitedly supported Mainly supported Not supported Not supported/ limitedly opposite Partially supported Not supported Not supported
Mainly supported
Mainly supported
>Mainly supportedo >Not supportedo >Not supportedo >Mainly supportedo Limitedly supported Partially supported
>Not supportedo >Not Supported/limitedly oppositeo
H8 H9 H10
Product adaptation Price adaptation Communication adaptation Market diversication Performance of the main product intensity & composite Niche, international orientation & international communication Product adaptation of product, price & communication
+ + +
>Not supportedo >Not supportedo Partially supported
>Not supportedo
H11 H12
+ Stronger relationships
Mainly supported Not supported
Partially supported Limited support for composite
Limitedly supported Limited support for main product & composite Not supported
Mainly supported Limited support for composite
Limitedly supported Not supported
H13
Stronger in BGs
Not supported
H14
Stronger in traditional exporters
Not supported
Not supported
51
52
JORMA LARIMO
a positive impact along any of the six performance measures. The earlier ndings have given strong support for the positive impact of management commitment on export performance. The mean level of commitment was high in this study. Thus, even management in poorly performing rms was highly committed to exports, which explains the non-signicant impact on export performance. Furthermore, the results related to the impact of adaptation are similar to those found by Shoham (2002). Furthermore, once (based on objective performance) the results indicated surprisingly a negative relation between international experience of the management and export performance. Finally, the industry (manufacturers of industrial goods or consumer goods or service eld companies) of the reviewed companies had only once inuenced the results. The results evidenced also some similarities and differences depending on the type of the SME. Among both traditional exporters and truly born international companies, the rm size and international orientation had the greatest positive impact on export performance. Additionally, the product/ service quality and market diversication had clearly more often a positive impact on export performance among traditional exporters, than in the case of truly born internationals. The less signicant impact of the quality issue in latter case is more difcult to explain, whereas the market diversication issue is at least partly justied by the fact that truly born international rms were in general more diversied than traditional exporters. An unexpected result was that in the truly born international sub-group the price adaptation had a positive impact on export performance using three different measures of performance (included both subjective and objective measures). The results among traditional exporters vs. born internationals, using both looser and tighter classications, indicated no impact of the niche focus, adaptation of product/service, price, and communication along any of the six measures of performance. International commitment was also signicant in only one case. Moreover, no differences could be found depending on the industry of the sample companies. The insignicant role of the product/service and communication adaptation may be partly explained by the fact that in the case of high-quality products there may be less pressure toward local adaptation than in the case of poorer quality products/services. A similar nding was also made in the study focusing on UKbased export award winners by Styles and Ambler (1994). They assumed that the result may have been inuenced by the fact that the most important export markets for the sampled rms were Western Europe and USA, thus culturally rather similar areas/countries to the UK. Some of the main target countries in this study, e.g., Sweden and Germany are also culturally close
53
to Finland, thus the same argument may be valid at least partially also in this study. In conclusion, the results of the study indicate that: (1) the export performance is inuenced by the rm, management, and export strategy-related variables, (2) the results are at least partly dependent on the measure of export performance, (3) there are some similarities, but also some differences in the variables inuencing export performance among traditional exporters and born internationals, and (4) the differences depend on the operationalization of the born international company. Especially the relation between the types of exporting SME traditional exporters vs. born internationals has been analyzed very limitedly so far. Furthermore, there has been a lot of variation in the operationalizations used for born internationals. Thus, this study provided with more detailed information about the inuence of various operationalizations of the born international company on the export performance. If the results of this study are compared with those of earlier studies, one has to remember that the companies in this study were rather small (all SMEs) but the foreign sales usually had a very signicant role at least partly based on the small size of the Finnish domestic market. Most foreign studies have focused on companies having bigger domestic markets, when compared to Finland, and also the sample companies have usually been bigger (maximum limit of 1,000 employees for the SMEs in some studies), but the role of foreign sales has not been as signicant as in the sample of this study. The results give also basis for some important managerial implications. The reliance on quality as a competitive strategy was great in the whole sample, but the best performing exporters had given even more weight to the quality of the product/service than the lower performing exporters. Furthermore, the results evidenced the importance of the international orientation of the management for the export performance. The results also showed that good export performance usually demands at least somewhat bigger size from the company in order to have enough nancial and managerial resources for the successful export entry and expansion. Moreover, the market spreading seemed to lead to better performance than the market concentration, especially among traditional exporters companies. Finally, the product adaptation seemed to be more important than the price or communication adaptation in the total sample. However, among born internationals the product adaptation did not seem to be as important as price adaptation. This explained mainly by the high quality of the products/services provided by the born internationals why country-specic adaptation of the product offering is not necessary for success. Furthermore, more standardized products are also
54
JORMA LARIMO
expected to be typical for born internationals in order to be able to expand sales rapidly to several target countries. From the government policy point of view, key issues are how to develop the export systems to t to support the increasing amount of born internationals and also to the increasingly diversied types of born international companies. Nevertheless, the present study has several limitations. The study did not include any information about, for instance, of the motives for exports, main export markets, competition, and type of customers and/or distribution arrangements. In this light, these aspects could be added in future analyses in order to have a more comprehensive view of the export-related variables and export performance. Additionally, the adaptation vs. standardization issues were analyzed only at a very general level. Thus a more detailed analysis is needed. Furthermore, the nancial performance concerned the total operation of the company. In truly born international companies this was no problem because of that, on average almost 70% of the total sales in our sample was generated abroad and it seems that they do not separate between domestic and foreign protability. Therefore, this may have possibly been a problem only in those cases where the foreign sales had a much smaller role in the total sales, and the protability of domestic vs. foreign sales could be analyzed separately in future studies. This study focused on traditional exporters vs. born international rms. There was no analysis of the so-called born-again global rms (see Bell et al., 2001). One possibility for further research would be to concentrate more on this sub-group of exporters. Finally, this study analyzed only the direct impact of reviewed variables on export performance. Consequently, in the future research more developed methods of analysis should be used to review also the indirect effects of various variables on export performance. This would give more detailed information, e.g., about the possible differences in the impact of product/ service, communication, and price adaptation on performance in different elds of industries, and in the case of various export age and market diversication groups. Additionally, it would be worth taking into account that the relationships between the various independent variables and the export performance are not necessarily linear but curvilinear. This avenue of research requires further consideration as well.
ACKNOWLEDGMENTS
The study is a part of the Finnish Companies and Challenges of Globalization (LIIKE) research project in Finland. The author expresses his
55
sincere thanks to Johanna Pulkkinen, Jukka Harju, and Huu Le Nguyen for their help in various states of the data collection and treatment. Furthermore, the author acknowledges for the nancial support provided by the Academy of Finland and the Foundation for Economic Education in Finland.
REFERENCES
Aaby, N.-E., & Slater, S. (1989). Management inuences on export performance: A review of the empirical literature 197888. International Marketing Review, 6(4), 726. Aspelund, A., & Moen, O. (2001). A generation perspective on small rm internationalization. From traditional exporters and exible specialists to born globals. In: C. Axinn & P. Matthyssens (Eds), Advances in international marketing 11: Reassessing the internationalization of the rm (pp. 197225). Greenwich, CT: JAI Press Inc. Aspelund, A., & Moen, O. (2005). Small international rms: Typology, performance and implications. Management International Review, 45(3), 3757. Baldauf, A., Cravens, D., & Wagner, U. (2000). Examining determinants of export performance in small open economies. Journal of World Business, 35(1), 6179. Bell, J., McNaughton, R., & Young, S. (2001). Born-again global rms: An extension to the born global phenomenon. Journal of International Management, 7(3), 173190. Bonaccorsi, A. (1992). On the relationship between rm size and export intensity. Journal of International Business Studies, 23(4), 605636. Buzzell, R., & Gale, B. (1987). The PIMS principles: Linking strategy to performance. New York: The Free Press. Calantone, R., Kim, D., Schmidt, J., & Cavusgil, T. (2006). The inuence of internal and external rm factors on international product adaptation strategy and export performance: A three-country comparison. Journal of Business Research, 59(2), 176185. Calantone, R., & Knight, G. (2000). The critical role of product quality in the international performance of industrial rms. Industrial Marketing Management, 29(6), 494506. Cavusgil, T., & Kirpalani, V. (1993). Introducing products into export markets: Success factors. Journal of Business Research, 27(1), 127. Cavusgil, T., & Zou, S. (1994). Marketing strategyperformance relationship: An investigation of the empirical link in export market ventures. Journal of Marketing, 58(1), 121. Chadee, D., & Mattsson, J. (1998). Do service and merchandise exporters behave and perform differently? A New Zealand investigation. European Journal of Marketing, 32(9/10), 830842. Chetty, S., & Hamilton, R. T. (1993). Firm-level determinants of export performance: A metaanalysis. International Marketing Review, 10(3), 2634. Contractor, F., Hsu, C. C., & Kundhu, S. (2005). Explaining export performance: A comparative study of international new ventures in Indian and Taiwanese software industry. Management International Review, 45(3), 83110. Special issue. Das, M. (1994). Successful and unsuccessful exporters from developing countries: Some preliminary ndings. European Journal of Marketing, 28(12), 1933. Dean, D., Menguc, B., & Myers, C. (2000). Revisiting rm characteristics, strategy, and export performance relationship: A survey of the literature and an investigation of New Zealand small manufacturing rms. Industrial Marketing Management, 29(5), 461477.
56
JORMA LARIMO
Diamantopoulos, A. (1999). Export performance measurement: Reective versus formative indicators. International Marketing Review, 16(6), 444457. Diamantopoulos, A., & Inglis, K. (1988). Identifying differences between high- and lowinvolvement exporters. International Marketing Review, 5(Summer), 5260. Dominguinhos, P., & Simeos, V. (2004). Born globals: Taking stock, looking ahead. Paper presented at the EIBA 2004 annual conference in Ljubljana, Slovenia, December 2004 (conference cd-rom). Douglas, S., & Wind, Y. (1987). The myth of globalization. Columbia Journal of World Business, 22(Winter), 1929. Erramilli, M. K. (1991). The experience factor in foreign market entry behavior of service rms. Journal of International Business Studies, 22(3), 479501. Fenwick, I., & Amine, L. (1979). Export performance and export policy: Evidence from the U.K. clothing industry. Journal of the Operational Research Society, 30(8), 747754. Francis, J., & Collins-Dodd, C. (2000). The impact of rms export orientation on the export performance of high-tech small and medium-sized enterprises. Journal of International Marketing, 8(3), 84103. Gabrielsson, M., Sasi, V., & Darling, J. (2004). Finance strategies of rapidly-growing Finnish SMEs: Born internationals and born globals. European Business Review, 16(6), 590604. Gemunden, H. (1991). Success factors of export marketing. A meta-analytic critique of the empirical studies. In: S. Paliwoda (Ed.), New perspectives on international marketing (pp. 3362). London: Routledge. Grifths, W., Hill, R., & Judge, G. (1993). Learning and practicing econometrics. New York: Wiley. Jolly, V., Alahuhta, M., & Jeanet, J.-P. (1992). Challenging the incumbents: How hightechnology start-ups compete globally. Journal of Strategic Change, 1, 7182. Katsikeas, C., Leonidou, L., & Morgan, N. (2000). Firm-level export performance assessment: Review, evaluation, and development. Journal of the Academy of Marketing Science, 28(4), 493511. Katsikeas, C., Piercy, N., & Ioannidis, C. (1996). Determinants of export performance in a European context. European Journal of Marketing, 30(6), 635. Kaynak, E., & Kuan, W. (1993). Environment, strategy, structure, and performance in the context of export activity: An empirical study of Taiwanese manufacturing rms. Journal of Business Research, 27, 3349. Keegan, W. (1995). Global marketing management. Upper Saddle River, NJ: Prentice-Hall. Kirpalani, V. H., & Balcome, D. (1987). International marketing success: On conducting more relevant research. In: P. J. Rosson & S. Reid (Eds), Managing export entry and expansion: Concepts and practice (pp. 387397). New York: Praeger. Knight, G. (1997). Emerging paradigm for international marketing: The born global rm. Unpublished doctoral dissertation, Michigan State University. Knight, G., & Calantone, R. (2000). The critical role of product quality in the international performance of industrial rms. Industrial Marketing Management, 29(6), 493506. Knight, G., & Cavusgil, T. (1996). The born global rm: A challenge to traditional internationalization theory. In: T. Cavusgil & T. Madsen (Eds), Advances in international marketing (Vol. 8). Greenwich, CT: JAI Press Inc. Knight, G., Madsen, T. K., & Servais, P. (2004). An inquiry into born-global rms in Europe and the USA. International Marketing Review, 21(6), 645665.
57
Koh, A., & Robicheaux, R. (1988). Variations in export performance due to differences in export marketing strategy: Implications for industrial marketers. Journal of Business Research, 17, 249258. Kundhu, S., & Renko, M. (2005). Explaining export performance: A comparative study of international new ventures in Finnish and Indian software industry. In: D. Shephard & J. Katz (Eds), Advances in entrepreneurship, rm emergence and growth 8: International entrepreneurship. Greenwhich, CT: JAI Press Inc. Lages, C., Lages, L., & Lages, C. (2006). European managers perspective on export performance determinants. Journal of Euromarketing, 15(2), 7592. Leonidou, L., & Kaleka, A. (1998). Behavioural aspects of buyerseller relationships: Their association with export involvement. International Marketing Review, 15(5), 373397. Leonidou, L., Katsikeas, C., & Samiee, S. (2002). Marketing strategy determinants of export performance: A meta analysis. Journal of Business Research, 55(1), 5167. Luostarinen, R., & Welch, L. (1990). International business operations. Helsinki: Kyriiri Oy. Madsen, T. K. (1987). Empirical export performance studies: A review of conceptualizations and ndings. In: T. Cavusgil (Ed.), Advances in international marketing (Vol. 2, pp. 177198). New York: JAI Press Inc. Madsen, T. K. (1989). Successful export marketing management: Some empirical evidence. Journal of International Marketing Review, 6(4), 4157. Madsen, T. K., & Servais, P. (1997). The internationalization of born globals an evolutionary process? International Business Review, 6(6), 561583. Majocchi, A., Bacchiocchi, E., & Mayrhofer, U. (2005). Firm size, business experience and export in SMEs: A longitudinal approach to complex relationships. International Business Review, 14(6), 719738. Manolova, T., & Manev, I. (2004). Internationalization and the performance of the small rm: A review of the empirical literature between 1996 and 2001. In: M. Jones & P. Dimitratos (Eds), Emerging paradigms in international entrepreneurship (pp. 3763). Cheltenham, UK: Edward Elgar. Menon, A., Bharadwaj, S., & Howell, R. (1996). The quality and effectiveness of marketing strategy: Effects of functional and dysfunctional conict in intraorganizational relationships. Journal of the Academy of Marketing Science, 24, 299313. Moen, O. (2002). The born globals: A new generation of small European exporters. International Marketing Review, 19(2), 156175. Moen, O., & Servais, P. (2002). Born global or gradual global? Examining the export behavior of small and medium-sized enterprises. Journal of International Marketing, 10(3), 4972. Mohr-Jackson, I. (1998). Conceptualizing total quality orientation. European Journal of Marketing, 32(1/2), 1322. Nakos, G., Brouthers, K., & Brouthers, L. (1998). The impact of rm and managerial characteristics on small and medium-sized Greek rms export performance. Journal of Global Marketing, 11(4), 2347. Oviatt, B., & McDougall, P. P. (1994). Towards a theory of international new ventures. Journal of International Business Studies, 25(1), 4564. Oviatt, B., & McDougall, P. P. (1995). Global start-ups: Entrepreneurs on a worldwide stage. Academy of Management Executive, 9(2), 3043. Oviatt, B., & McDougall, P. P. (1997). Challenges for internationalization process theory: the case of international new ventures. Management International Review 37(special issue 2), 8599.
58
JORMA LARIMO
Oviatt, B., & McDougall, P. P. (2005). Dening international entrepreneurship and modeling the speed of internationalization. Entrepreneurship Theory & Practice, 29(5), 537553. Piercy, N. (1982). Export strategy: Markets and competition. London: George Allen & Unwin. Piercy, N., Kaleka, A., & Katsikeas, C. (1998). Sources of competitive advantage in high performing exporting companies. Journal of World Business, 33(4), 378393. Porter, M. (1990). The competitive advantage of nations. Harvard Business Review, 68(2), 7393. Preece, S., Miles, G., & Baetz, M. (1998). Explaining the international intensity and global diversity of early-stage technology-based rms. Journal of Business Venturing, 14, 259281. Rialp, A., Rialp, J., & Knight, G. (2005). The phenomenon of international new ventures, global start-ups, and born globals: What do we know after a decade (19932002) of exhaustive scientic inquiry? International Business Review, 14(2), 147166. Samiee, S., & Anckar, P. (1998). Currency choice in industrial pricing: A cross-national evaluation. Journal of Marketing, 62(3), 112127. Shoham, A. (1995). Global marketing standardization. Journal of Global Marketing, 9(1/2), 91119. Shoham, A. (1996). Marketing-mix standardization: Determinants of export performance. Journal of Global Marketing, 10(2), 5373. Shoham, A. (1999). Bounded rationality, planning, standardization of international strategy, and export performance: A structural model examination. Journal of International Marketing, 7(2), 2450. Shoham, A. (2002). Standardization of international strategy and export performance: A metaanalysis. Journal of Global Marketing, 16(1/2), 97120. Sousa, C. (2004). Export performance measurement: An evaluation of the empirical research in the literature. Academy of Marketing Science Review, 9. Available at http://www. amsreview.org/articles/sousa09-2004.pdf Styles, C., & Ambler, T. (1994). Successful export practice: The UK experience. International Marketing Review, 11(6), 2347. Thirkell, P., & Dau, R. (1998). Export performance: Success determinants for New Zealand manufacturing exporters. European Journal of Marketing, 32(9/10), 813829. Tookey, D. A. (1964). Factors associated with success in exporting. Journal of Management Studies, 1(1), 4866. Voerman, L. (2004). The export performance of European SMEs. Unpublished doctoral dissertation, University of Groningen. White, S., Grifth, D., & Ryans, J. (1998). Measuring export performance in service industries. International Marketing Review, 15(3), 188204. Young, S., Hamill, J., Wheeler, S., & Davies, J. R. (1989). International market entry and development. Upper Saddle River, NJ: Prentice-Hall. Zou, S., & Stan, S. (1998). The determinants of export performance: A review of the empirical literature between 1987 and 1997. International Marketing Review, 15(5), 333356.
59
APPENDIX A. OPERATIONALIZATIONS OF DEPENDENT AND INDEPENDENT VARIABLES

Dependent variable: Export performance Performance related to the main goals. The management was asked to rate their degree of satisfaction with the export performance related to the goals set in the main markets of the company on a 5-point Likert scale where 1 very disappointed y 5 very satised. General foreign performance. The management was asked to rate their degree of satisfaction with the export performance in all foreign markets on a 5point Likert scale where 1 very disappointed y 5 very satised. Performance of the main product. The management was asked to rate their degree of satisfaction with the export performance of their main export product on a 5-point Likert scale where 1 very disappointed y 5 very satised. Foreign sales intensity. The foreign sales intensity was measured asking the exact share of foreign sales from total sales of the company in 2001. Objective nancial performance. The management was asked to inform the level of the operating prot of the company in 19992001 on a 5-point scale where 1 negative, 2 04%, 3 59%, 4 1019%, 5 20% or more. Composite export performance. The composite export performance was built using the answers related to the above ve export performance measures. For the total scale, the export sales intensity was transferred to ve categories where a foreign sales share below 20% was rated as 1, a share of 2039% as 2 y and a foreign sales share of 80100% was rated as 5. Five groups were built: the rst group included cases where the total score was from 5 to 7, the second category included cases where the total score was from 8 to 12 y and the fth group included cases where the total score was from 22 to 25. Independent variables Firm size (FSIZE). The rm size was measured based on the total sales of the company in 2001 in million euros. Because it may be expected that the inuence is not linear, a logarithmic version was used. Product adaptation (PRODUCTADP). The product/service adaptation was measured on a scale from 0 to 5 where 0 indicated that there was no adaptation at all, 1 that there was only extremely limitedly adaptation, whereas 5 meant very signicant adaptation.
60
JORMA LARIMO
Product/service quality (QUALITY). The level product/service quality was measured based on the evaluations of the competitiveness of the company based on the level of product/service quality on a scale from 1 to 5 where 1 very low/poor y 5 very high/good. Niche product/service (NICHE). The level of niche focus in the operations was measured on the evaluations how well the following statement described the company: our product/service serves some special need that the competitors have problems to offer on a scale from 1 to 5 where 1 describes extremely poorly y 5 describes very well. Price adaptation (PRIADAPT). The level of price adaptation was measured based on a scale from 0 to 5 where 0 indicated that there was no adaptation at all, 1 that there was only extremely adaptation, whereas 5 meant very signicant adaptation. Communication adaptation (COMADAPT). The level of communication adaptation was measured on a scale from 0 to 5 where 0 indicated that there was no adaptation at all, 1 that there was only extremely adaptation, whereas 5 meant very signicant adaptation. International orientation (INTORIENT). The level of international orientation was measured based on the evaluations of the international orientation of the management on a scale from 1 very low/ poor y 5 very high/good. International commitment (INTCOMMIT). The level of international commitment was measured based on the evaluations of the international commitment of the management on a scale from 1 very low/ poor y 5 very high/good. International experience (INTEXP). The level of international experience was measured based on the evaluations of the international experience of management on a scale from 1 very limited y 5 very extensive. Export age (EXPORT AGE). The export age was measured based on the length of time the company has been exporting in years. Market diversication (MARDIV). The level of market diversication was measured based on the amount of target countries of exports in 2001.
61
APPENDIX B. DESCRIPTIVE SAMPLE INFORMATION

Total Sample N 358 Truly Traditional Exporters N 164 3.18 64.6 4.02 2.96 15,41 3.76 3.88 3.49 1.75 8.83 3.22 3.33 2.41 116 28 20 Born Internationls N 194 Traditional Exporters N 281 Truly Born Internationls N 77
Firm size (log) Firm size Product/service quality Niche product/service Export age International orientation International commitment International experience Market diversication (log) Market diversication Product adaptation Price adaptation Promotion adaptation Industrial Consumer Service
3.16 60.0 3.99 2.89 15.11 3.79 1.91 3.53 1.83 9.75 3.29 3.28 3.43 239 69 50
3.14 56.2 3.97 2.84 14,86 3.82 3.93 3.56 1.89 10.6 3.34 3.25 2.44 123 41 30
3.16 59.6 3.96 2.91 15,57b 3.73 3.85 3.46 1.81 8.94 3.14 3.19 2.38 192 46 43
3.14 57.5 4.09 2.84 13,43 4.03 4.13b 3.77b 1.89 12.9 3.53b 3.40 2.53 47 23 7
Note: 1 Exports Within 3 years, no share limits; 2 Exports within 3 years, shareX25%, FSALES in 2001X50%. Level of statistical signicance: a 0.1; b 0.05; c 0.01; d 0.001.
62
APPENDIX C. CORRELATION MATRIX

Mean STDV Firm size Product/ service quality Niche product/ service Export International International Inter-national Product age orientation commitment experience adaptation Price adaptation Communication Market adaptation diversication
Firm size 3.17 1.31 1 Product/service quality 3.98 0.67 0.12 Niche product/service 2.93 1.18 0.01 Export age 15.93 14.05 0.37 International orientation 3.72 0.98 0.16 International commitment 3.85 1.00 0.21 International experience 3.50 0.98 0.18 Product adaptation 3.29 1.19 0.23 Price adaptation 3.27 1.10 0.22 Communication 2.46 1.32 0.11 adaptation Market diversication 1.82 0.94 0.31
1 0.10 0.10 0.26 0.30 0.25 0.23 0.04 0.15 0.01
1 0.11 0.18 0.12 0.18 0.11 0.06 0.13 0.17
1 0.08 0.11 0.14 0.01 0.09 0.05 0.18
1 0.69 0.74 0.33 0.23 0.31 0.09
1 0.066 0.36 0.19 0.32 0.17
1 0.33 0.21 0.27 0.17
1 0.42 0.44 0.13
1 0.34 0.12
1 0.10 1
JORMA LARIMO
NICHE STRATEGY AND EXPORT PERFORMANCE

Antonella Zucchella and Giada Palamara
ABSTRACT
Small rms can approach foreign markets notwithstanding their limited resources by adopting a niche strategy. This permits to understand how SMEs can reach high levels of export intensity and broad geographic scope. Moreover, a global niche approach permits to explain among other factors why and how infant rms can be international or even global since their inception. The case studies analysis shows a positive relation between niche strategy and high international performance, in terms of export intensity, precocity, speed, and scope. The international expansion of niche rms is based on an horizontal micro-segmentation of the global market: they move internationally following global customers, independently from the psychic/geographical distance, and compete mostly on a non-price basis.
INTRODUCTION
Small rms have frequently approached international markets adopting a niche strategy. The latter can explain how small rms, which are traditionally described as constrained in terms of nancial and managerial resources, may reach a sustainable competitive positioning in global markets. Focusing on a narrow market segment through the offer of specialised goods or
63
64
ANTONELLA ZUCCHELLA AND GIADA PALAMARA
services enables small rms to compete globally. Moreover, the smaller the niche, the stronger the need to reach a global scope in order to break even. In market niches, small rms can deploy the resources and capabilities which characterise their organisations, in particular strong orientation to global customers, wherever they are located, an offer of customised goods, a customised pre- and after-sale service. The adoption of a niche strategy protects small rms from the competition of large multinationals. In markets dominated by economies of scale and scope in production, research and design, and marketing, small rms would not be able to compete. As Penrose (1959) suggested, the existence of market niches permits the survival and prosperity of small rms in the world economic systems. Actually, a number of authors recognise that the adoption of a niche strategy is not just a defensive move of rms looking for protected market spaces, but it entails entrepreneurial proactivity and innovativeness in shaping market niches (and not just discovering them). Moreover, the dynamism of global competition and the behaviours of large MNEs, which proved able of entering both small market niches, and mass markets, impose on a dynamic strategic approach to small rms in their niches. Building on these premises, the present contribution aims at verifying, through theoretical survey and 17 case studies research, the hypotheses that the adoption of a niche strategy enables small rms to better perform their export activity when compared to other SMEs according to the following dimensions: Precocity of internationalisation, i.e., niche rms are born international or start international activity from their rst years of life; Path of internationalisation, i.e., niche rms show a prevailing serial (simultaneous and fast) internationalisation process and not a sequential (incremental, step by step) one; Intensity of internationalisation, i.e., niche rms perform better in terms of foreign sales to total sales ratios; and Scope of internationalisation, i.e., niche rms show a broader geographic scope, since they focus on global customers wherever they are located, and not on foreign markets.
SMALL FIRMS AND NICHE STRATEGY

The academic literature has widely explored SMEs internationalisation, with special emphasis on its main outcome, typically represented by the
Niche Strategy and Export Performance
65
exporting activity. Among the different issues considered we nd the following ones: the drivers of export activity, which for some authors lie in the experience, attitudes, and traits of the entrepreneur and of the management team, while for others in business and industry specic factors (Tesar & Moini, 1998; Westhead, Wright, & Ucbasaran, 2001; Audretsch, 2002; Kundu & Katz, 2003); the strategic approach to internationalisation, outlining the path followed by rms facing international markets (Johanson & Vahlne, 1977; Cavusgil, 1980); the construct of export performance that represents one of the most meaningful indicators of international performance, especially in the case of small rms (Dess & Robinson, 1984; Venkatraman & Ramanujam, 1986; Shoham, 1988; Sullivan, 1994); and the emerging eld of international entrepreneurship (Venkataraman, 1997; Shane & Venkataraman, 2000; McDougall & Oviatt, 2000), and the related debate on born global rms, which leads to the identication of new patterns of international expansion, characterised by high precocity, speed, and wide scope (Oviatt & McDougall, 1994; Madsen & Servais, 1997; Rialp, Rialp, & Knight, 2005). According to the above-mentioned developments, the drivers and the dimensions of SMEs international success have been studied quite extensively, moving the focus from why do small rms export to how they perform this activity successfully (Audretsch, Prince, & Thurik, 1998). Issues regarding the decision maker and the entrepreneurial aspect, the industry- and countryspecic issues and the different stages of the internationalisation process have been widely studied (Miesenbock, 1988). At the same time, the relationship between business strategy and export performance is not adequately explored, and in particular, the literature about niche positioning as an important driver of the internationalisation strategy of SMEs is still scarce. SMEs are able to offer an attractive offering in terms of differentiation, which can be obtained only through the continuous improvement of the product, due to technology and know-how protection, and through the continuous attention to the clients needs. Global niche strategy can guarantee a leading position not only in the regional market, but also in the global one, notwithstanding the small dimension of rms. The internationalisation process for SMEs is primarily represented by the export strategy that still constitutes the most common foreign-market entry mode (Leonidou & Katsikeas, 1996; Wolff & Pett, 2000). Even though in the
66
last two decades small rms have diversied the avenues of their international growth paths and entry modes, exporting still represents the main way through which these rms sell abroad. It indeed offers a greater degree of exibility and an effective means for rms to achieve an international positioning without overextending their capabilities or resources (Young, Hamill, Wheeler, & Richard, 1989). Exporting is frequently coupled with importing (Monczka & Trent, 1991; Luthje & Servais, 2005), evidencing that the competitiveness of small rms in global markets depends actually also on their capacity to nd the needed resources and capabilities in different markets and/or to be strategic units in international value chains. Some studies have demonstrated that a rms size and export intensity as measured by the ratio of exports to sales are not correlated (Bonaccorsi, 1992; Calof, 1994; Wolff & Pett, 2000). This shows that, inside the world of SMEs, the smallest rms are not necessarily hindered from being strong exporters, and that export capabilities are more correlated with the rms organisational, entrepreneurial, managerial, and strategic elements. Schmalensee (1985) argues that market share and protability are strictly correlated: some contributions have been devoted to the emerging case of niche rms that, although small in size, appeared to be leaders in their market segment on a global scale (Maleksadeh & Nahavandi, 1985; Simon, 1996; Calof, 1994; Gomes Casseres, 1997; Kohn, 1997). Small rms can directly compete with large companies in narrow segments, and can reach a high share in international markets by choosing a niche strategy (Porter & Caves, 1977; Porter, 1979). On the other side, competing in a mass market is not sustainable for SMEs, because of the high entry barriers, mostly represented by the need of relevant nancial, technological, and managerial resources (La Marca & Palamara, 2005). Competition in global markets and scale advantages of large rms tend to push rms into niche markets (Christensen, 1991). In this case, the domestic market can be too small in terms of customer base, and rms need to sell in multiple countries to survive. A niche strategy involves (Dalgic & Leeuw, 1994; Mattiacci, 2000) specialisation, both endogenous (supply-side niche) and exogenous to the business (demand-side niche); scarcity, i.e., a production that is not quantitatively adequate for a potential mass market; competitive isolation, i.e., there is no actual substitutive product; strong customer orientation; and originality, i.e., a quality/innovativeness perception that the customer feels as exclusive to the rms offering.
67
The concept of niche strategy has assumed different meanings in the literature. Born as a sort of defensive reaction (Newman, 1978; Caves & Pugel, 1980) where the niche is considered as a protected area it has developed into a truly strategic concept, where the niche is the result of the segmentation of the market, in order to anticipate customers value functions. From a reactive strategy to compensate SMEs size disadvantage (Acs & Audretsch, 1990), to a strategic answer (entrepreneurial action) to the complexity of the global competitive arena. Niche strategy is a key competitive strategy for SMEs, allowing them to maintain a long-term competitive advantage also with reference to bigger rms (Kotler & Scott, 1993). It is a mix of unique resources, competences, and abilities, which create a favourable competitive environment for SMEs. This strategy reduces the degree of complexity of the international market, in terms of knowledge and control of the segment created; it reduces the complexity in the competitive environment, thanks to non-price competition and the absence of direct competitors; it reduces the degree of organisational complexity, thanks to the centralised organisational structure, typical of SMEs. Niche competitors generally select market segments characterised by nonprice competition, since they orient their products towards high levels of quality, technology content, and service for customers. Even in the case of manufacturing activities the cost competitive capability is not crucial (Ward, Bickford, & Keong Leong, 1996). The strategic approach to niche market assumes that no traditional barriers exist in entering a niche (as in the defensive approach): barriers exist in surviving and building a competitive advantage. A niche market is characterised by non-price competition: quality, innovation, and adaptation to customers needs are considered the fundamental competences that build new barriers to survival. Choosing to compete on a niche market, in a certain way, means selecting a competitive system based on intangible and competence-based barriers. The global niche approach does not in fact imply that there are any stable niches protected from competition (Maccarini, Scabini, & Zucchella, 2003). The dynamism of markets, the transversality of new technologies, and the recently acquired ability of large rms to respond with exible strategies and penetrate the market at niche level (also by acquiring small rms) create a continual challenge for small enterprises. The underlying goal is to reach a better and faster product adaptation to foreign customer needs, and to learn from the foreign context in order to improve their technology to meet local customers objectives and conditions.
68
The niche approach implies a non-generic vision of the market, turned to interpreting its complexity and variety, and to orienting the entrepreneurial action to the proactive clients service (Maccarini et al., 2003). These are fundamental characteristics both of a demand-based niche strategy where the niche is born from a specic and explicit need of the market and of a supply-based niche strategy determined by the research of a potential market of implicit needs.
NICHE STRATEGY AND SMALL FIRMS INTERNATIONALISATION

Operating within a narrowly dened market niche leads to an international market horizon in order to break even, since the domestic one at a small niche level does not permit adequate sales volumes to be reached even for small enterprises. All over the world, niche rms share some fundamental similarities: they possess unique assets, focus on narrow global market segments and are strongly customer-oriented. High export intensity, precocity, and global scope are thus expected elements characterising export performance of small rms, which have opted for a niche strategy. Moreover, we can expect that their internationalisation path shows different features, when compared to the one of non-niche small rms. The need of quickly reaching a relatively small number of customers dispersed across the globe, in order to break even, leads these rms to follow a simultaneous and fast international growth, while other rms more frequently adopt an incremental process of internationalisation, both in terms of timing and in terms of entry markets, starting from the closest and gradually moving the more distant ones. Johanson and Vahlne (1977) describe internationalisation as a process that begins and develops in incremental stages, slowly and gradually, grounded on the issue of commitment building. The rst step of the international process consists in exporting towards those countries that are physically and culturally the nearest; then, in the next phases, more committed forms are deployed. In addition to the deepening of their presence in foreign countries, rms move gradually from closest to more distant countries, i.e., they engage in widening processes as well. Both widening and the deepening processes are supposed incremental. The Uppsala model is based on the concepts of experience and knowledge: the internationalisation process is conceived as a gradual process of
69
acquisition, integration, and utilisation of knowledge concerning markets and externally performed transactions (experience accumulation). In the long run, this path leads to dimensional growth, to the increase of management know-how, to the constant growth of knowledge and to the increase of nancial resources. According to Johanson and Vahlne, the psychic distance that separates the home country from the foreign market causes the cautious attitude (grounded on the risk aversion hypothesis) at the beginning of the international process. This psychic distance is the sum of factors that disturb or, sometimes, hamper the ow of information between the rms and the markets in which they intend to operate. In the last two decades, the sequential model was challenged by the evolution of global markets and of rms strategies and did not always t internationalisation processes of small and infant rms. The global market, indeed, offers new chances to create and gain value, and SMEs started to develop a new approach to internationalisation, characterised by a serial process: a simultaneous internationalisation in different markets, frequently independent of both from the age of the rm and from psychic distance. Researchers and practitioners have started to study the phenomenon of the so-called born global rms, which are international from their inception (Oviatt & McDougall, 1994; Hordes, Clancy, & Baddaley, 1995; Madsen & Servais, 1997; Preece, Miles, & Baetz, 1998; Madsen, Rasmussen, & Servais, 2000; Larimo, 2001; Kuemmerle, 2002). Niche strategy frequently characterises born global rms as well as the ones that internationalise quickly and intensely. It is an approach to the international market that suits the characteristics of SMEs: dynamism, exibility, centralised governance, focus on product quality, and specialised production (Ward et al., 1996). In terms of approach to internationalisation, the main impact of a niche strategy is a shift towards a market vision in terms of clusters of homogeneous clients and no longer of countries. The latter issues help in explaining why they can quickly reach a global scope and also why they typically show a more widening attitude rather than a deepening one in foreign markets: their global customers are too much dispersed and each single market usually does not make increase in commitment economically and strategically convenient for these small rms. Relationships are kept with customers and not with countries, they are based on personal contacts and gradual commitment is built at the customer relationship level. The adoption of a niche strategy can explain why small rms are international from their inception, and reach high export performances in terms
70
of intensity and scope. It also explains how small rms internationalise, i.e., through which kind of process (notably a serial one, as opposed to the sequential process).
THE DIMENSIONS OF THE EXPORT PERFORMANCE: INTENSITY, PRECOCITY, SPEED, AND SCOPE
Even if internationalisation of an enterprise can cover different aspects of a rms life (export, international sourcing, international agreement, IDE, joint ventures, etc.), export strategy is still the primary foreign-market entry mode used by small businesses in their internationalisation efforts (Leonidou & Katsikeas, 1996; Wolff & Pett, 2000); it represents also the predominant international activity of small rms (Tesar & Moini, 1998; Westhead et al., 2001; Audretsch, 2002; Kundu & Katz, 2003). According to a recent European survey, the contemporary existence of import ows is quite common in SMEs (Servais, Zucchella, & Palamara, 2005) and supports the hypothesis that export growth and competitiveness depend among others on the capacity of the rm to access the best (optimal price quality combination) resources worldwide. Measuring export performance is not easy because export performance is a multidimensional construct (Shoham, 1988). Even if export intensity is the most accepted measurement, it is possible to dene three additional dimensions that contribute to better specify this measure and to explain its level (see Fig. 1). They are export precocity, speed, and scope. Export intensity has widely been used in international business literature as an indicator of SMEs international performance (Czinkota & Johnson, 1983; Calof, 1993), and the ratio of exports to total sales, is a typical measure of the degree of internationalisation (Ramaswamy, Kroeck, & Renforth, 1996), particularly for SMEs. The recent debate on international new ventures leads us to assume that international performance can be explained also through other dimensions, viz. precocity, speed, (Madsen & Servais, 1997; Lommelen, Matthyssens, & Pauwels, 2002) and scope (Tallman & Li, 1996). Oviatt and McDougall (1994, p. 49) describe the born global rm as a business organization that from inception seeks to derive signicant competitive advantage from the use of resources and the sale of outputs in multiple countries. Madsen and Servais (1997) speak about born globals as rms that adopt an international or even global approach right from their birth or very shortly thereafter.
71
Export intensity
Export performance dimensions Export precocity and speed Export scope
Fig. 1.
Research Model.
Literature on born global rms argues that precocity matters in international competition and that temporal dimensions of internationalisation, like precocity and speed, are important as much as the quantitative (export intensity) and geographic ones (scope) for a thorough understanding of export performance. Moreover, precocity and speed are positively correlated to export performance, and contribute to determining its level, as some empirical studies demonstrate (Denicolai, Palamara, & Zucchella, 2005). The sooner and the broader a rm goes international, the higher its export intensity is supposed to be. Precocity and speed inuence intensity because both are connected with the accumulation of experiential learning (Johanson & Vahlne, 1977; Kolb, 1984), which is crucial to maintaining competitive advantage in a global market. According to the fourth dimension of export performance, i.e., geographic scope, some authors divide rms in regional and global players (Maccarini et al., 2003; Servais et al., 2005), or between international and global players (Loustarinen & Gabrielsson, 2002; Gabrielsson & Gabrielsson, 2004), depending on the choice to operate on a macro-regional area that usually correspond to the home country or on the global market. The former are rms competing on a macro-regional market in our case the UE where natural and articial barriers have weakened. In these cases, new supranational and integrated spaces can be considered as the new domestic market. For an Italian rm, exporting to France, just like to a number of EU countries, is something different than exporting to extra-European markets, where a different currency and barriers to trade are present. In our research model, these rms are operationalised as having a narrow scope. The global player shows a global commitment, based on a strategic search of global customers. They have a broader scope, which means that they have different customers in one or more continents other than Europe.
72
Fig. 1 illustrates the research model developed through the aboveoutlined literature review on niche strategy and export performance dimensions. The model is tested in the following section though an exploratory empirical study on 17 case studies.
RESEARCH DESIGN AND METHODOLOGY

The empirical section of this work is based on the case study research (Yin, 1994). According to this methodological approach, we use a qualitative scale to measure all the variables, and we use a matrix positioning approach to summarise the main results. According to Marshall and Rossman (1998), qualitative research is appropriate for understanding phenomena such as entrepreneurs and rms orientation, action and behaviour, and is useful in investigations of interpretations and meanings of events. Our sample is a group of 17 manufacturing rms headquartered in Italy. All of them are SMEs according to UE parameters, which dene small rm as an independent organisation with a number of employees ranging between 10 and 49, and revenues less than h10 millions; the medium-sized rm is dened as an organisation with a number of employees ranging between 50 and 249, and sales volume less than h50 millions. All the rms in the sample have in fact less than 250 employees, total revenues and assets below the mentioned ceilings, and are independent (not belonging to national/ multinational groups). The group of rms was selected from a database reported by the Chambers of Commerce.1 According to this dataset, the selection made is representative of the population of small- and medium-sized international rms of Piedmont region (North West Italy). In these rms, entrepreneurs and top management usually coincide, due to their small scale. Table 1 summarises the relevant information about the surveyed group. The methodology of investigation adopted is a structured interview, submitted through a questionnaire. We have directly interviewed the entrepreneur of each rm; if an export manager exists, he/she was interviewed about the international activity. Each interview lasted on an average 1 hour and a half. The questionnaire is part of a wider research project going on in our department on the process of internationalisation of small rms and contains ve different sections: business activities and governance; strategic positioning; ownership and relations with other rms; international activities; and entrepreneurship.
Table 1.
Firm Product Year of Foundation
The Sample.
Time to Export Precocity Export Intensity Scope Approach to International Market Serial Serial Sequential Sequential Sequential Serial Serial Serial Serial Sequential Serial Serial Sequential Sequential Serial Serial Sequential Niche Strategy
Employees
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Ring for mechanic pad Toys Car components Wine Lever for steering gear Informatics for banking Printed circuits Injection systems Cosmetics Steel tubes Machinery for jewellery Laser machineries Car components Valves and engines Cages for champagne and sparkling wine Diamonds threads Numeric controls
1882 1968 1968 1963 1925 1980 2004 1999 2000 1930 1945 2001 1990 1953 1956 1984 1996
63 60 210 9 142 220 90 30 12 114 104 7 26 64 80 35 130
88 2 29 17 21 0 0 0 0 43 3 0 1 13 0 0 0
Low High Low Low Low High High High High Low High High High Low High High High
High Low Low High Low High High Low Low Low High High Low High High High Low
Broad Broad Narrow Broad Broad Broad Broad Broad Broad Narrow Broad Broad Narrow Broad Broad Broad Narrow
Yes Yes No No No Yes Yes No No No Yes Yes No No Yes Yes No
73
74
The questionnaire was designed after a review of the literature on international business, and through discussions and focus groups with practitioners, researchers, and entrepreneurs. Some problems stem from operationalising the concept of niche rm. In order to identify the niche competitors we analysed some key variables suggested by the above-mentioned literature: the specialised product, the target customer description, and the share on global market (Kotler & Scott, 1993; Dalgic & Leeuw, 1994). This was further integrated by the structured interviews in order to verify the effective niche orientation of the rms. Measuring international performance is complex because its operationalisation may refer to different aspects of the organisational effectiveness of the rms (Dess & Robinson, 1984; Venkatraman & Ramanujam, 1986; Shoham, 1988; Sullivan, 1994). In this work, as mentioned before, we accepted the theoretical framework that considers export performance as an indicator of the international performance (Czinkota & Johnson, 1983; Calof, 1993), but we also took into consideration the recent literature on born global rm, dening three more dimensions of export performance, viz. precocity, speed, and scope. We grouped the interviewed rms according to export performance indicators, on the basis of the following operationalisation criteria: low export performance (domestic share>export share on total sales) versus high export performance (export share>domestic share on total sales); early internationalising (time to exportp3 years) versus late internationalising (time to export >3 years); fast internationalising (export sales ratioX25% in the rst three years) versus slow internationalisation (export sales ratioo25% in the rst three years); and narrow scope (international players) versus broad scope (more than one continent, markets variety, global players). In addition to these measures we also distinguished rms according to the kind of internationalisation process, combining the constructs of scope, precocity, and speed (Zucchella, 2005). The serial approach to international market corresponds to a broad (global) scope and high growth rate from inception in international markets, while the sequential one corresponds to a narrow scope (international players, gradually internationalising from closest markets in terms of geographic and psychic distance). We are aware that there could be a number of borderline situations in terms of precocity, speed, intensity, and scope; and we are also aware of the limitations involved in adopting cut-off values in operationalising the variables.
75
Some of them are based on the literature, viz. early internationalising and speed (Madsen & Servais, 1997; Rialp et al., 2005) and scope (Gabrielsson & Gabrielsson, 2004; Maccarini et al., 2003; Servais et al., 2005). Others are the results of an arbitrary choice, inspired from the necessity of simple and clear representative measures, and based on a meaningful representation of the Italian business market. For this reason, a preliminary case study research grounded on qualitative information should help in dening better quantitative criteria and in rening research hypotheses.
RESEARCH FINDINGS
Fig. 2 shows the distribution of the case studies according to three static indicators of export performance (intensity, precocity, and scope). Intensity
Fig. 2.
The Three Static Dimensions of Export Performance.
76
(Czinkota & Johnson, 1983; Calof, 1993), precocity (Madsen & Servais, 1997), and scope (Tallman & Li, 1996; Gabrielsson & Gabrielsson, 2004) are positively linked. In particular, there is a cluster of rms in the area corresponding to high precocity (born global), broad scope, and high export intensity. This group portrays the international performance of the typical (and successful) niche rms, since the rms all share this strategic option. One of these cases is described in the box below. There are two minor groups in this representation: one is in an area where three rms show precocity and broad scope but they do not perform well in terms of intensity (only one is niche-oriented). The other one corresponds to three rms with high export intensity and broad scope, but characterised by late internationalisation (two are niche-oriented). Fig. 3 considers the dynamic dimension of export performance, as described by the features of the internationalisation process, serial versus sequential, according to the above-mentioned speed criteria, coupled with an analysis of the simultaneity of entry in different and distant foreign markets.
Fig. 3.
Niche Strategy Approach to International Market Matrix.
77
Niche rms, indeed, have had a fast internationalisation process, regardless the geographic or cultural distance: the case studies show that a sequential approach does not coexist with the niche strategy (Fig. 3). Our results conrm that niche strategy is strictly linked with the serial approach to internationalisation: the international expansion is based on a horizontal micro-segmentation of the global market, and rms move internationally following global customers, independent from the country, and the psychic/geographical distance. In the global niche approach, the rm overcomes the psychic distance problem and the difculty of dealing with the diversity of foreign markets. According to the recent literature on international business, the serial approach is typical of rms which internationalise from the inception (Oviatt & McDougall, 1994; Madsen & Servais, 1997), following global customers. Empirical evidence shows that in different business context, niche and customer orientation represent a core strategy for international new ventures. Literature on born global rms, indeed, has underlined the relevance of niche strategy in explaining the early and fast internationalisation processes. Other empirical evidences (Rennie, 1993) argue that the niche orientation is a critical factor for born global rms, this being the strategic model that permits to address narrow segments of clients wherever they are localised. Our ndings conrm the above-mentioned results, showing that the niche strategy is typical of born global rms (Fig. 4). It is interesting that more than the half of our rms (8 out of 17) are international from their inception, but not all of them have reached high export intensity ratios. In some cases it depends on their very short life, in other cases it seems to depend on the adoption of a sequential internationalisation strategy. In addition to this, a niche positioning is positively linked with a high speed of internationalisation, and, according to the typical customer orientation of the niche strategy (Aspelund & Moen, 2001), rms that compete in these markets have a broader scope, following their clients wherever they are located (Fig. 5). A niche approach does not coexist with a narrow scope of export activities. An entrepreneur said Why do you ask me in how many and which countries I sell my products? I have no idea y I can tell you how many customers I have, their names, their problems, obviously I know where they live but that is not a primary issue for us. Country matters only rarely when we need to adapt our product to some different foreign technical standards y when we deal with local bureaucracy y thats all!
78
Fig. 4.
Niche Strategy/Precocity Matrix.
Finally, all the above-mentioned items precocity, speed, and scope contribute in determining high export intensity for niche rms. According to Fig. 6, nearly all the niche rms interviewed show high export intensity ratios, except one, which is struggling to survive in a business (toys) where marketing investments and distribution channels tend to favour large rms. It is interesting to notice that all the other (performing) niche rms belong to the business-to-business market. In such markets, direct contact with global industrial customers supports customer orientation, problem solving, and strengthens client relationships without the need of large investments in marketing and promotion. Other studies seem to conrm the role of network embeddedness for niche rms performance (Echols & Tsai, 2005) and underline the importance of building a system of embedded ties together with the selection of product/market segment strategy. Global networks are deemed crucial for the development of born global rms (Knight & Cavusgil, 1996). High export intensity ratios are the result of such a strategy and yield on early and fast internationalisation and broad scope, which permitted to
79
Fig. 5.
Niche Strategy and Geographic Scope Matrix.
quickly cumulate experiential learning about different foreign customers. According to Fig. 4, the niche rms interviewed are more frequently early international than others (six niche rms out of eight). All the niche rms have a broad geographic scope, beyond the European market, and they all show a serial internationalisation process, because they went international quickly and with simultaneous entries in different markets, both close and distant in geographic and in psychic terms. There are two non-niche rms which show a serial internationalisation process and broad scope, but without reaching high export intensity: this may appear contradictory, but it could be explained with the young age of these rms. On the other hand, no niche rm shows a sequential pattern of international growth. An export manager describes the serial process as y discovery of our global customers, through business contacts and participating in international trade fairs. Shortly after foundation our rm was selling products both in Germany and India, with a growth rate of foreign sales around 20% per year, but we didnt make any country analysis to plan this entry y we established a contact with customers and met their requests, which were
80
Fig. 6.
Niche Strategy and Export Performance Matrix.
basically similar, regardless of their country differences. This is what we mean by global customer, someone who needs something you can sell, no matter where he/she is. The interviewed rms did not consider the possibility to move from export to more committed entry modes, apart from a couple of cases of light investments in sales and assistance subsidiaries in the main macro-regions, for two main reasons: The distinctive competences, which enable the offer of highly specialised and customised goods, derive from a unique combination of domestic (notably local) unique resources (human capital, social capital, specialised suppliers y). Producing at home and exporting is for many rms still the main international competitive strategy, as an entrepreneur declares y for our rm FDIs and JVs, and especially the manufacturing oriented ones, are out of question. It is not a matter of costs, it is a matter of nonreproducibility of our product uniqueness in foreign contexts. This outcome is conrmed by other studies which revealed that rms especially
81
during their founding period tend to rely on resources (skills, services, networks) which are typical of the cluster they belong to (Servais & Rasmussen, 1999) and develop competences and competitive offers which are rare and difcult to imitate. Commitment building is not made at the country level but at the customer relationship level, according to the global customer orientation characterising these rms (Aspelund & Moen, 2001). One entrepreneur declared, when we start a customer relationship, we know it is there we have to invest in order to build a long term and protable relationship. Our prot margin is highly indebted to long term relationships with customers: they are our main assets. Investing in long term relationships means travelling a lot, regularly visiting these people, talking with them on the phone weekly, understanding their needs and sometimes anticipating their needs. It is an investment in time and human resources, in research, design and technical assistance.
CASE NUMBER 16
It is a small rm with 35 employees, a turnover of h12 million. Born in 1984, it operates in the granite mining and rening industry. This rm has all the characteristics of Italian SMEs: a strong role of the entrepreneur, family business, and it does not belong to any group, even if it has some minority share in other Italian SMEs. Initially it was the only producer of diamonds threads to cut granite, even though later on other enterprises began to imitate the technology. International Activities and Performance The enterprise is the leader in the domestic market, and among the top 10 in the world market. The latter is divided with other four direct foreign competitors, but they are big rms, that operate on more market segments and product lines. The rm had a serial approach to the global market We were the only producer of diamonds threads, our natural market is the world, for such reason we have always looked at global customers, independently from their location. Barriers in the international activities were found only in tariffs and protectionism, while geographic or cultural distance where not considered a problem at all. It is a born global rm: export and import started in same the year of foundation, with a broad scope (UE, USA), and a percentage of export
82
on total sales, in the rst 3 years form inception, of 33%. In a few years, its products were present all over the world, with a EU share on total sales of 25%, an extra UE share of 70%, and only a 5% on the national market. The rm sells mostly in Spain, Portugal, France, Brazil, China, India, and in USA with a sales subsidiary, while it has representatives and agents in the other countries.
Niche Orientation The rm has focused its business only on the production of diamond threads, with the purpose to reach a critical mass of clients, to specialise production and to have a development in line with the demands of the market and with an immediate adjustment to the particular demands of the client. The customer orientation is very strong; the strategic approach is more customer than country oriented: the product can vary according to the demands of the clients. The strong customer orientation is the reason why the enterprise besides the activity of production also offers services to clients. The rm is aware that its technology is not well known yet: the enterprise sells the machinery to the client, and assists him/her/them to correctly use the diamond thread. The entrepreneur thinks that the competitive advantages of his rm are quality, innovation, customer adaptation, and assistance. Price is not considered an inuential element for competing in this business. The entrepreneur has travelled a lot, in order to learn from his customers, and to explore markets. He thinks an open mind, the ability to create a global and local network, commercial relationship, and the ability to have dialogue with other rms, also of other businesses, are fundamental to develop an international commitment. The entrepreneur himself is the manager of the international activity and of the strategic choices. He has driven his production to a global market from the inception, because the local one has never been considered as the only market. The rm tries to solve the problem of the technological risk that could derive from the introduction of an innovation, with a dynamic attitude and proactivity. The rm develops continuous research and aims at leading the innovation in stone cutting over time.
83
CONCLUSIONS
This contribution shows through case studies research that there is a positive relation between niche strategy and high international performance, in terms of export intensity, precocity, speed, and scope. Among the niche rms, most reveal what we consider the typical expected traits of successful niche rms: high export intensity, early and fast internationalisation, and broad geographic scope. The niche strategy is strictly linked with a serial approach to internationalisation: the international expansion is based on an horizontal microsegmentation of the global market, and rms move quickly following global customers, independent of the country and the psychic/geographical distance, and compete mostly on a non-price basis. The exploratory research permits to better rene the criteria of operationalisation of the variables: it is conrmed that export intensity is a signicant measure, but export performance is made of (depends from) precocity, speed, and scope, which appear important indicators for nicheoriented rms. The serial internationalisation path is a complex construct depending on all the above-mentioned variables, and supported by the interviews made, which permitted to improve its denition. The main limitation of this study is that the ndings of the case studies research cannot be generalised and they need to be tested through quantitative analysis with a signicant number of observations. The value of this qualitative research lies with the possibility it gives to build testable research hypotheses as well as with its capacity to highlight behaviour and strategic issues that are difcult to test quantitatively. The ndings have both policy and managerial implications. From the rst perspective, the frequency of a niche strategy in small rms represents an issue to be considered for designing appropriate services to support their internationalisation. Getting in touch with global customers is a core issue for niche-oriented rms, while accessing to foreign countries information is not so important. In addition to this, the joint effect of precocity and speed in internationalisation, which characterise many of the niche rms, involve some rethinking about a system of public (but also private) support services where start-up assistance is coupled with foreign trade support. Managerial implications of the study are also relevant. The actual arena of competition requires a global dimension of niche market: global niche allows rms to maximise opportunities in terms of experiential learning and segmentation. The niche strategy requires for rms management an
84
entrepreneurial drive through the discovery or creation of market niches, which can be viewed as business opportunities and the exploitation of their potentialities in terms of learning from customers and protable international growth. The niche positioning involves a strong customer orientation, which unfolds many opportunities of product/service improvement and innovation through customer needs understanding. This valuable experiential learning process is predominantly customer based and not country based and may lead to higher margins due to the lower competitive pressures in niche markets. A global niche is, on the other hand, a dynamic concept, and not a safe refuge from global competition. It involves the monitoring of potential competition and customer needs evolution (Maccarini et al., 2003): niche orientation not only requires a continuous customer orientation, but also continuous efforts in research and development, in product, process, marketing and organisational innovation, and in actual and potential customers monitoring.
NOTES
1. According to the Italian regulation, in each province a Chamber of Commerce is established by law and is a public institution. They represent the local system of rms and each rm has to apply for registration to this institution, providing data and paying an annual fee. They receive from the Chambers of Commerce a number of services, including foreign trade support services.
REFERENCES
Acs, Z. J., & Audretsch, D. B. (1990). Innovation and small rms. Cambridge: MIT Press. Aspelund, A., & Moen, O. (2001). A generation perspective of small rm internationalisation: From traditional exporters and exible specialists to born globals. In: C. Axinn & P. Matthyssens (Eds), Reassessing the internationalisation of the rm. London: JAI Elsevier Science. Audretsch, D. B. (2002). Entrepreneurship: A survey of the literature. London: European Commission, Enterprise Directorate General. Audretsch D. B., Prince, Y., & Thurik, A. R. (1998). Do small rms compete with large rms? Working Paper. Tinbergen Institute at Erasmus University Rotterdam. Bonaccorsi, A. (1992). On the relationship between rm size and export intensity. Journal of International Business Studies, 23(4), 605635. Calof, J. L. (1993). The impact of size in internationalization. Journal of Small Business Management, 31(October), 6069. Calof, J. L. (1994). The relationship between rm size and export behaviour revisited. Journal of International Business Studies, 25(2), 367387.
85
Caves, R. E., & Pugel, T. (1980). Intra-industry differences in conduct an performance: Viable strategies in U.S. manufacturing industries. Salomon brothers Centre for Study of Financial institutions. New York University, Monograph Series in Financial Economics, n.2. Cavusgil, S. T. (1980). On the internationalisation process of rms. European Research, 8(6), 273281. Christensen, P. R. (1991). The small and medium sized exporters squeeze: Empirical evidence and model reections. Entrepreneurship and Regional Development, 3, 4965. Czinkota, M. R., & Johnson, W. J. (1983). Exporting: Does sales volume make a difference? Journal of International Business Studies, (Spring/Summer), 47153. Dalgic, T., & Leeuw, M. (1994). Niche marketing revisited: Concept, applications and some European cases. European Journal of Marketing, 20(1), 3955. Denicolai, S., Palamara, G., & Zucchella, A. (2005). Drivers and dimension of export performance. Paper presented at the 8th Vaasa conference on international business, August 2123. Dess, G. G., & Robinson, R. B. (1984). Measuring organizational performance in the absence of objective measures: The case of privately held rms and conglomerate business units. Strategic Management Journal, 5, 265273. Echols, A., & Tsai, W. (2005). Niche and performance: The moderating role of network embeddedness. Strategic Management Journal, 26, 219238. Gabrielsson, P., & Gabrielsson, M. (2004). Globalizing international: Business portfolio and marketing strategies in the ICT elds. International Business Review, 13, 661684. Gomes Casseres, B. (1997). Alliance strategies of small rms. Small Business Economics, 9(1), 3344. Hordes, M., Clancy, J., & Baddaley, J. (1995). A primer for global start ups. Academy of Management Executive, 9, 711. Johanson, J., & Vahlne, E. (1977). The internationalization process of the rm. A model of knowledge development and increasing foreign market commitment. Journal of International Business Studies, 8, 2332. Knight, G., & Cavusgil, S. T. (1996). The born global rm: A challenge to traditional internationalisation theory. Advances in International Marketing, 8, 1126. Kohn, T. O. (1997). Small rms as international players. Small Business Economics, 9(1), 4551. Kolb, D. A. (1984). Experiential Learning: Experience as the source of learning and development. Englewood Cliffs, NJ: Prentice-Hall. Kotler, P., & Scott, W. G. (1993). Marketing management. Torino: Isedi. Kuemmerle, W. (2002). Home base and knowledge management in international ventures. Journal of Business Venturing, 17, 99122. Kundu, S. K., & Katz, J. A. (2003). Born-international SMEs: BI-level impacts of resources and intentions. Small Business Economics, 20(1), 2547. La Marca, P., & Palamara, G. (2005). Strategia di nicchia nel settore moda. Quaderni di ricerca ` del Dipartimento di Ricerche aziendali, Universita degli studi di Pavia, 4. Larimo, J. (2001). Internationalisation of SMEs two case studies of Finnish born global rms. Paris: EIBA Conference Proceedings. Leonidou, L. C., & Katsikeas, C. S. (1996). The export development process: An integrative review of empirical models. Journal of International Business Studies, 27(3), 517551. Lommelen, T., Matthyssens, P., & Pauwels, P. (2002) Accelerations in the internationalization process of logistics rms. 28th annual meeting of the European International Business Academy (EIBA), Athens, Greece.
86
Loustarinen, R., & Gabrielsson, M. (2002). Born globals of small and medium economies. Globalisation and marketing strategies. 28th annual meeting of the European International Business Academy (EIBA), Athens, Greece. Luthje, T., & Servais, P. (2005). Firms international sourcing and intra-industry trade. Quaderni ` di ricerca del Dipartimento di Ricerche aziendali, Universita degli studi di Pavia, 2 gennaio. Maccarini, M.E., Scabini, P., & Zucchella, A. (2003). Co-location and business specic issues in the internationalization of SMEs. Paper presented at the conference on Clusters, Industrial Districts and Firms: The Challenge of Globalisation. Modena, Italy, September 1213. Madsen, T. K., Rasmussen, E. S., & Servais, P. (2000). Differences and similarities between born globals and other types of exporters In: A. Yaprak, & J. Tutek (Eds), Globalization, the multinational rm, and emerging economies (pp. 247265). Advances in International Marketing, Vol. 10. Amsterdam: JAI/Elsevier Inc. Madsen, T. K., & Servais, P. (1997). The internationalization of born globals: An evolutionary process? International Business Review, 6(6), 561583. Maleksadeh, A., & Nahavandi, A. (1985). Small business exporting. Misconceptions are abundant. American Journal of Small Business, 9(4), 714. Marshall, C., & Rossman, G. B. (1998). Designing qualitative research. Thousand Oaks: Sage. Mattiacci, A. (2000). Il marketing strategico del business di nicchia. Padova: Cedam. McDougall, P. P., & Oviatt, B. M. (2000). International entrepreneurship: The intersection of two research paths. Academy of Management Journal, 43, 902908. Miesenbock, K. J. (1988). Small business and exporting: A literature review. International Small Business Journal, 6(2), 4261. Monczka, R., & Trent, R. (1991). Global sourcing: A development approach. International Journal of Purchasing and Materials Management, 27(2), 28. Newman, H. (1978). Strategic groups and the structureperformance relationship. Review of Economics and Statistics, 60, 417427. Oviatt, B. M., & McDougall, P. P. (1994). Toward a theory of international new ventures. Journal of International Business Studies, 25, 4564. Penrose, E. T. (1959). The theory of the growth of the rm. New York: Wiley. Porter, M. E. (1979). The structure within industries and companies performance. Review of Economics and Statistics, 61, 214227. Porter, M. E., & Caves, R. E. (1977). From entry barriers to mobility barriers. Quarterly Journal of Economics, 91, 241261. Preece, S. B., Miles, G., & Baetz, M. C. (1998). Explaining the international intensity and global diversity of early stage technology based rms. Journal of Business Venturing, 14, 259281. Ramaswamy, K., Kroeck, K. G., & Renforth, W. (1996). Measuring the degree of internationalization of a rm: A comment. Journal of International Business Studies, 27, 167177. Rennie, M. W. (1993). Global competitiveness: Born global. The McKinsey Quarterly, 4, 4552. Rialp, A., Rialp, J., & Knight, G. (2005). The phenomenon of early internationalising rms: What do we know after a decade (19932003) of scientic inquiry? International Business Review, 14(2), 147166. Schmalensee, R. (1985). Do markets differ much? American Economic review, 75, 341351. Servais, P., & Rasmussen, E. S. (1999). Born Globals connectors between various industrial districts? London, RENT XIII conference on entrepreneurship.
87
Servais, P., Zucchella, A., & Palamara, G. (2005). International entrepreneurship and sourcing: International value chain of small rms. Paper presented at the 8th Uddevalla symposium & McGill conference, September 1517. Shane, S., & Venkataraman, S. (2000). The promise of entrepreneurship as a eld of research. Academy of Management Review, 25, 217226. Shoham, A. (1988). Export performance: A conceptualization and empirical assessment. Journal of International Marketing, 6(3), 5981. Simon, H. (1996). Hidden champions. Boston: HBS Press. Sullivan, D. (1994). Measuring the degree of internationalisation of a rm. Journal of International Business Studies, 25, 325342. Tallman, S., & Li, J. (1996). Effects of international diversity and product diversity on the performance of multinational rms. Academy of Management Journal, 39(1), 179196. Tesar, G., & Moini, A. H. (1998). Longitudinal study of exporters and non exporters: A focus on smaller manufacturing enterprises. International Business Review, 7(3), 291313. Venkataraman, S. (1997). The distinctive domain of entrepreneurship research. In: J. A. Katz (Ed.), Advances in entrepreneurship, rm emergence and growth (pp. 119138). Greenwich: JAI Press. Venkatraman, N., & Ramanujam, V. (1986). Measurement of business performance in strategy research: A comparison of approaches. Academy of Management Review, 11, 801814. Ward, P., Bickford, D. J., & Keong Leong, G. (1996). Conguration of manufacturing strategy, business strategy, environment and structure. Journal of Management, 22(4), 597626. Westhead, P., Wright, M., & Ucbasaran, D. (2001). The internationalization of new and small rms: A resource-based view. Journal of Business Venturing, 16, 333358. Wolff, J. A., & Pett, T. L. (2000). Internationalization of small rms: An examination of export competitive patterns, rm size, and export performance. Journal of Small Business Management, 4, 3447. Yin, R. (1994). Case study research: Design and methods (2nd ed.). Beverly Hills, CA: Sage. Young, S., Hamill, J., Wheeler, c., & Richard, D. (1989). International market entry and development. Englewood Cliffs, NJ: Prentice-Hall. Zucchella, A. (2005). Non sequential internationalisation processes between competition and co-operation: A tentative re-conciliation of alternative approaches. In: T. Morrow, S. Loane, J. Bell & C. Wheeler (Eds), International business in an enlarging Europe. Basingstoke: Palgrave.
FACTORS AFFECTING THE TIMING OF THE EXPORT DEVELOPMENT PROCESS IN SPANISH MANUFACTURING FIRMS
Aristides Olivares-Mesa and Sonia Suarez-Ortega
ABSTRACT
We study entry timing in the export development process of Spanish manufacturing rms. We interpret this process as a sequential path which allows us to identify the following export stages: (I) the pre-engagement phase, where rms do not export; (II) the initial phase, where rms export via an agent; and (III) the advanced phase, where rms export via a sales subsidiary. This study explores factors, which can accelerate or decelerate the decision to change phases. Data are taken from the Spanish Survey on Business Strategies that comprises 1,478 rms in 2002. Event history analysis is applied to our dataset. Obtaining product or process innovations is the most signicant motivation for an early entry in the initial and advanced phases of the export development process. Network ties, a broader scope of products, rm size and foreign ownership participation are also key factors in accelerating entries.
89
90
ARISTIDES OLIVARES-MESA AND SONIA SUAREZ-ORTEGA
1. INTRODUCTION
Internationalization is one aspect of strategic behavior that is considered of great importance regarding the growth of businesses. In this sense, many empirical models have been proposed to describe and explain the export development process of manufacturing rms (see e.g., Johanson & Wiedersheim-Paul, 1975; Wiedersheim-Paul, Olson, & Welch, 1978; Cavusgil, 1982; Hyvaerinen, 1994; Bell, 1995; Korhonen, Luostarinen, & Welch, 1996; Coviello & Munro, 1997; Fontes & Coombs, 1997; Gankema, Snuit, & van Dijken, 1997; Zafarullah, Ali, & Young, 1998). Although there are differences among the various models as to the number, nature and content of the stages, it can be concluded that the export development process comprises three broad phases: (a) the pre-engagement phase, which includes two types of rms. First, those selling their goods solely in the domestic market; that is, those rms involved in the domestic market with no interest in exporting, and second, those that exported in the past but no longer do so; (b) the initial phase, where rms are involved in sporadic export activity and can be classied as having the potential to increase their overseas involvement, and are unable to cope with the demands of exporting, leading to marginal export behavior or withdrawal from selling abroad altogether; and nally (c) the advanced phase, where rms are regularexporters with extensive overseas experience, and frequently consider more committed forms of international business (Leonidou & Katsikeas, 1996). Although researchers agree that the export development process is highly dynamic and time-dependent, paradoxically almost all models are static in nature (Leonidou & Katsikeas, 1996). Only Berra, Piatti, and Vitali (1995) and Gankema et al. (1997) specically analyze longitudinal patterns. This condition is a concern given that internationalization is dened to be a process occurring over time, and cross-sectional data ultimately limits the depth of our understanding of that process (Coviello & McAuley, 1999). Another time-related issue is the velocity at which the rm moves between stages. This factor has been ignored by almost all models, although variations in technological intensity, product life cycles, research and development costs, and other factors can affect a rms progress along the internationalization path (Young, 1987). Also, even though there are an increasing number of studies on rapid internationalization (see Rialp, Rialp, & Knight, 2005 for a literature review), it could be said that the impact of time has been overlooked (HurmerintaPeltomaki, 2003).
Factors Affecting the Timing of the Export Development Process
91
This paper investigates entry timing by Spanish manufacturing rms into foreign markets. Using the Survey on Business Strategies, which offers details on the rms situation at the moment of the survey but does not date relevant events, we analyze how different factors identied in the literature affect the transition between stages. We use event history analysis (also called survival analysis) in the research. This methodology allows us to take into account not only rms that have entered into foreign markets but also domestic rms. In this way, we avoid the bias produced when studying the speed of internationalization by looking exclusively to exporting rms and discarding information from domestic rms, as previous studies do (e.g., Bell, McNaughton, & Young, 2001; Aspelund & Moen, 2001; Zahra, Matherne, & Carleton, 2003). Survival analysis also allows us to overcome the difculties due to the lack of information about the age of transition from one phase to another. These technical problems are solved in the context of the doubly censored models of duration, as we will see further on. The paper is organized in the following way. In the next section we present a review of the literature on the process and the factors affecting internationalization according to a resource-based view of the rm. This approach provides the basis for stating several hypotheses about the inuence of these factors on the speed of the export development process. The methodology of the empirical study is then set out, followed by an exposure of the main results. We then highlight the studys contribution and limitations and, nally, draw some conclusions from the study.
2. FACTORS AFFECTING THE SPEED OF THE EXPORT DEVELOPMENT PROCESS

The Uppsala model and other stage approaches to internationalization are the basis for the idea of internationalization as a process of acquiring international knowledge and accumulating tangible and intangible resource stocks (Fernandez & Nieto, 2005; Westhead, Wright, & Ucbasaran, 2001). According to the resource-based view of the rm, the strategic decisions taken depend on the characteristics of the resources and capabilities controlled by the rm (Barney, 1991; Grant, 1991; Peteraf, 1993). Similarly, these resources and capabilities can inuence decisions about whether the rm internalizes its exports or not, the mode of internationalization chosen and the control of distribution channels abroad (Campa & Guillen, 1999). Foreign investment decisions by rms are partly inuenced by unique
92
competencies developed from their founding. These competencies emphasize the role of foreign market knowledge needed by the rm to successfully undertake international expansion. Concretely, a review of the literature can help identify several factors associated with the rms resources that can affect the speed of the internationalization process. These factors are described in the following paragraphs. Size. The relationship between rm size and export behavior has been one of the most widely analyzed relationships in international marketing literature. Although there is numerous research on the topic, little consensus exists concerning the actual impact of size on export behavior, the exception being the positive association with propensity to export and the negative one with export intensity (Bonaccorsi, 1992; Calof, 1994). Past research, however, has not examined the potential relationship between size and timing of the export development process. On the one hand, rm size is often an indicator of its nancial, physical, human and tangible resources. The large size of a rm reects its capacity to absorb the high costs and risks involved in internalizing international expansion (Buckley & Casson, 1976), while resource scarcity limits smaller rms ability to reach advanced stages of internationalization requiring foreign direct investment (e.g., Dunning, 1980, 1988). Hypothesis1a. The larger the rm, the earlier the rm will proceed to advanced stages of internationalization. On the other, from a resource-based view of the rm, smaller size typical of young rms appears to confer a sort of exibility that provides key benets for succeeding in foreign markets (Knight & Cavusgil, 2004). This success is seen despite the scarce resources typical of young rms. It should also be highlighted that entering into export activities usually occurs through the use of intermediaries in the domestic market (Leonidou & Katsikeas, 1996), and this mode of entry does not require a large amount of tangible and nancial resources. Thus, Hypothesis1b. Timing for the rm to enter into export activities is independent of rm size. Technological resources. From a resource-based view, high technological intensity, typically indicated by high R&D expenditures, provides the rm with unique technological know-how, which often promotes the expansion of the rm overseas (Dhanaraj & Beamish, 2003). R&D expenditures lead to
93
the creation of valuable rent-yielding intangible assets, such as knowledge related to new product and process technologies (Kogut & Chang, 1991). According to Hitt, Hoskisson, and Ireland (1994), rms will internationalize their operations to obtain higher returns on their innovations and reduce the risk of selling a product in a single market. Internationalization also safeguards rms against rapid imitation (Zahra, 1996). Thus, Hypothesis 2. The greater the amount of technological resources possessed by the rm, the earlier the rm will enter into export activities and proceed to advanced stages of internationalization. Network ties. Networks include the formal and informal relationships that develop among rms as they interact (Johanson & Mattsson, 1987). Network ties are resources that are rm specic and difcult to imitate, and have consequences along three dimensions (Burt, 1997): (1) the information that is available to the rm, as networks are a source of information about what goes on in the market, (2) its timing, as a particular piece if information could reach the rm earlier through its network ties, and (3) referrals, that imply that rms interests are represented in a positive light, at the right time, and in the right place. In accordance with Sharma and Blomstermo (2003), this may inuence the internationalization process of rms. In relation with the export development process, network members can help a rm by identifying areas where changes in its product are desirable, making foreign market penetration easier (Zahra et al., 2003). Network members can also help by identifying export potential markets, especially when they are established in foreign markets. In this respect, Bell (1995), in a cross-national study into the export behavior of small computer software rms in Finland, Ireland and Norway, found evidence that contacts with foreign suppliers to obtain hardware, local software distribution rights or production licenses led to export initiation. The author is convinced that these contacts accelerated the decision to export, and it is unclear whether exporting would have occurred without these relationships. Thus, Hypothesis 3. The greater the amount of network ties, the earlier the rm will enter into export activities and proceed to advanced stages of internationalization. Firms with foreign ownership participation are more likely to obtain technical, economic, personnel and information resources that are less available,
94
or even unavailable, to rms without such participation (Fernandez & Nieto, 2005; Rialp, Axinn, & Thach, 2002). These resources allow rms to take advantage of capital ow increases, and develop the internationalization process. Thus, Hypothesis 4. The greater the amount of foreign ownership participation in the rm, the earlier the rm will enter into export activities and proceed to advanced stages of internationalization. Product diversication can bring benets to the rm because it takes advantage of economies of scope in common resources (Grant, 2002). Following Gaba, Pan, and Ungson (2002), the impact of economies of scope on internalization timing comes from two sources. Firstly, a broader scope of products and services means a wider portfolio of offerings to choose from and, thus, a better chance of providing the right product to the foreign market. In this case, rms with a broader strategic scope are better prepared to handle the uncertainty about the types of product that are needed in the foreign market (Kerin, Varadarajan, & Peterson, 1992). Secondly, a broader scope of products enables the rm to develop synergy across different product sectors (Shaver, Mitchell, & Yeung, 1997). This synergy gives rise to both efciency and quality in product development, product line extension, production, distribution and market support (Lambkin, 1988; Green, Barclay, & Ryans, 1995; Reddy, Holak, & Bhat, 1994). Taken together, rms with a broader product scope have a greater likelihood of early entry into a foreign market. Thus, Hypothesis 5. The broader the scope of products, the earlier the rm will enter into export activities and proceed to advanced stages of internationalization. Human capital resources include training, experience, judgment, intelligence, relationships, and insight of individual managers and workers in a rm (Barney, 1991). Education, training and experience of employees determine the skills available to the rm (Grant, 2002). Several studies indicate a positive relationship between the main decision makers education and internationalization (Cavusgil & Naor, 1987; Simpson & Kujawa, 1974). The quality of the workforce can be positively related to entry into foreign markets if the goods for export are of higher quality. Also, if exporting requires new product design and other forms of technical assistance, a highly skilled staff can be considered a sign of quality, vertical
95
differentiation and a measure of the rms level of production skills (Manez, Rochina, & Sanchis, 2002). Thus, Hypothesis 6. The higher the educational, technical and professional qualication of employees, the earlier the rm will enter into export activities and proceed to advanced stages of internationalization.
3. METHODOLOGY
3.1. Database The database to investigate the timing of the export development process of Spanish manufacturing rms was obtained from the Survey on Business Strategies (henceforth SBS), which has been carried out annually by the Spanish Ministry of Science and Technology since 1990. We have used the SBS for the year 2002, the last data available. The rm must indicate if it has exported in 2002. This information allowed us to classify rms in two groups: exporters and non-exporters. Moreover, the rm is asked to indicate the means employed to access foreign markets in that year, according to one of the following ve options: (1) own channel agents or sales subsidiary; (2) parent company in a foreign country foreign-owned rms; (3) export intermediary located in Spain; (4) export collaborative agreement exporters association, sectorial agreement or export consortium; and (5) others specify. Given that the purpose of this paper is to analyze the export development process of the Spanish rms, we decided to exclude all foreign-owned rms, that is, those that exported through a parent company located in a foreign country (option 2 from the prior paragraph). Regarding entry modes, if a rm answers yes to option 1 (own channel), we consider it exports directly and is in the advanced phase; if it answers no to option 1 and yes to options 3 (export intermediary located in Spain) or 4 (export collaborative agreement), we consider it exports indirectly and is in the initial phase. This classication from our sample size of 1,478 rms breaks down into 602 non-exporter rms, 330 which export indirectly and 546 direct exporters. 3.2. Method of Analysis A sequential process with three phases is assumed: rms begin with a preengagement phase, where they do not export; followed by an initial phase,
96
where rms export indirectly; and ending with an advanced phase, where rms export directly. Two events are of interest to us: the rst time a rm exports indirectly, and the rst time a rm exports directly. The aim is to determine how the occurrence and timing of exports depend on several predictor variables (covariates). One possibility to analyze such data using conventional methods is to perform a logit analysis with a dichotomous-dependent variable: exporter or non-exporter. However, this analysis ignores information on the timing of exports. One solution to this problem is to make the dependent variable the length of time between the foundation of the rm and the time of the rst export (indirect or direct, depending on what we are studying), and then estimate a conventional linear regression model (let T be this dependent variable). But what do we do with the rms that are non-exporters when they are surveyed? Such cases are referred to as right censored. Or what do we do with the rms that are exporters, but we do not know the exact date of the rst export? Such cases are referred to as left censored. In both cases, we know the age of the rm (let t be the age of the rm), but we do not know the duration T. Survival analysis devises a procedure that combines the information of the right- and left-censored cases in a way that produces consistent estimates of the parameters of interest. In right- and left-censored cases, the time of censoring is the rms age, but they differ in how the information is incorporated into the likelihood function. For right-censored cases, such information is incorporated into the estimates using their survival function values at the rms age, that is, the probability that rst export will occur at some time beyond the rms age [P(T>t) S(t)]. Information for left-censored cases is incorporated using their cumulative distribution function values at the rms age, that is, the probability that rst export occurred at some time before the rms age [P(Tpt) F(t)]. Thus, the natural logarithm of the likelihood function is X X ln L ln Sti ln F ti right-censored left-censored where ti is the age of rm i. A specication of the distribution for the survival time is required. The Weibull distribution is a two-parameter distribution that is particularly exible and used in this study. The survival function for this distribution is Sti explti 1=s and F ti 1 Sti
97
Table 1.
Name
Name and Denition of Independent Variables.

Denition Number of employees Ratio of research and development expenditures to the sales of the rm Number of patents and utility models registered by the rm in Spain or abroad Dummy variable equal to 1 if the rm has obtained product or process innovations; 0 otherwise Ratio of import expenditures to the purchases of the rm Percentage of foreign capital participation in the rm Let pj be the ratio of sales of product j to total sales. This ratio is available for the ve core products. Then we use a Berry-Herndhal P5 2 pj index:DIVERSIF 1 P j1 2
5
Size of rm (EMPLOY) Technological resources Research and development intensity (R&DINT) Number of patents (PATENTS)
Innovations (INNOVAT) Network ties (IMPINT) Foreign ownership (FORCAP) Product diversication (DIVERSIF)
p j1 j
Human capital (HUMANCAP)
Ratio of employees with a university degree to total employees of the rm
where l is the scale parameter and s the shape parameter. In our model, independent variables affect the l parameter in the following way: l expb0 b0 X i where Xi is the vector of covariates associated with the ith rm and b the vector of coefcients associated with each independent variable. A positive (negative) coefcient implies that the covariate exercises a positive (negative) inuence on waiting time. Thus, a unit increase in the covariate is interpreted as a rm delaying (hastening) entry into a more advanced phase in the export development process. A denition of the independent variables based on the questions available in the SBS is given in Table 1. Once the survival function is specied, estimation proceeds by maximizing the log-likelihood for the censored data. We use the LIFEREG procedure from SAS 8.0 version in our analyses. b0 and s are referred to as INTERCEPT and SHAPE by the LIFEREG procedure, respectively.
98
4. RESULTS
Table 2 shows means and standard deviations of variables, for non-exporters, indirect exporters and direct exporters. Note that as commitment with international activities deepens, mean values of variables are greater. Table 3 shows correlations between the independent variables. The magnitudes of the correlations among variables are not large enough to pose a serious threat of multicollinearity. Table 2.
Variable EMPLOY R&DINT PATENTS INNOVAT IMPINT FORCAP DIVERSIF HUMANCAP Number of rms
a
Mean and Standard Deviationa of Variables.

Non-Exporters 55.44 (171.66) 0.36 (2.14) 0.06 (0.56) 0.20 2.86 (8.62) 3.11 (16.19) 0.15 (0.23) 7.75 (13.02) 602 Indirect Exporters 170.86 (364.29) 0.60 (1.84) 0.43 (3.83) 0.41 7.95 (12.21) 8.31 (26.18) 0.28 (0.27) 11.87 (14.81) 330 Direct Exporters 389.98 (828.32) 1.14 (2.76) 1.06 (6.10) 0.53 11.97 (13.53) 20.15 (38.80) 0.31 (0.27) 13.97 (13.71) 546
All rms 207.56 (567.80) 0.71 (2.36) 0.52 (4.19) 0.37 7.39 (12.15) 9.78 (28.80) 0.24 (0.27) 11.01 (13.96) 1,478
Standard deviation in brackets, except for dummy variables.
Table 3.
(1) (1) (2) (3) (4) (5) (6) (7) (8) EMPLOY R&DINT PATENTS INNOVAT IMPINT FORCAP DIVERSIF HUMANCAP 1.00 0.24 0.11 0.17 0.20 0.09 0.13 0.16
Correlation Matrix (All Firms).

(2) (3) (4) (5) (6) (7) (8)
1.00 0.24 0.24 0.14 0.02 0.06 0.26
1.00 0.10 0.06 0.01 0.04 0.10
1.00 0.14 0.01 0.17 0.16
1.00 0.05 0.11 0.16
1.00 0.07 0.04
1.00 0.13
1.00
99
Table 4 shows the results of our model estimates for entry timing in the initial and advance phases of the export development process. We have added a likelihood-ratio test of the null hypothesis that all the covariates have zero coefcients. This test is analogous to the F-test that is routinely reported for linear regression models. To calculate this statistic, we t a null model that includes no covariates. The statistic is then twice the positive difference in the two log-likelihoods. With eight degrees of freedom (the number of covariates excluded from the null model), the p-value is less than 0.001, so we reject the null hypothesis and conclude that at least one of the coefcients is nonzero. The results indicate that rm size has an impact on the speed of the export development process, especially in the advance phase (po0.01) where mustering resources is more necessary. These results support Hypothesis H1a, and reject Hypothesis H1b. In relation to Hypothesis H2, which relates early entry with the technological resources of the rm, we found that registering patents and utility Table 4. Factors Affecting Entry Timing. Event History Analysis: Weibull Distribution 2002.
Initial Phase Coefcient INTERCEPT EMPLOY (H1) R&DINT (H2) Technological resources PATENTS (H2) INNOVAT (H2) Network ties IMPINT (H3) Foreign ownership FORCAP (H4) Product diversication DIVERSIF (H5) Human resources HUMANCAP(H6) Shape parameter (standard error) Log-likelihood No-covariate log-likelihood Log-likelihood w2 Right censored observations Left censored observations Number of observations Size Note: Negative coefcients indicate early entry. * po0.1. po0.05. po0.01. w
2
Advanced Phase Coefcient w

2
+8.387 21.36 0.003 7.10 +0.019 0.04 0.668 4.68 3.711 9.22 0.156 8.93 5.88 0.034 6.503 8.05 0.009 0.28 6.004 (1.831) 787.9 966.95 358.10 602 876 1,478
+10.344 23.06 0.003 8.68 0.070 0.70 0.024 0.49 2.713 9.89 0.091 9.26 0.029 7.49 3.969 7.90 0.006 0.15 4.566 (1.256) 810.6 941.83 262.46 932 546 1,478
100
models accelerate entry timing only in the initial phase of the export development process, while obtaining product or process innovations are actions that accelerate entry timing in both phases. The most signicant variable in our study is INNOVAT. Hypotheses H3, H4 and H5 relate early entry with network ties, higher foreign ownership participation, and broader scope of products. Our results support these three hypotheses. Specically, we found that contact with foreign suppliers is an important factor for foreign market entry decisions, as shown by the signicance of the variable IMPINT (po0.01). Hypothesis H6, which relates early entry with human capital of the rm, is not supported. However, this does not necessarily mean that a higher quality labor force will not help accelerate entry into international markets. A higher qualied labor force can contribute to the skills available to the rm, although these skills may be better captured by other variables included in the analysis.
5. CONTRIBUTION AND LIMITATIONS

This paper deals with the timing stages of the export development process, an aspect that has been neglected in the literature. Timing as the dependent variable in a study of export behavior is an important contribution as the time concept has a crucial role in todays appraisal of internationalization research (Hurmerinta-Peltomaki, 2003). Event history analysis is the ap propriate methodological tool to approach this subject. This econometric technique has an important advantage; it allows us to measure the length of each stage of the export development process including rms that have not yet evolved to a particular stage. Prior research on rms internationalization has reviewed multinational rms that have already gone through the whole process. The results of this type of research have a bias problem, as rms that have not yet decided to internationalize are not included in the studies (e.g., Bell et al., 2001; Aspelund & Moen, 2001). This bias problem is especially relevant when the purpose is to analyze entry timing into foreign markets. Despite these signicant contributions, the paper has several limitations. The data were collected from a survey that was not carried out for the specic purpose of the paper. It does not allow us to use all the richness of the resource-based view of the rm. The operationalization of the construct can be improved by generating specic indicators. Thus, we use R&D spending as a proxy for tangible technological resources. Other potential
101
measures of this construct include a rms manufacturing facilities and equipment, and R&D facilities and equipment. In addition, we use import intensity to operationalize network ties, although asking directly about the relationships developed by the company could be more appropriate. Another limitation of the data concerns the lack of information about the age of transition between phases. This is an important technical problem that affects the estimation procedure and, as we saw, can be overcome by applying doubly censored models of duration. However, it has been acknowledged that left censoring can be a problem unless cases are negligibly small (Tuma & Hannan, 1979). Therefore, future surveys should include date of transitions, which would provide us with a more accurate estimation of the duration of the stages in the export development process.
6. CONCLUSIONS
We study the export development process of Spanish manufacturing rms and postulate that such a process follows a sequential path which can be divided into three general phases: the pre-engagement phase (when a rm does not export), the initial phase (a rm exports indirectly, exerting little or no control over the marketing of its product) and the advanced phase (a rm exports directly through sales branches or subsidiaries). This study explores the factors which can accelerate or decelerate the decision to switch from one phase to another. The results offer some evidence about the key factors affecting the speed of the internationalization process. In this sense, capacity to innovate, involving the introduction of new products or the improvement of a rms existing product range, plays a signicant role in accelerating entry timing in foreign markets. On the same lines, Basile (2001) also found that innovation capabilities are very important competitive factors and help explain heterogeneity in export behavior among Italian manufacturing rms in the 1990s. One such common observation is that internationalization is the process of increasing the accumulation of knowledge in markets and institutions abroad. Network ties that rms have may help them to go international by supplying information about clients and markets. Firms that operate in an international network may enjoy a learning advantage and nd it easier to go abroad than rms whose exchange partners are domestic rms. Our ndings point out the importance of network ties in the timing of the internationalization process of rms. We have measured network ties through import intensity, which is a factor not usually included in previous studies
102
even though international business transactions involve both exporters and importers. In our research, rms seem to start their internationalization by purchasing machinery, raw materials and components from abroad, and only then do they move on to the outward side. This result demonstrates that inward and outward internationalization are interrelated processes through which companies become involved, step-by-step, in more demanding product, operation and market strategies. In line with previous studies (e.g., Katsikeas, Piercy, & Leonidou, 1996), our results show that rm size has a statistically signicant association with early entry into both phases of the export development process. Therefore, this study gives support to the postulate that resource scarcity limits smaller rms capacity to reach foreign markets, while rejecting the idea that the smaller size typical of young rms appears to confer a sort of exibility that provides key benets for succeeding in foreign markets (Knight & Cavusgil, 2004, p. 125). A broader scope of products and services also provides the rm with greater opportunities to nd, in a wider portfolio of offerings, the right product for the foreign market. Finally, another early-entry incentive is foreign ownership participation, as rms with such participation are more likely to obtain technical, economic, personnel and information resources that give rise to an advantage that the rm can use to accelerate the export development process. In summary, this study highlights the resources of the rm that lead to early internationalization. The most signicant resources are product innovation and network ties. The ability to develop new products and formal and informal relationships is an important managerial skill that should be developed by the manager in ways that give the rm a competitive advantage.
REFERENCES
Aspelund, A., & Moen, O. (2001). A generation perspective on small rms internationalization from traditional exporters and exible specialist to born globals. In: C. N. Axinn & P. Matthyssens (Eds), Reassessing the internationalization of the rm. Advances in International Marketing (Vol. 11, pp. 197225). Amsterdam: JAI. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99120. Basile, R. (2001). Export behaviour of Italian manufacturing rms over the nineties: The role of innovations. Research Policy, 30(8), 11851201. Bell, J. (1995). The internationalization of small computer software rms a further challenge to Stage theories. European Journal of Marketing, 29(8), 6075.
103
Bell, J., McNaughton, R., & Young, S. (2001). Born-again global rms. An extension to the born global phenomenon. Journal of International Management, 7, 173189. Berra, L., Piatti, L., & Vitali, G. (1995). The internationalization process in the small and medium-sized rm: A case study of the Italian clothing industry. Small Business Economics, 7(1), 6775. Bonaccorsi, A. (1992). On the relationship between rm size and export intensity. Journal of International Business Studies, 23, 305636. Buckley, P. J., & Casson, M. C. (1976). The future of multinational enterprise. New York: Holmer & Meier Publisher. Burt, R. S. (1997). The contingent value of social capital. Administrative Science Quarterly, 42, 339365. Calof, J. L. (1994). The relationship between rm size and export behavior revisited. Journal of International Business Studies, 25, 367387. Campa, J. M., & Guillen, M. F. (1999). The internationalization of exports: Firm- and locationspecic factors in a middle-income country. Management Science, 45(11), 14631478. Cavusgil, S. T. (1982). Some observations on the relevance of critical variables for internationalization stages. In: M. R. Czinkota & G. Tesar (Eds), Export management: An international context (pp. 276285). New York: Praeger. Cavusgil, S. T., & Naor, J. (1987). Firm management characteristics as discriminators of export marketing activities. Journal of Business Research, 15(3), 221235. Coviello, N., & McAuley, A. (1999). Internationalisation processes and the smaller rm: A review of contemporary empirical research. Management International Review, 39(3), 223256. Coviello, N. E., & Munro, H. J. (1997). Network relationships and the internationalization process of small software rms. International Business Review, 6(4), 361386. Dhanaraj, C., & Beamish, P. W. (2003). A resource-based approach to the study of export performance. Journal of Small Business Management, 41, 242261. Dunning, J. H. (1980). Towards an eclectic theory of international production: Some empirical tests. Journal of International Business Studies, 11, 931. Dunning, J. H. (1988). The eclectic paradigm of international production: A restatement and some possible extensions. Journal of International Business Studies, 19, 931. Fernandez, Z., & Nieto, M. J. (2005). Internationalization strategy of small and medium-sized family businesses: Some inuential factors. Family Business Review, XVIII(1), 7789. Fontes, M., & Coombs, R. (1997). The coincidence of technology and market objectives in the internationalisation of new technology-based rms. International Small Business Journal, 15(4), 1435. Gaba, V., Pan, Y., & Ungson, G. R. (2002). Timing of entry in international market: An empirical study of U.S. fortune 500 rms in China. Journal of International Business Studies, 33(1), 3955. Gankema, H. G. J., Snuit, H. R., & van Dijken, K. A. (1997). The internationalisation process of small and medium sized enterprises: An evaluation of the stage theory. In: R. Donkels & A. Miettinen (Eds), Entrepreneurship and SME research: On its way to the next millennium (pp. 185197). Aldershot: Ashgate Publishing Ltd. Grant, R. M. (1991). The resource-based theory of competitive advantage: Implications for strategy formulation. California Management Review, 33(Spring), 114135. Grant, R. M. (2002). Contemporary strategy analysis. Massachusetts: Blackwell. Green, D. H., Barclay, D. W., & Ryans, A. B. (1995). Entry strategy and long term performance: Conceptualization and empirical evidence. Journal of Marketing, 59, 116.
104
Hitt, M. A., Hoskisson, R. E., & Ireland, R. D. (1994). A mid-range theory of the interactive effects of international and product diversication on innovation and performance. Journal of Management, 20, 297326. Hurmerinta-Peltomaki, L. (2003). Time and internationalisation. Theoretical challenges set by rapid internationalisation. Journal of International Entrepreneurship, 1, 217236. Hyvaerinen, L. (1994). Internationalization of Finnish SMEs: Commitment, internationalization paths and innovation in SMEs. In: J. M. Veciana (Ed.), Internationalization, networks and strategy (pp. 76100). Aldershot: Ashgate Publishing Ltd. Johanson, J., & Mattsson, L. G. (1987). Interorganizational relations in industrial systems: A network approach compared with the transaction-cost approach. International Studies of Management and Organization, 17, 3448. Johanson, J., & Wiedersheim-Paul, F. (1975). The internationalization of the rm: Four Swedish cases. Journal of Management Studies, 12(October), 305322. Katsikeas, C., Piercy, N. F., & Leonidou, L. C. (1996). Determinants of export performance in a European context. European Journal of Marketing, 30(6), 635. Kerin, R. A., Varadarajan, P. R., & Peterson, R. A. (1992). First mover advantage: A synthesis, conceptual framework, and research propositions. Journal of Marketing, 56, 3352. Knight, G. A., & Cavusgil, S. T. (2004). Innovation, organizational capabilities, and the bornglobal rm. Journal of International Business Studies, 35, 124141. Kogut, B., & Chang, S. J. (1991). Technological capabilities and Japanese foreign direct investment in the United States. Review of Economics and Statistics, 73, 401413. Korhonen, H., Luostarinen, R., & Welch, L. (1996). Internationalization of SMEs: Inward outward patterns and government policy. Management International Review, 36(4), 315329. Lambkin, M. (1988). Order of entry and performance in new market. Strategic Management Journal, 9, 127140. Leonidou, L. C., & Katsikeas, C. S. (1996). The export development process: An integrative review of empirical models. Journal of International Business, 3, 517551. Manez, J. A., Rochina, M. E., & Sanchis, J. A. (2002). Hysteresis in the Spanish manufacturing exports. Working paper. University of Valencia, Spain. Peteraf, M. A. (1993). The cornerstones of competitive advantage: A resource-based view. Strategic Management Journal, 14, 179191. Reddy, S. K., Holak, S. L., & Bhat, S. (1994). To extend or not to extend: Success determinants of line extensions. Journal of Marketing Research, 31, 243262. Rialp, A., Axinn, C., & Thach, S. (2002). Exploring channel internalization among Spanish exporters. International Marketing Review, 19(2), 133155. Rialp, A., Rialp, J., & Knight, G. (2005). The phenomenon of early internationalizing rms: What do we know after a decade (19932003) of scientic inquiry? International Business Review, 14(2), 147166. Sharma, D. D., & Blomstermo, A. (2003). The internationalization process of Born Globals: A network view. International Business Review, 12, 739753. Shaver, J. M., Mitchell, W., & Yeung, B. (1997). The effect of own-rm and other-rm experience on foreign direct investment survival in the United States, 19871992. Strategic Management Journal, 18, 811824. Simpson, C. L., & Kujawa, D. (1974). The export decision process: An empirical enquiry. Journal of International Business Studies, 5(1), 107117.
105
Tuma, N. B., & Hannan, M. T. (1979). Approaches to the censoring problem in analysis of event histories. In: K. F. Schuessler (Ed.), Sociological methodology (pp. 209240). San Francisco: Josey Bass. Westhead, P., Wright, M., & Ucbasaran, D. (2001). The internationalization of new and small rms: A resource-based view. Journal of Business Venturing, 16, 333358. Wiedersheim-Paul, F., Olson, H. C., & Welch, L. S. (1978). Pre-export activity: The rst step in internationalisation. Journal of International Business Studies, 9(1), 4758. Young, S. (1987). Business strategy and the internationalisation of business: Recent approaches. Managerial and Decision Economics, 8, 3140. Zafarullah, M., Ali, M., & Young, S. (1998). The internationalization of the small rm in developing countries: Exploratory research from Pakistan. Journal of Global Marketing, 11(3), 2140. Zahra, S. (1996). Technology strategy and performance: A study of corporate-sponsored and independent biotechnology ventures. Journal of Business Venturing, 11, 289321. Zahra, S., Matherne, B. P., & Carleton, J. M. (2003). Technological resource leveraging and internationalisation of new ventures. Journal of International Entrepreneurship, 1, 163186.
THE ROLE OF MARKET AND LEARNING ORIENTATIONS IN RELATIONSHIP QUALITY: EVIDENCE FROM VIETNAMESE EXPORTERS AND THEIR FOREIGN IMPORTERS
Trang T. M. Nguyen, Nigel J. Barrett and Tho D. Nguyen
ABSTRACT
This study examines the roles of market and learning orientations in relationship quality between exporters in transition economies and their foreign importers and subsequently, export performance. A random sample of 283 export rms in Vietnam provides evidence to support the hypothesized main effects. The results further indicate that learning orientation plays a role in building high-quality relationships for both new and mature relationships. However, the impact of market orientation on relationship quality is found only in the new relationship. In addition,
107
108
TRANG T. M. NGUYEN ET AL.
rm-ownership structure does not moderate the relationships between learning orientation, market orientation, relationship quality, and export performance.
INTRODUCTION
It is impossible for rms not to have relationships with other rms as they cannot operate in isolation (e.g., Hakansson, 1982). Research has shown that rms competitive advantage can be created through keeping long-term relationships with customers and that many rms are moving from transactional discrete exchanges to relational ones (Ganesan, 1994). This is because loyal customers will bring more prots to rms than the price-sensitive and deal-prone switcher (Reicheld, 1996). In addition, committed relationships are among the most durable advantages because they are difcult for competitors to understand, to copy, or to displace (Day, 2000; Srivastava, Fahey, & Christensen, 2001). Consequently, the role of market relationships has emerged as a top priority for most business rms around the world. The emergence of this trend indicates that rms realize the importance of relationship quality between them and their customers. In an export setting, unlike relationships in domestic markets, relationships developed with partners in international markets are more complex due to the difference in cultural, economic, and other environmental factors (Lages, Lages, & Lages, 2005). Therefore, understanding how to initiate, develop, and maintain highquality buyerseller relationships in international markets is critical to successful export involvement (Styles & Ambler, 1994). As a result, it has been recommended by several researchers that additional research is required (e.g., Bigne & Blesa, 2003; Frazier, 1999; Sanzo, Santos, Vazquez, & Alvarez, 2003; Simpson, Siguaw, & Baker, 2001). Research on inter-rm relationships in export markets, particularly research on relationship quality, however, has largely focused on the developed world, e.g., North American and European economies (Leonidou & Kaleka, 1998). Little research has been undertaken in developing and transition economies like Vietnam, one of Asias fastest growing and export-led economy. In 2005, Vietnams total value of exports surged to US$ 31.5 billion from US$ 2.4 billion in 1990 (Saigon Tiepthi, 2006). The export sector accounts for nearly 50% of the countrys gross domestic product in 2004 and plays a key role in Vietnams economic growth, which has remained at about 7% in recent years (Field, 2006). Owing to nearly becoming
Role of Market and Learning Orientations in Relationship Quality
109
a member of the World Trade Organization, Vietnam expects that its exports will grow rapidly in 2006. The trade ministry of Vietnam has projected that Vietnams exports will increase US$ 38 billion in 2006. Recently, the trade ministry has decided to use US$ 60 million for export promotion programs and Vietnams export markets have expanded to 127 countries in the world (Hieu Long, 2006). Therefore, research on export relationships in Vietnam will enhance our understanding of business-to-business relationships in such an under-investigated economy (Tsang, 2005). Further, although market and learning orientations have been widely studied (e.g., Slater & Narver, 1995; Sinkula, Baker, & Noordewier, 1997), little attention has been paid to the role of these two organizational factors in the quality of business-to-business relationships. In addition, most previous research examines the competitive value of market and learning orientations under the lens of the operating environment condition. Drawing on the resource-based view of the rm (Wernerfelt, 1984), this study investigates the role of market and learning orientations through internal capabilities of the rm. Specically, it views market and learning orientations as forms of organizational culture and explores their impacts on the quality of business relationships between exporters in developing economies and their foreign importers, and subsequently export performance. The rest of the article is organized as follows: rst, we review the literature and propose the hypotheses. Subsequently, we present the method, data analysis, and the results. We conclude the article by discussing a number of implications, limitations, and directions for future research.
LITERATURE REVIEW AND HYPOTHESES

Fig. 1 depicts a conceptual model that presents the role of market and learning orientations in building high-quality business relationships. Learning orientation and market orientation are expected to have direct impacts on relationship quality. Learning orientation is also hypothesized to be an antecedent of market orientation. Finally, export performance is the outcome of relationship quality. Relationship Quality Relationship quality is an important aspect in maintaining and evaluating buyerseller relationships. It can be dened as an overall assessment of the strength of a relationship and the extent to which it meets the needs or
110
Market Orientation
H1 H4
H3
Relationship Quality
Export Performance
Learning Orientation
H2
Fig. 1.
A Conceptual Model.
expectations of the parties based on a history of successful or unsuccessful events (Smith, 1998, p. 78). Several conceptualizations of relationship quality have been proposed, such as trust, commitment, and satisfaction (e.g., Smith, 1998), willingness to invest, conict, expectation of continuity (e.g., Kumar, Scheer, & Steenkamp, 1995), and minimal opportunism (e.g., Dwyer & Oh, 1987). However, trust, commitment, minimal opportunism, and satisfaction are widely accepted as the dimensions of relationship quality because they have received strong empirical support (e.g., Dorsch, Swanson, & Kelly, 1998; Dwyer & Oh, 1987; Nguyen, Barrett, & Nguyen, 2004). Trust can be dened as the willingness of an export rm to be vulnerable to the actions of another party based on the expectations that the other party will behave in a right (good) way toward the rm (Kumar et al., 1995). This denition of trust reects two essential dimensions honesty and benevolence. Honesty is based on the belief that the rm stands by its words (Anderson & Narus, 1990), fullls promised role obligations, and is sincere (Dwyer & Oh, 1987). Benevolence is the belief that the rm is interested in its partners welfare, and will not take unexpected actions that would have a negative impact on the partner (e.g., Anderson & Narus, 1990). Trust plays a central role in inter-rm relationships and is essential for the development of enduring partnerships (Morgan & Hunt, 1994) because it facilitates constructive dialogue and cooperative problem-solving, enabling the rm not only to work together with other rms more effectively but also to reduce perceived uncertainty and complexity in the future. Commitment is another dimension of relationship quality. It is central to successful relationship marketing because it enables independent channel members to work together to serve customers better and to achieve a higher
111
level of performance (Morgan & Hunt, 1994). Commitment to a relationship entails a desire to develop a stable relationship, and a willingness to sacrice short-term benets to maintain stability in a long-term relationship (Anderson & Weitz, 1992). Three main types of commitment have been found in the literature, i.e., continuance, behavioral, and affective (e.g., Kim & Frazier, 1997). Continuance commitment is reected in the stability of a relationship and is dened as an exporters desire to continue the relationship with foreign customers, which brings the exchange partners closer together (Anderson & Weitz, 1992). Any rms interested in ongoing exchange relationships with their customers would have a propensity for relation continuity. Behavioral commitment is reected in the actual behavior of an exporter toward an importer. It is dened as the extent to which an exporter provides special assistance to its foreign customer in times of need. High behavioral commitment is indicated by the exporters expressive behavior that shows it cares about the importer. The exporter not only performs its pre-agreed roles but also provides extra help for its foreign customer as required under various situations. Behavioral commitment is realized in the concrete behavior through which the partners become committed. Gundlach, Achrol, and Mentzer (1995) note that remaining in a relationship and compliance with contractual stipulations indicate behavioral commitment. Affective commitment is the attitudinal aspect of an exporters business ties to its foreign customer. Affective commitment refers to the sense of unity binding an exporter to its importer (Kim & Frazier, 1997). Highly affective commitment means that the exporter feels a strong unity of interests and goals with the importer and can work with the importer in harmony (Anderson & Narus, 1990). The third dimension of relationship quality is satisfaction. It can be dened as the extent of a partners overall affective evaluation of the relationship (Anderson & Narus, 1990). Satisfaction is also a key aspect of successful buyerseller relationships as it motivates satised parties to commit more to benecial exchange relationships (Leuthesser, 1997). Satisfaction is considered to be an indicator of how a rm assesses some of the other costs and benets of its relationship beyond economic performance and conict levels (Cullen, Johnson, & Sakano, 1995). The nal dimension of relationship quality is minimal opportunism. Opportunism is dened in general terms, as self-interest seeking with guile (Williamson, 1975, p. 6). Opportunistic behavior refers to the taking of unexpected actions that will generate negative outcomes for a rm that is involved in a transaction or relationship. Successful business relationships provide enhanced efciencies for both buyers and sellers (Kalwani &
112
Narayandas, 1995). In spite of such benets, it has been revealed that business relationships between buyers and sellers have a high failure rate (Parkhe, 1993). Though divergent long-term goals can be used for explaining these failures, a primary reason for the failure of many alliances is the inability to check opportunism by the alliance partners (Bucklin & Sengupta, 1993, p. 33). This means that opportunistic behavior decreases the possibility of achievement of common goals. The consequences of opportunistic behavior include the failure of exchange partners to fulll promises and obligations, and the possible termination of the relationship. In an international context, an exporter works with its foreign importers in far distance in terms of geography and culture, allowing international channel members to easily engage in opportunistic behavior (Cavusgil, Deligonul, & Zhang, 2004). Therefore, the development of satisfying relational exchange also requires minimal opportunism (Dwyer & Oh, 1987). Relationship quality is considered to be the essence of relationship marketing (Jap, Manolis, & Weitz, 1999) and serves as an indicator of the health and future well-being of long-term relationships (Crosby, Evans, & Cowles, 1990). Accordingly, several researchers have attempted to investigate possible predictors of relationship quality. For example, a salespersons expertise and relational selling behavior have been found to have positive impacts on relationship quality between salespersons and customers in the life insurance industry (Crosby et al., 1990). Dwyer and Oh (1987) found that the quality of channel relationships is affected adversely by the degree of partners bureaucratization. Procedural fairness has been examined to have a positive impact on relationship quality, while environmental uncertainty has a negative effect on relationship quality (Kumar et al., 1995). Smith (1998) found that relational bonds have a positive effect on relationship quality. Following this stream of research, this study examines the impact of two key organizational factors market and learning orientations on relationship quality. Market and Learning Orientations The resource-based view of the rm (Wernerfelt, 1984) posits that when a rm owns valuable, rare, inimitable, and non-substitutable resources and is able to implement value-creating strategies that cannot be easily duplicated by competitors, the rm can achieve sustainable competitive advantage (e.g., Barney, 1991; Eisenhardt & Martin, 2000). The valuable, rare, inimitable, and non-substitutable attributes of resources form the basis of unique valuecreating strategies (Eisenhardt & Martin, 2000; Teece, Pisano, & Shuen, 1997). In this study, both market and learning orientations are viewed as
113
forms of organizational culture and are rm-level resources, which are sources of sustainable competitive advantage (Barney, 1986; Hult, Hurley, & Knight, 2004; Menguc & Auh, 2006; Slater & Narver, 1995). Market-oriented rms continuously collect information about target customers needs and competitors capabilities, and then utilize this information to continuously create superior customer value (Slater & Narver, 1995). Narver and Slater (1990, p. 21) dene market orientation as the organizational culture that most effectively and efciently creates the necessary behaviors for the creation of a superior value for buyers and, thus, continuous superior performance for the business. These researchers propose three behavioral dimensions: customer orientation, competitor orientation, and, inter-functional coordination. Each of these dimensions plays its role in intelligence generation, dissemination, and responsiveness to the collected information. Together with market orientation, learning orientation has also been considered as another organizational culture that assists a rm in achieving competitive advantage because it plays a key role in the creation of knowledge (e.g., Hurley & Hult, 1998; Sinkula, 1994; Slater & Narver, 1995). Learning orientation refers to organization-wide activity of creating and using knowledge to enhance competitiveness (Calantone, Cavusgil, & Zhao, 2002, p. 516). Learning-oriented rms are better equipped to manage their organizational knowledge, than rms which have not yet learned to learn. Sinkula et al. (1997) argue that learning orientation causes a set of organizational values that inuence the propensity of rms to create and use knowledge. Firms with strong learning orientation encourage, or even require, employees to constantly question the organizational norms that guide organization actions (Sinkula, 1994; Sinkula et al., 1997; Baker & Sinkula, 1999). Learning orientation is comprised of three dimensions commitment to learning, shared vision, and open-mindedness, which direct a rm to create and encourage a learning environment throughout the rm (Sinkula et al., 1997). The rm continuously promotes the organizational learning process, that is, information acquisition, information dissemination, and shared interpretation (Sinkula, 1994). The rm endlessly creates and uses new knowledge about customers and competitors that has the potential to inuence the rms performance (Emden, Yaprak, & Cavusgil, 2005; Sinkula et al., 1997). Therefore, both market and learning orientations are related to specic and routine processes that create superior values to customers, and thus, assist a rm in gaining sustainable competitive advantage (Baker & Sinkula, 1999; Celuch, Kasouf, & Peruvemba, 2002; Hult et al., 2004; Slater & Narver, 1995; Yeniyurt, Cavusgil, & Hult, 2005).
114
In export markets, sustainable competitive advantage created by market and learning orientations can be obtained through exporters capabilities of building high-quality relationships with their foreign importers. Both orientations are related to customer interface behaviors (Celuch et al., 2002) and assist exporters to achieve competencies in foreign market knowledge (Yeniyurt et al., 2005). These capabilities will stimulate exporters to join efforts with their partners in order to achieve mutual and individual goals successfully, which discourage opportunistic behaviors (Stern & Reve, 1980). Market- and learning-oriented exporters are willing to understand the needs and wants of their partners, and to satisfy these needs more effectively and efciently than their competitors. These exporters value the benets of exporterimporter relationships and will seek to put their partners needs as a priority in the organizational concerns (Emden et al., 2005; Deshpande, Farley, & Webster, 1993; Hult, Nichols, Giunipero, & Hurley, 2000; Kalwani & Narayandas, 1995; Slater & Narver, 1995; Siguaw, Simpson, & Baker, 1998). In addition, learning orientation can foster market-oriented behavior in an organization (Jaworski & Kohli, 1996). A key component of learning orientation is a rms ability to engage in adaptive as well as generative learning (Slater & Narver, 1995). This enables the rm to acquire, process, and subsequently use market intelligence, which is reected in its market orientation (Jaworski & Kohli, 1996). In addition, the learning-oriented rm is more likely to leverage the use of all resources, including the behaviors that accompany a market orientation (Baker & Sinkula, 1999). Thus, H1. A positive relationship between market orientation and relationship quality is expected. H2. A positive relationship between learning orientation and relationship quality is expected. H3. A positive relationship between learning orientation and market orientation is expected. Relationship Quality and Export Performance Performance in international marketing channels is dened as the accomplishments real and perceived that have resulted from the manufacturer distributor relationship (Rosson & Ford, 1982, p. 61). A number of performance measures have been proposed in the literature. There are two approaches, objective and subjective, which are used to measure export performance. The objective approach is based on nancial indicators, which
115
include such measures as return on investment, prots, margins, sales, sales growth, and asset turnover. The subjective approach is employed to measure a rms absolute and relative performance on tasks, such as market share and new product development. This study focuses largely on the rms performance in dealing with its specic partner and, therefore, employed the latter approach (Madsen, 1998). Moreover, Geringer and Hebert (1991) found that objective and subjective measures of performance correlate highly. A high level of relationship quality is likely to have positive consequences for the relationship. The benets of a high level of relationship quality should translate into the economic performance for both partners. A highquality relationship leads to efcient transactions, such as shortened response time, advantages in logistics management, and marketing programs that contribute to the rms efciency and effectiveness in serving its market. These, in turn, can create a strong market position, which will be reected in the rms performance. When conditions required for a high-quality relationship are met, the exporter or its foreign importer (or both) is more likely to be attracted to an existing relationship, and such relationships can be expected to continue in the future. As a result, agreement on such matters as decision-making, and mutual dependence should increase, moving the relationship closer to long-term partnership. Research has shown that longterm partnerships lead to increased mutual protability (Anderson & Weitz, 1992) and enhance the performance outcomes in buyerseller relationships (Noorderwier, John, & Nevin, 1990). Likewise, Kalwani and Narayandas (1995, p. 14) found that maintaining close relationships with customers in the long-run lead to high protability through better understanding and servicing of customer needs. Suppliers who develop better relationships with their foreign customers are likely to enjoy superior performance in terms of ultimate outcomes such as sales and share of customer business (Leuthesser & Kohli, 1995). Therefore, H4. A positive relationship between relationship quality and export performance is expected.
METHOD
The Sample A systematic sample of 283 Vietnamese export rms in Ho Chi Minh City, the major business center of Vietnam, was surveyed. The sampling frame,
116
based on the Business Directories in Ho Chi Minh City, consisted of about 5,000 rms in all industries. The single-key informant approach, the most commonly used method in organizational research (Kumar, Stern, & Anderson, 1993), was employed in this study. Respondents were export managers of the rms. Partial self-administered surveys, in which questionnaires are mailed to the target respondents and are collected by interviewers, were used for this study. Follow-up telephone calls to remind respondents to complete the questionnaires prior to collection were also utilized. Three hundred and twenty completed questionnaires were collected from 400 distributed questionnaires, yielding a response rate of 80%. Among 320 completed questionnaires, 37 were found invalid because these respondents were not members of management in charge of export activities. Consequently, the remaining 283 valid completed questionnaires comprised the sample for this research. The sample was comprised of 145 (51.2%) stateowned rms and 138 (48.8%) rms with other types of ownership (joinstock, limited-proprietary, and private-owned rms). In terms of rm age, 173 (61.1%) rms were in business less than 10 years and 93 (30.8%) rms were in business from 11 to 20 years. Only 17 (6.1%) rms were in business more than 30 years. In terms of relationship duration, 166 (58.7%) rms had business relationships with their partners less than ve years and 117 (41.3%) rms had business relationships with their business partners more than ve years. Using w2 tests with respect to rm ownership and rm age we found that no signicant difference between the population percentages and the sample percentages. This suggests that the sample and population proles (based on these two key variables) are not signicantly different. Measurements Learning orientation was measured based on Sinkula et al.s (1997) scale. It was a second-order construct consisting of three dimensions, i.e., commitment to learning (measured by four items), shared vision (four items), and open-mindedness (three items). Market orientation was also a second-order construct and was measured using Narver and Slaters (1990) scale. It comprised three dimensions: customer orientation (seven items), competitor orientation (four items), and inter-functional coordination (ve items). Relationship quality was a high-order construct comprising two second-order constructs (trust and commitment), and two rst-order constructs (satisfaction and minimal opportunism). Trust had two dimensions: honesty and benevolence. Honesty was measured by 12 items and benevolence was measured by 10 items. These measures were developed by Kumar et al.
117
(1995). Commitment was measured by using Kim and Fraziers (1997) scale, and was comprised of three dimensions: continuance commitment (six items), behavioral commitment (10 items), and, affective commitment (seven items). Satisfaction was measured based on the scale (seven items) developed by Gaski and Nevin (1985). Minimal opportunism was measured using Lees (1998) scale (13 items). Finally, export performance was measured by seven items. This scale was based on Raven, McCullough, and Tansuhaj (1994) scale with some modications. The modications were based on the nancial and strategic dimensions of the scale developed by Zou, Taylor, and Osland (1998). The questionnaire was initially prepared in English and then translated into Vietnamese by an academic who is uent in both languages. Back translation was undertaken to ensure the equivalence of meanings.
DATA ANALYSIS AND RESULTS

Measurement Validation All measures used were rst rened via Cronbachs alpha and then tested by conrmatory factor analysis (CFA). The Cronbachs alphas of all measures of the rst-order constructs and the dimensions of the second-order constructs were high (X.86) (Table 1). The screening process shows that the data exhibited slight deviations from multinormality, however, all univariate kurtoses and skewnesses were within the range of [1, 1]. Therefore, the maximum likelihood estimation method was used (Muthen & Kaplan, 1985). As discussed previously, relationship quality was a high-order construct comprising four dimensions: trust, commitment, satisfaction, and minimal opportunism. Trust and commitment were second-order constructs, and satisfaction and minimal opportunism were rst-order constructs. The CFA results indicate that these measurement models of these constructs t the data well. Market and learning orientation were also second-order constructs, and export performance was a rst-order construct. The CFA results also indicate that their measurement models t the data well (Table 1). In addition, all factor loadings were high and substantial (the lowest loading was .59) (see the appendix for the standardized factor loadings). These results indicate that the measures of all rst-order constructs and all dimensions of second-order constructs were unidimensional and their convergent validity was achieved (Steenkamp & van Trijp, 1991).
118
Table 1.
Constructs Market orientation: 96.02 (p 0.23); CFI 0.997; GFI 0.956; RMSEA 0.019 Learning orientation: w2 50.06 (41) (p 0.15); CFI 0.996; GFI 0.968; RMSEA 0.028 Trust: w2 (208) 238.05 (p 0.075); CFI 0.991; GFI 0.929; RMSEA 0.023 Commitment: w2 (227) 270.83 (p 0.024); CFI 0.989; GFI 0.921; RMSEA 0.026 w2 (87)
Summary of Scale Validation.

Correlations mcs mcm mco lcm lsv lop hon ! ! ! ! ! ! ! mcm mco mcs lsv lop lcm ben r(se) 0.53(.077) 0.74(.091) 0.65(.085) 0.71(.086) 0.58(.083) 0.63(084) 0.74(.101) a 0.94 0.89 0.87 0.92 0.91 0.86 0.94 0.89 0.93 0.88 0.91 0.96 0.96 0.93 rc 0.94 0.89 0.87 0.92 0.92 0.86 0.94 0.89 0.93 0.88 0.91 0.96 0.96 0.94 rvc 0.70 0.68 0.64 0.75 0.73 0.67 0.56 0.45 0.70 0.43 0.60 0.67 0.78 0.62
Dimensions Customer orientation (mcs) Competitor orientation (mcm) Interfunctional coordination (mco) Commitment to learning (lcm) Shared vision (lsv) Open mindedness (lop) Honesty (hon) Benevolence (ben) Continuance (con) Behavioral (beh) Affective (aff)
con con aff
! ! !
beh aff beh
0.60(.089) 0.79(.097) 0.68(.095)
Minimal opportunism: w2 77.62 (p 0.135); CFI 0.996; GFI 0.960; RMSEA 0.026 (65) Satisfaction: w2 23.16 (p 0.057); CFI 0.996; GFI 0.976; RMSEA 0.048 (14) 2 Export performance: w(27) 33.09 (p 0.194); CFI 0.996; GFI 0.974; RMSEA 0.028
Note: r(se), correlation (with standard error); a, Cronbachs alpha; rc, composite reliability; rvc, average variance extracted.
119
The correlations (with standard errors) between the dimensions of the second-order constructs indicate that they were signicantly different from unity. Therefore, the within-construct discriminant validity was achieved (Steenkamp & van Trijp, 1991) (Table 2). Because the measures of all rstorder constructs and the dimensions of all second-order constructs were unidimensional, summates were used to the measurement model of relationship quality and the nal measurement model. It is noted that two summates were used for minimal opportunism and satisfaction, and three summates was used for export performance. These summates were formed by randomly summing the indicators in each scale into groups, and each group was represented by one indicator. The use of two and three summates (instead of one) makes the model identied without using additional constraints and is referred to as partial disaggregation (Bagozzi & Edwards, 1998). The CFA results of the measurement model of relationship quality t the data well. The correlations (with standard errors) between the dimensions of relationship quality indicate that they were signicantly different from unity. Therefore, the discriminant validity between the dimensions of relationship quality was also achieved. Finally, the nal measurement model also received a good t to the data. The correlations between constructs together with their standard errors indicate that they were signicantly different from unity. These ndings support the across-construct discriminant validity (Steenkamp & van Trijp, 1991) (Table 2). Table 2. Correlations between Constructs.
Measurement Models Relationship quality: w2 28.07 (p 0.138); (21) CFI 0.996; GFI 0.978; RMSEA 0.035 Trust Commitment Minimal opportunism Satisfaction Satisfaction Satisfaction Market orientation Learning orientation Learning orientation Relationship quality Market orientation Market orientation Correlation ! ! ! ! ! ! ! ! ! ! ! ! Commitment Minimal opportunism Trust Trust Minimal opportunism Commitment Learning orientation Relationship quality Export performance Export performance Export performance Relationship quality r(se) 0.80(0.091) 0.50(0.073) 0.51(0.078) 0.68(0.082) 0.52(0.071) 0.77(0.080) 0.54(0.082) 0.35(0.076) 0.22(0.070) 0.58(0.079) 0.30(0.069) 0.38(0.074)
Final measurement model: w2 (125) 176.61 (p 0.002); CFI 0.986; GFI 0.936; RMSEA 0.038
Note: r(se), correlation (with standard error).
120
Test of Hypotheses Structural Results The SEM results indicate that the hypothesized model received an acceptable t to the data: w2 194.72 (p 0.000); CFI 0.981; GFI 0.929; and, [127] RMSEA 0.043. In addition, all hypotheses were supported. Specically, consistent with H1 and H2, positive impacts of both market and learning orientation on relationship quality were found: b 0.28 (po0.01) and g 0.20 (po0.05), respectively. Concerning H3, learning orientation was positively related to market orientation (g 0.54, po0.001). Finally, with regard to H4, relationship quality was positively related to export performance (b 0.58, po0.001) (see Fig. 2). Table 3 shows the unstandardized structural paths and Fig. 2 shows the standardized structural paths of the model. Multigroup Analysis The duration of a relationship is a factor that may affect relationship quality. For example, Wray, Palmer, and Bejou (1994) found that duration of a
mcs mcm mco
0.65 0.72
aff beh con sat2 sat1 0.80 0.67 0.91 0.94

Commitment
0.91
0.99
0.29S Market orientation 0.28**(H1) 0.54**(H3) 0 .20*(H2) Learning orientation 0 .82 0.81 lcm lsv 0.64 lop
Satisfaction
0.91
0.84 0.58**(H4) 0.34S Export performance 0.91 0.91
0.18S Relationship quality 0.59

Minimal opportunism
0.85
Trust
0.92
Eperf3 Eperf2 Eperf1
0.94
0.99
0.82
0.82 ben
mop1 mop2 hon
2(127) = 179.23 (p = 0.002); CFI = 0.986; GFI = 0.935; RMSEA = 0.038 *: significant at p < 0.05; **: significant at p < 0.001; S : squared multiple correlations
Fig. 2.
Structural Results (Standardized Estimates).
121
Table 3.
Unstandardized Structural Paths.

Estimate 0.31 0.23 0.56 0.48 Standard Error 0.091 0.095 0.072 0.051 p-value 0.000 0.015 0.000 0.000
Structural Paths H1 H2 H3 H4 Market orientation Learning orientation Learning orientation Relationship quality Relationship quality Relationship quality Market orientation Export performance
relationship affect trust. However, Lagace, Dahlstrom, and Gassenheimer (1991) found no impact of relationship duration on any dimensions of relationship quality. In addition, rm-ownership structure has been considered an important factor that affects rm performance (e.g., Aulakh & Kotabe, 1997; Luo, Grifth, Liu, & Shi, 2004). Moreover, the private sector has only recently been present in Vietnam. It is argued that the private sector has more idiosyncratic behaviors than the state-owned sector (Friedman, 2004). Therefore, an additional purpose of the research is to examine the difference, if any, based on the duration of the relationship (new and more mature), and on the structure of rm ownership (stateowned and other types of ownership). In order to compare the structural paths in the model between these groups (new versus more mature relationships and state-owned versus other types of ownership), the multigroup analysis in SEM was utilized. It is noted that we measured rm-ownership structure by a nominal scale (1: stateowned rms and 2: others) and duration of relationship by a ratio scale (number of years engaged in the relationship). Therefore, a median split was used to bifurcate the sample into new and more mature relationships. The new relationship group consisted of relationships that were less than or equal to ve years. The more mature relationship group included relationships that had lasted for more than ve years. Two steps of analysis were conducted. First, these two samples were used to estimate the structural paths in the model with no structural paths constrained (Model A). Second, constraints were imposed on the structural paths in both groups new and more mature relationship groups. This was done by setting the hypothesized structural paths between learning orientation and market orientation, between learning orientation and relationship quality, between market orientation and relationship quality, and between relationship quality and export performance to be equal for both groups (Model B). This procedure was also utilized for rm-ownership structure. It is also noted that no constraints were set for the measurement models (partial invariance). In terms of relationship duration, the results of the multigroup analysis show that Model
122
Table 4.
Unstandardized Structural Paths of New and More Mature Relationships.

Structural Paths New Relationships Est(se)a More Mature Relationships Est(se) 0.10(0.134) 0.49(0.156) 0.64(0.119) 0.35(0.078) p-value 0.457 0.002 0.000 0.000
p-value 0.000 0.053 0.000 0.000
H1 H2 H3 H4
a
Market orientation Learning orientation Learning orientation Relationship quality
Relationship quality Relationship quality Market orientation Export performance
0.42(0.115) 0.21(0.109) 0.45(0.090) 0.57(0.068)
Estimate (standard error).
A (w2 390.80) was selected over Model B (w2 401.11) because it [254] [258] received a better t: Dw2 4] 10.31 (po0.05). This result conrms the [Ddf moderating effect of the duration of relationship. Table 4 presents the structural paths of new and more mature relationships. In terms of rmownership structure, the results of the multigroup analysis show that Model B (w2 350.43) was selected over Model A (w2 343.02) because it [258] [254] received a better t: Dw2 4] 7.11 (p>0.11). This result indicates that [Ddf the structure of rm ownership did not moderate the relationships hypothesized in the model.
DISCUSSION, IMPLICATIONS, AND CONCLUSIONS

The major objective of this study is to investigate key antecedents of the quality of business relationship between developing-economy exporters and their foreign importers. Following the resource-based view of the rm and extant research on market and learning orientations, we developed a model that explains the role of market and learning orientations in relationship quality, and subsequently, export performance. The empirical ndings support all of the hypotheses, substantially adding to the literature on the importance of adopting market and learning orientations, especially in the exportimport relationship in developing economies. First, it has been found that market and learning orientations are inuential forces driving exporters to build and maintain strong relationships
123
with their foreign partners. As forms of organizational culture, market and learning orientations are rm-level resources and are related to specic and routine processes that enable exporters to become competent in understanding and responding to their partners needs. These competencies will create superior values to their foreign partners (Celuch et al., 2002; Hult et al., 2004; Yeniyurt et al., 2005). The effort of an exporter to understand and to satisfy its partner will also result in the partners beliefs that the exporter is an expert in performing its obligations and behaves in the best interests of its import partner. Therefore, the import partner is likely to trust and commit to the working relationship with the exporter. As such, market and learning orientations will smooth the exporterimporter relationships and enhance the quality of these relationships, leading to high export performance. Consistent with the results found in advanced economies (e.g., Calantone et al., 2002; Sanzo et al., 2003), the results of this study indicate that the role of market and learning orientations in relationship quality is not limited to advanced economies. Adopting and nurturing these orientations are important for exporters in developing economies who aim to strengthen business ties with foreign partners in competitive environments because these orientations are rm-level resources that create sustainable competitive advantage. The results also indicate that learning orientation facilitates a marketoriented approach. A superior learning environment will leverage the use of all rm resources, including the behaviors that accompany a market orientation (Baker & Sinkula, 1999). Moreover, learning-oriented exporters in developing economies like Vietnam (i.e., a transition economy where business values are still affected by a centrally planned approach), tend to adopt new ways of looking at the market, involving a market-oriented approach. The exporter is likely to withdraw from its routine ways of doing business, which have become embedded in its previous business approach. Therefore, exporters in developing economies, such as Vietnam, who wish to enhance relationship quality with their foreign partners, should dedicate themselves to learning orientations. In short, the importance of market and learning orientations for export rms is so crucial that managers should promote the culture of such orientations. The values and norms of such philosophy must be translated into specic actions in order to attain sustainable competitive advantage. This will strengthen the quality of relationships between them and their import partners, and thus, enhancing their export performance. Further, the multigroup analysis results indicate that no signicant difference between stated-own rms and rms with other types of ownership
124
was found. This implies that adopting and nurturing market and learning orientations are equally important for exporters in both sectors, in their efforts to establish high-quality relationships with their foreign importers. Nevertheless, the results show that learning orientation plays its role in building high-quality relationships for both new and mature relationships. Export managers should take this into consideration. At whatever stage of their business relationships, creating a learning-oriented environment can assist export managers in building and fostering the relationships with their foreign importers. However, the impact of market orientation on relationship quality is signicant only at the early stage of the relationship. This may be a particular characteristic of Vietnam, a developing and transition economy. Prior to the transformation into a market-oriented economy, all export activities of Vietnamese rms had been arranged by the Vietnamese government with other socialist governments within the Soviet trading block. The break-up of the Soviet trading block led to the liberalization of foreign trade by the Vietnamese government in 1989, creating an open economy in Vietnam. This required Vietnamese rms to nd new import partners because the Soviet trading block could no longer provide outlets for exports (Grifn, 1998). Instead of focusing on production and relying primarily on the government, Vietnamese rms are now required to nd their own export markets for their products. As such, market orientation plays an important role in the early stage of their relationships with foreign importers because market orientation facilitates an understanding between two parties. When their relationships are more mature, other investments such as asset specicity and innovativeness (e.g., Haugland, 1999; Hult et al., 2004) may be more important, and market orientation may support these investments rather than directly enhances their relationships. This study has a number of limitations. First, generalizability of the ndings to other settings must be undertaken with the utmost caution because only one city in one developing economy was sampled. Replication and extension to other developing economies is a direction for future research. Second, limitations relate to the examination of the relationship from only one side of the dyad, the exporters. Establishment of the validity of the hypotheses is limited by the single model viewed from only the exporter side. Future research should consider the use of dyadic information. Further, the results of this study show that the impact of market orientation on relationship quality is not signicant at the more mature stage of the relationship. Therefore, other factors, such as asset specicity and innovativeness, can mediate this impact. Further research should examine possible mediators. Finally, the cross-sectional design employed inhibits strong
125
inference regarding the direction of the causal relationships of the constructs. Longitudinal research designs would better enable researchers to make inferences about the causal sequence.
ACKNOWLEDGMENTS
The authors would like to thank the special issue editors, Alex Rialp and Josep Rialp, the two anonymous reviewers, and the participants of the CIMaR 2005 at the Autonomous University of Barcelona, Spain, for their constructive and insightful comments on the previous version of the article.
REFERENCES
Anderson, E., & Weitz, B. (1992). The use of pledges to build and sustain commitment in distribution channel. Journal of Marketing Research, 29(February), 1834. Anderson, J. C., & Narus, J. A. (1990). A model of distributor rm and manufacturer rm working partnerships. Journal of Marketing, 54(January), 4258. Aulakh, P. S., & Kotabe, M. (1997). Antecedents and performance implications of channel integration in foreign markets. Journal of International Business Studies, 28(1), 145175. Bagozzi, R. P., & Edwards, J. R. (1998). A general approach for representing constructs in organizational research. Organizational Research Methods, 1(1), 4587. Baker, W. E., & Sinkula, J. M. (1999). The synergistic effect of market orientation and learning organization on organizational performance. Journal of the Academy of Marketing Science, 27(4), 411427. Barney, J. B. (1986). Organizational culture: Can it be a source of sustained competitive advantage? Academy of Management Review, 11(3), 656665. Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99120. Bigne, E., & Blesa, A. (2003). Market orientation, trust and satisfaction in dyadic relationships: A manufacturerretailer analysis. International Journal of Retail & Distribution Management, 31(11), 574590. Bucklin, L. P., & Sengupta, S. (1993). Organizing successful co-marketing alliances. Journal of Marketing, 57(April), 3246. Calantone, R. J., Cavusgil, S. T., & Zhao, Y. (2002). Learning orientation, rm innovation capacity, and rm performance. Industrial Marketing Management, 31(6), 515524. Cavusgil, S. T., Deligonul, S., & Zhang, C. (2004). Curbing foreign distributor opportunism: An examination of trust, contracts, and the legal environment in international channel relationships. Journal of International Marketing, 12(2), 727. Celuch, K. G., Kasouf, C. J., & Peruvemba, V. (2002). The effects of perceived market and learning orientation on assessed organizational capabilities. Industrial Marketing Management, 31(6), 545554. Crosby, L. A., Evans, K. R., & Cowles, D. (1990). Relationship quality in services selling: An interpersonal inuence perspective. Journal of Marketing, 54(3), 6882.
126
Cullen, J. B., Johnson, J. L., & Sakano, T. (1995). Japanese and local partner commitment to IJVs: Psychological consequences of outcomes and investments in the IJV relationship. Journal of International Business Studies, 26(1), 91115. Day, G. S. (2000). Managing market relationships. Journal of the Academy of Marketing Science, 28(1), 2430. Deshpande, R., Farley, J. U., & Webster, F. E. (1993). Corporate culture, customer orientation and innovativeness in Japanese rms: A quadrad analysis. Journal of Marketing, 57(1), 2327. Dorsch, M. J., Swanson, S. R., & Kelly, S. W. (1998). The role of relationship quality in the stratication of vendors as perceived by customers. Journal of the Academy of Marketing Science, 26(2), 128142. Dwyer, R. F., & Oh, S. (1987). Output sector municence effects on the internal political economy of marketing channels. Journal of Marketing Research, 24(November), 347358. Eisenhardt, K. M., & Martin, J. A. (2000). Dynamic capabilities: What are they? Strategic Management Journal, 21(6), 11051121. Emden, Z., Yaprak, A., & Cavusgil, S. T. (2005). Learning from experience in international alliances: Antecedents and rm performance implications. Journal of Business Research, 58(7), 883892. Field, M. A. (2006). Field of dreams. Journal of Commerce, 7(1), 1618. Frazier, G. L. (1999). Organizing and managing channels of distribution. Journal of the Academy of Marketing Science, 27(2), 226240. Friedman, J. (2004). Firm ownership and internal labour practices in a transition economy an exploration of worker skill acquisition in Vietnam. Economics of Transition, 12(2), 333366. Ganesan, S. (1994). Determinants of long-term orientation in buyerseller relationships. Journal of Marketing, 58(April), 119. Gaski, J., & Nevin, J. (1985). The differential effects of exercised and unexercised power sources in a marketing channel. Journal of Marketing Research, 22(May), 130142. Geringer, M. J., & Herbert, L. (1991). Measuring performance of international joint ventures. Journal of International Business Studies, 22(2), 249264. Grifn, K. (1998). Restructuring and economic reforms. In: K. Grifn (Ed.), Economic reform in Vietnam (pp. 119). London: Macmillan. Gundlach, G. T., Achrol, R. S., & Mentzer, J. T. (1995). The structure of commitment in exchange. Journal of Marketing, 59(January), 7892. Hakansson, H. (1982). International marketing and purchasing of industrial goods: An interaction approach. Chichester, UK: Wiley. Haugland, S. A. (1999). Factors inuencing the duration of international buyerseller relationships. Journal of Business Research, 46(3), 273280. Hieu Long. (2006). Opportunities for export growth. Vietnam Investment Review, April 12. Available at www.vir.com.vn (in Vietnamese). Hult, G. T. M., Hurley, R. F., & Knight, G. A. (2004). Innovativeness: Its antecedents and impact on business performance. Industrial Marketing Management, 33(5), 429438. Hult, G. T. M., Nichols, E. L., Jr., Giunipero, L. C., & Hurley, R. F. (2000). Global organizational learning in the supply chain: A low versus high learning study. Journal of International Marketing, 8(3), 6183. Hurley, R. F., & Hult, G. T. M. (1998). Innovation, market orientation, and organizational learning: An integration and empirical examination. Journal of Marketing, 62(July), 4254.
127
Jap, S. D., Manolis, C., & Weitz, B. A. (1999). Relationship quality and buyerseller interactions in channels of distribution. Journal of Business Research, 46(3), 303313. Jaworski, B. J., & Kohli, A. K. (1996). Market orientation: Review, renement, and roadmap. Journal of Market Focused Management, 1(2), 119135. Kalwani, M. U., & Narayandas, N. (1995). Long-term manufacturersupplier relationships: Do they pay off for supplier rms? Journal of Marketing, 59(January), 116. Kim, K., & Frazier, G. F. (1997). Measurement of distributor commitment in industrial channels of distribution. Journal of Business Research, 40(2), 139154. Kumar, N., Scheer, L. K., & Steenkamp, J.-B. E. M. (1995). The effects of supplier fairness on vulnerable resellers. Journal of Marketing Research, 32(February), 5465. Kumar, N., Stern, L. W., & Anderson, J. C. (1993). Conducting organizational research using key informants. Academy of Management Journal, 36(6), 16331651. Lagace, R. R., Dahlstrom, R., & Gassenheimer, J. B. (1991). The relevance of ethical salesperson behavior on relationship quality: The pharmaceutical industry. Journal of Personal Selling & Sales Management, 11(4), 3947. Lages, C., Lages, C. R., & Lages, L. F. (2005). The RELQUAL scale: A measure of relationship quality in export market ventures. Journal of Business Research, 58(8), 10401048. Lee, D.-J. (1998). Developing international strategic alliances between exporters and importers: The case of Australian exporters. International Journal of Research in Marketing, 15(4), 335348. Leonidou, L. C., & Kaleka, A. A. (1998). Behavioral aspects of international buyerseller relationships: Their association with export involvement. International Marketing Review, 15(5), 373397. Leuthesser, L. (1997). Supplier relational behavior: An empirical assessment. Industrial Marketing Management, 26(3), 245254. Leuthesser, L., & Kohli, A. K. (1995). Relational behavior in business markets. Journal of Business Research, 34(3), 221233. Luo, X., Grifth, D. A., Liu, S., & Shi, Y.-Z. (2004). The effects of customer relationships and social capital on rm performance: A Chinese business illustration. Journal of International Marketing, 12(4), 2545. Madsen, T. K. (1998). Managerial judgment of export performance. Journal of International Marketing, 6(3), 8293. Menguc, B., & Auh, S. (2006). Creating a rm-level dynamic capability through capitalizing on market orientation and innovativeness. Journal of the Academy of Marketing Science, 34(1), 6373. Morgan, R. M., & Hunt, S. D. (1994). The commitmenttrust theory of relationship marketing. Journal of Marketing, 58(July), 2038. Muthen, B., & Kaplan, D. (1985). A comparison of some methodologies for the factor analysis of non-normal Likert variables. British Journal of Mathematical and Statistical Psychology, 38(May), 171180. Narver, J. C., & Slater, S. F. (1990). The effect of a market orientation on business protability. Journal of Marketing, 54(October), 2035. Nguyen, T. T. M., Barrett, N. J., & Nguyen, T. D. (2004). Cultural sensitivity, information exchange, and relationship quality. Journal of Customer Behaviour, 3(3), 281303. Noorderwier, T. G., John, G., & Nevin, J. R. (1990). Performance outcomes of purchasing agents in industrial buyervendor relationships. Journal of Marketing, 54(October), 8093.
128
Parkhe, A. (1993). Strategic alliance structuring: A game theoretic and transaction cost examination of inter-rm cooperation. Academy of Management Journal, 36(4), 794829. Raven, P. V., McCullough, J. M., & Tansuhaj, P. S. (1994). Environmental inuences and decision making uncertainty on satisfaction and performance. Journal of International Marketing, 2(3), 3759. Reicheld, F. F. (1996). The loyal effect. Boston, MA: Harvard Business School Press. Rosson, P. J., & Ford, D. I. (1982). Manufacturer-overseas distributor relationships and export performance. Journal of International Business Studies, 13(Fall), 5772. Saigon Tiepthi. (2006). Foreign trade drive for economic growth. Saigon Tiepthi, April 10. Availalble at www.sgtt.com/web/tintuc (in Vietnamese). Sanzo, J. M., Santos, M. L., Vazquez, R., & Alvarez, L. I. (2003). The role of market orientation in business dyadic relationships: Testing an integrator model. Journal of Marketing Management, 19(1/2), 73107. Siguaw, J. A., Simpson, P. M., & Baker, T. L. (1998). Effects of supplier market orientation on distributor market orientation and the channel relationship: The distributor perspective. Journal of Marketing, 62(3), 99111. Simpson, P. M., Siguaw, J. A., & Baker, T. L. (2001). A model of value creation: Supplier behaviors and their impact on reseller-perceived value. Industrial Marketing Management, 30(2), 119134. Sinkula, J. M. (1994). Market information processing and organisational learning. Journal of Marketing, 58(1), 3545. Sinkula, J. M., Baker, W. E., & Noordewier, T. G. (1997). A framework for market-based organizational learning: Linking values, knowledge, and behavior. Journal of the Academy of Marketing Science, 25(4), 305318. Slater, S. F., & Narver, J. C. (1995). Market information processing and organizational learning. Journal of Marketing, 59(January), 3545. Smith, B. (1998). Buyerseller relationships: Bonds, relationship management, and sex-type. Canadian Journal of Administrative Sciences, 15(1), 7692. Srivastava, R. K., Fahey, L., & Christensen, H. K. (2001). The resource-based view and marketing: The role of market-based assets in gaining competitive advantage. Journal of Management, 27(6), 777802. Steenkamp, J.-B. E. M., & van Trijp, H. C. M. (1991). The use of LISREL in validating marketing constructs. International Journal of Research in Marketing, 8(4), 283299. Stern, L. W., & Reve, T. (1980). Distribution channels as political economies: A framework for competitive analysis. Journal of Marketing, 44(Summer), 5264. Styles, C., & Ambler, T. (1994). Successful export practice: The U K experience. International Marketing Review, 11(6), 2449. Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509533. Tsang, E. W. K. (2005). Inuences on foreign ownership level and entry mode choice in Vietnam. International Business Review, 14(4), 441463. Wernerfelt, B. (1984). A resource-based view of the rm. Strategic Management Journal, 5(2), 171180. Williamson, O. E. (1975). Markets and hierarchies: Analysis and antitrust implications. New York: The Free Press. Wray, B., Palmer, A., & Bejou, D. (1994). Using neural network analysis to evaluate buyer seller relationships. European Journal of Marketing, 28(10), 3248.
129
Yeniyurt, S., Cavusgil, S. T., & Hult, G. T. M. (2005). A global market advantage framework: The role of global market knowledge competencies. International Business Review, 14(1), 119. Zou, S., Taylor, C. R., & Osland, G. E. (1998). The EXPERF scale: A cross-national generalized export performance measure. Journal of International Marketing, 6(3), 3758.
APPENDIX: CFA FACTOR LOADINGS OF ITEMS (STANDARDIZED)

Items Market orientation (second-order, 7-point Likert scale) Customer orientation We closely monitor and assess our level of commitment in serving customers needs Business strategies are driven by the goal of increasing customer value Our competitive advantage is based on understanding customers needs Our business objectives are driven by customer satisfaction We pay close attention to after-sales service We respond quickly to customer needs We rapidly adapt our products in response to customers needs Competitor orientation In our rm, our sales people share information about competitors We respond rapidly to competitor actions Top management regularly discuss competitors strengths and weaknesses Customers are targeted when we have an opportunity for competitive advantage Interfunctional coordination Our top managers from each business function regularly visit customers Business functions within our rm are integrated to serve our target market needs Our managers understand how employees can contribute to customers value We share resources with other business units Loading t-value
0.81 0.81 0.85 0.86 0.87 0.82 0.84 0.84 0.80 0.87 0.79
16.45 16.43 17.96 18.10 18.44 16.78 15.54 17.62 15.27
0.79 0.80 0.83 0.76
14.26 14.85 13.44
130
APPENDIX (Continued )
Items Learning orientation (second-order, 7-point Likert scale) Commitment to learning Managers basically agree that our rms ability to learn is the key to our competitive advantage The basic values of our rm include learning as a key to improvement In our rm, employee learning is an investment, not an expense Learning in our rm is seen as a key commodity necessary to guarantee organizational survival Shared vision There is a commonality of purpose in our rm There is total agreement on our organizational vision across all levels, functions, and divisions All employees are committed to the goals of our rm Employees view themselves as partners in charting the direction of our rm Open mindedness We are not afraid to reect critically on the shared assumptions we have made about our markets Personnel in our rm realize that the very way they perceive the marketplace must be continually questioned We always collectively question our own biases about the way we interpret market information Trust (second-order, 7-point Likert scale) Honesty Our rm has often provided importer A information that has later proven to be accurate Importer A has often provided us information that has later proven to be accurate Our rm usually keeps the promises we make to importer A Importer A usually keeps the promises they make to our rm Whenever our rm gives importer A advice on business operations, they know we are sharing our best judgment Whenever importer A gives us advice on business operations, we know they are sharing their best judgment Our rm can count on importer A to be sincere Our rm think that importer A can count on us to be sincere Our managers feel that the importer A offers our rm reliable recommendations Our rm offers to importer A reliable recommendations Loading t-value
0.85 0.87 0.88 0.87
18.51 18.87 18.59
0.82 0.84 0.89 0.87
16.49 17.98 17.51
0.81 0.84 0.81
14.70 14.22
0.71 0.69 0.75 0.74 0.84 0.76 0.66 0.76 0.80 0.66
11.15 12.17 11.94 13.59 12.32 10.67 12.33 13.03 10.66
131
Items Our rm deals fairly and sincerely with importer A Our rm feels that importer A deals fairly and sincerely with our rm Benevolence Though circumstances change, our rm will be ready and willing to offer importer A an assistance and support When making important decisions, our rm is concerned about importer As welfare When importer A share their problems with our rm, we will respond with understanding When it comes to things that are important to us, we can depend on importer As support When we share our problems with importer A, we know that they will respond with understanding Though circumstances change, our rm believes that importer A will be ready and willing to offer us assistance and support When making important decisions, importer A is concerned about our rms welfare When it comes to things that are important to us, importer A can depend on our rms support We can count on importer As decisions and actions which will be favorably affect us in the future Importer A can count on our rms decisions and actions which will be favorably affect them in the future Commitment (second-order, 7-point Likert scale) Continuance commitment We are going to continue the relationship with importer A for many years We think that importer A is going to continue the relationship with our rm for many years We expect the business relationship with importer A to last for a long time Our rm is certain that our relationship with importer A will last a long time Our rm believes that if another exporter offered importer A a better deal, they would take them on even if it meant dropping us If another importer offered us a better deal, we would take them on even if it meant dropping importer A Loading 0.81 0.81 t-value 13.15 13.11
0.69 0.59 0.65 0.68 0.69 0.71 0.77 0.70 0.63 0.60
9.13 9.98 10.41 10.61 10.91 11.76 10.72 9.77 9.34
0.76 0.84 0.89 0.90 0.79
15.19 16.36 16.51 14.05
0.82
14.80
132
Items Behavioral commitment We respond quickly to importer As requests for help We think that the importer A responds quickly to our rms requests for help We devote more time to importer A when they need help We adjust our marketing program for importer A when necessary We provide special aid to importer A when they are in trouble We provide customized product or services as requested by importer A We give advice and suggestions when importer A has problems We think that importer A devoted more time when we need help We think that importer A provided special aid to us when we were in trouble Importer A has given advice and suggestions when we have problems Affective commitment A high sense of unity exists between importer A and us Our relationship with importer A is a long-term alliance We have a strong business link with importer A We want to remain a member of importer As network because we genuinely enjoy our relationship with them We think that importer A wants to remain a member of the our rms business network because they genuinely enjoy their relationship with us We think importer A has developed a close business relationship with us Importer A and our rm share common business interests Minimal opportunism (7-point Likert scale) We have always provided importer A with a completely truthful picture of our business Importer A has always provided our rm with a completely truthful picture of their business We carry out the duties of our relationships even if importer A does not check up on us Our rm think that importer A carries out the duties of relationships even if our rm does not check up on them We have never promised importer A that we would do things, that we actually had no intention of following through Loading t-value
0.69 0.63 0.70 0.54 0.67 0.62 0.60 0.67 0.68 0.73
9.61 10.59 8.40 10.16 9.47 9.23 10.22 10.35 11.07
0.77 0.82 0.79 0.81 0.75
14.54 13.95 14.36 13.21
0.81 0.70 0.78 0.78 0.79 0.81 0.81
14.39 12.22 14.56 14.69 15.24 15.21
133
Items Importer A has never promised us that they would do things, that they actually had no intention of following through To get the necessary support from importer A, we rarely mask the true nature of our needs To get the necessary support from our rm, importer A rarely masks the true nature of their needs Regardless of its impact on importer As business (e.g., protability, sales volume, and market share), they always conscientiously perform their duties to us In order to get what importer A needs from our rm, importer A rarely changed the facts In order to get what we need from importer A, we have rarely changed the facts On rare occasions, we have had to lie to importer A about certain things in order to protect our interests Satisfaction (7-point Likert scale) Overall, I believe we are both quite satised with our working relationship This is among the best importer relationships that our managers have experienced Our rms relationship with importer A has been a happy one Our rms relationship with importer A has fully met our rms expectations Our rm is proud of having this working relationship with importer A We are very pleased with what this importer A does for us If we had to do it all over again, we would still choose to use the importer A Export Performance (7-point Likert scale) Protability Sales revenue stability/growth Improvement in market share Competitive advantage over other rms Economies of scale Overall survival of the rm Reputation Market development Product development Loading 0.82 0.87 0.87 0.85 t-value 15.54 16.65 16.67 16.10
0.85 0.83 0.81
16.09 15.64 15.16
0.87 0.88 0.91 0.86 0.89 0.91 0.87
20.33 20.98 22.44 20.00 21.65 22.73
0.74 0.77 0.81 0.81 0.68 0.73 0.83 0.85 0.81
13.28 13.97 13.87 11.58 12.41 14.39 14.79 13.99
THE INTERNATIONALISATION OF BIOTECH SMES: A COMPARATIVE ANALYSIS OF UK AND US FIRMS

Calin Gurau and Ashok Ranchhod
ABSTRACT
The accelerated globalisation of world markets in the last 30 years has increased the importance of internationalisation models for both academics and practitioners. The internationalisation process of SMEs is one of the newest developments in this area, with major implication for the strategic orientation of small rms. However, this phenomenon has to be considered in relation with the specic characteristics for various market environments and industrial sectors. This study attempts to analyse the impact of the domestic market prole on the internationalisation process of biotech SMEs in US and UK, outlining the similarities and the differences between these two countries.
THE INTERNATIONALISATION MODELS

The accelerated globalisation of world markets in the last 30 years has increased the importance of internationalisation theories for both academics and practitioners. The academic research has developed three main
137
138
CALIN GURAU AND ASHOK RANCHHOD
categories of models to describe the process of internationalisation: the experiential learning models, the systematic planning models and the contingency models (Li, Li, & Dalgic, 2004). There are two experiential learning models: the Uppsala model, developed by Johanson and Wiedersheim-Paul (1975), and then rened by Johanson and Vahlne (1977, 1990) and Johanson and Mattsson (1986); and the Management Innovation model, described in the works of Bilkey and Tesar (1977), Cavusgil (1980), Czinkota (1982) and Reid (1981). The systemic approach of these two models is almost similar. Both of them describe the internationalisation process as a gradual evolution of the rm through a series of stages, that correspond to an increasing foreign involvement of the company. The Uppsala model explains the internationalisation process considering two main strategic dimensions: the market knowledge possessed by the rm about different foreign markets and the commitment of rms resources to those markets. The evolution of the company from a mainly domestic activity to a fully international prole is described as a slow, incremental process, which involves the gradual acquisition, integration and the use of knowledge concerning the characteristics of foreign markets as well as an increasing commitment of the companys resources towards international activities. The model also predicts that a rm will rst target the markets that are most familiar in terms of language, culture, business practice and industrial development, in order to reduce the perceived risk of international operations and to increase the efciency of information ows between the rm and the target market (Johanson & Vahlne, 1977, 1990). On the other hand, the Innovation model (Cavusgil, 1980; Czinkota, 1982; Reid, 1981) tends to explain the initiation of the internationalisation process through a series of management innovations implemented within the rm. The original model created by Bilkey and Tesar (1977) emphasises the evolution of a rms internationalisation through a series of successive learning stages. The experiential learning models have been extensively challenged over the years, with numerous scholars advancing various criticisms of their validity and assumption (Anderson, 1993; Axinn & Matthyssens, 2002; Fillis, 2001; Knight & Cavusgil, 1996; Welch & Luostarinen, 1988). Some authors questioned the deterministic nature of the Uppsala and the Innovation models, outlining that many rms do not follow a consistent path to internationalisation (Anderson, 1993; Reid, 1983; Turnbull, 1987). Many studies have found that the management of internationalising rm considers a variety of strategic approaches (Douglas et al., 1982; Root, 1987; Welch & Luostarinen, 1988), the internationalisation process representing a strategic answer to the
The Internationalisation of Biotech SMEs
139
evolving market conditions (Reid, 1983, 1984). Other critics demonstrated that the rst step in the internationalisation process may not be exporting, but one of several alternative international activities, such as licensing, franchising, joint-venture or network relationships (Carstairs & Welch, 1982/ 1983; Hakansson, 1982; Nordstrom, 1991; Reid, 1984; Root, 1987). These criticisms show that it is difcult and even dangerous to draw a unique recipe for the internationalisation process (Buckley, Newbould, & Thurwell, 1979; Garnier, 1982; Root, 1987; Roux, 1979; Varaldo, 1987), and that the rms stage of internationalisation is largely determined by the operating environment, industry structure and its own marketing strategy (Turnbull, 1987). These criticisms have been crystallised in two additional orientations: the strategic planning models, based on the research published by Root (1987), Miller (1993) and Yip, Biscarri, & Monti (2000), which claim that systematic planning based on a careful analysis of competitive factors and circumstantial conditions can signicantly increase the success of the internationalisation process; and the contingency models, developed by Turnbull (1987), and then rened by Roberts (1999), Jones (1999), Welch and Luostarinen (1993), and Boter and Holmquist (1996), which argue that the internationalisation process is inuenced by a series of contextual factors, such as the operating environment and the industry structure (Turnbull, 1987). The Born Global model, proposed by Knight and Cavusgil (1996) is another important contribution to the contingency perspective. Based on a study conducted in Australia, the Born Global companies are dened by the following specic characteristics (Knight & Cavusgil, 1996, p. 18): 1. Management views the world as its marketplace from the outset of the rms founding; unlike traditional companies, they do not see foreign markets as simple adjuncts to the domestic market. 2. Born Globals begin exporting one or several products within two years of their establishment and tend to export at least a quarter of total production. 3. They tend to be small manufacturers, with average annual sales usually not exceeding $100 million. 4. The majority of Born Globals are formed by active entrepreneurs and tend to emerge as a result of a signicant breakthrough in some process or technology. 5. They may apply cutting-edge technology to developing an unique product idea or to a new way of doing business. 6. The products that Born Globals sell typically involve substantial value adding; the majority of such products may be intended for industrial uses.
140
Knight and Cavusgil (1996) argued that this phenomenon was facilitated by series of recent trends that have increased the capacity of many companies to initiate early international activities: 1. The increasing role of niche markets, especially in the countries of the developed world (Holstein, 1992; Robinson, 1986), determined by increased demand for specialised or customised products. 2. The recent advances in process technology that have created the possibility of exible, low-scale and low-cost production. 3. The recent advances in communication technologies have reduced the costs of information transmission between distant markets. 4. The inherent advantages of small companies such as quicker response time, exibility, adaptability and direct customer relations facilitate the international operations of Born Global companies, and offer them an important competitive edge in comparison with the larger multinationals (McKinsey and Co., 1993). 5. The means of internationalisation such as knowledge, technology, tools and facilitating institutions have become more accessible to all rms, regardless of their size or activity sector (Czinkota & Ronkainen, 1995; Nordstrom, 1991). 6. The recent emergence of complex transnational networks of strate gic alliances (Coviello & Munro, 1997; Hakansson, 1982; Thorelli, 1990). The Born Global model is supported by numerous studies that show that these types of companies are emerging in many national economies including France (Roux, 1979), Canada (Denis & Depelteau, 1985; Garnier, 1982), Taiwan (Chang & Grub, 1992), Finland (Gabrielsson, Sassi, & Darling, 2004), the United Kingdom (Buckley et al., 1979) and the United States (Holstein, 1992; Norton, 1994). Despite the growing interest in the internationalisation process of SMEs, very few studies have been conducted at the transnational level (Knight, Madsen, & Servais, 2004; Moen, 2002; Rasmussan, Madsen, & Evangelista, 2001). Knight et al. (2004) have investigated the phenomenon of Born Global rms based on data from case- and survey-based studies in Denmark and the USA. The results of this study indicate that the international performance of rms is enhanced by the managerial focus on foreign customer and marketing competence. Moen (2002) has analysed the export behaviour of SMEs from France and Norway identifying the global orientation of the manager and the market conditions as important factors for a quick
141
internationalisation. Finally, Rasmussan et al. (2001) have studied the founding process of Born Globals in Denmark and Australia, outlining the importance of networking and sensemaking. Many authors emphasised that the internationalisation of SMEs has to be considered specically in relation to the industry sector and the local market prole (Boter & Holmquist, 1996; Moen, 2002; Reid, 1983, 1984; Turnbull, 1987). However, there are yet no transnational studies concerning the particular inuence of the domestic market prole on the internationalisation of SMEs from a specic industry. This research project starts from the assumption that the differences among various countries in terms of resource availability, market size and sector development will determine a specic pattern of the internationalisation process: the rms from rich, large markets will have no reason to become international at an early stage of their corporate life, because they can nd locally the necessary resources for product development, as well as a highly responsive demand. On the other hand, the start-ups emerging in countries with more limited facilities and resources may attempt a quicker internationalisation in order to obtain the necessary assets for their survival and development. In order to verify this assumption, this article compares some aspects (speed, degree and main reasons) of the internationalisation process of biotech SMEs in two countries that have a different domestic market prole: US the worlds largest biotech market and UK comparatively, four times smaller than the UK market (Table 1). Through this methodological approach, this study attempts to rene the existing internationalisation models, demonstrating that in the case of biotech SMEs, the internationalisation process is signicantly inuenced by the prole of the domestic market. The choice of the biotech sector was determined by its specic international Table 1. A Comparison of the Main Indicators Dening UK and the US Biotech Market in 2003.
UK 455 22,400 224 5.4 billion dollars 588 million dollars 36 US 1830 172,400 1,110 52 billion dollars 9.2 billion dollars 83
Indicators (2003) Number of companies People employed Dew drugs in clinical development or expecting approval Generated revenues Equity raised New start-ups Source: DTI (2003).
142
characteristics (Gurau & Ranchhod, 1999): the sources fuelling the biotechnology industry are international (i.e. nance, knowledge, legal advice, etc.) (Acharya, 1998, 1999; Russel, 1988); the marketing of biotechnology products and services is international (Acharya, 1999; Daly, 1985); the competition in the biotechnology sector is international (Acharya, 1999; Russel, 1988); and the international community (Acharya, 1999; Bauer, 1995; Nelkin, 1995; Russel, 1988) closely scrutinises the scientic or industrial developments in biotechnology. After this discussion of the internationalisation theory, the paper presents the specic conditions of internationalisation for biotech SMEs, and the differences between various world regions in terms of bio-entrepreneurship. Based on these elements, three research hypotheses are formulated, which attempt to outline the inuence of the domestic market prole on the level, speed and main reasons of internationalisation for the UK and US biotech SMEs. The research methodology and the research ndings are then discussed and analysed. The paper concludes with a summary of the ndings, outlining their relevance for the development of academic models, as well as for the management and government practice.
THE INTERNATIONALISATION OF BIOTECH SMES

Biotechnology is the systematic industrial use of biological processes and organisms to manufacture medical, agricultural and consumer products (Biobox, 1998; Oakey, Faulkner, Cooper, & Walsh, 1990). The range of biotechnology applications includes bulk and specialty chemicals, healthcare products, many food and drink products, extraction processes, waste and pollution treatment, and agriculture (Mantegazzini, 1986). Considering the internationalisation process of biotech rms, it is important to present the reasons that explain the SMBEs need to initiate and develop international operations: 1. Biotechnology is a global sector (Daly, 1985). 2. All the industries that use biotechnology applications are highly internationalised or globalised (Russel, 1988).
143
3. The limited human, technological and nancial resources available on the domestic market may force a company to search abroad the necessary inputs for the development of biotechnology activities (Fontes & Coombs, 1997). On the other hand, a domestic business environment with excessive safety standards and regulations may encourage SMBEs to search abroad for more permissive markets. 4. The domestic market is often too small in number of customers or purchasing power in order to allow a rapid return on investments (Kobrin, 1997). In the rst few years of activity, many biotech SMEs are developing a nal product, a project that requires time and complex ressources (Dorabjee, Lumley, & Cartwright, 1998; Persidis, 1996); therefore, their international activities will not consist in product exports, but rather in contacts and collaboration with foreign organisations that can supply them with necessary resources, such as money, information, technology or specialised services. Once these rms have developed a nal product or a service platform, they attempt to commercialise it in order to obtain a quick return on investment. In some cases, the number of customers within the national market might be too small to ensure a proper return on investment, especially when the investment is large. Only by integrating potential customers from many different countries into a global market segment with a homogeneous demand, is a company able to compete and to efciently develop its activities. Regarding this aspect Kobrin (1997, p. 151) wrote:
y the evidence strongly suggests that the minimum size of markets needed to support technological development in industries such as aerospace, semiconductors, and pharmaceuticals is now larger than the largest national markets. In an emerging global economy international integration is a requisite of a competitive R&D budget; national markets are fused transnationally rather than linked through cross-border transactions.
This is especially true for the European biotechnology rms. Considering the limited size of some national markets (the Netherlands or Switzerland compared with the large US market), the local biotech rms are obliged to expand their activities abroad and became internationalised, or, in some cases, even globalised. 5. The complexity and high risks of innovation and technological development within the biotechnology sector determine the development of an integrated system of transnational strategic alliances it appears that even the global integration of markets by a single rm may no longer be sufcient to offset the huge costs and risks of technological development in a number of strategic industries (Kobrin, 1997, p. 150).
144
Considering the specic characteristics of biotechnology, it may be expected that some of the biotech rms can be dened as Born Global rms. However, many authors emphasised that the phenomenon of Born Global rms has to be considered specically in relation to the industry sector and the local market prole.
BIO-ENTREPRENEURSHIP: AN INTERNATIONAL PERSPECTIVE

The SMBEs have many common characteristics with the general prole of SMEs (Carson, Cromie, McGowan, & Hill, 1995). As most of the SMEs, the SMBEs have limited resources. The biotechnology activities usually require a high level of technological expertise, and the SMBEs tend to ignore the importance of marketing operations, concentrating their resources on the process of research and development (Briggs, 1994). The entrepreneurial model of marketing activities also applies to SMBEs. Most of these companies are managed by biotechnology specialists (Briggs, 1994; Bullock & Dibner, 1995; Euromonitor, 1988; Saviotti, 1998), which have either created their own rm in order to develop promising research discoveries, or have abandoned their positions in large hierarchical structures (big companies, universities) in order to enjoy the independence of entrepreneurial activity (Acharya, 1999; McKelvey, 1996; Witholt, 1999). The United States leads in bio-entrepreneurship, with Europe signicantly behind but closing the gap, and Japan a very distant third. The market conditions, business environment (nancial and legal), social response, regulatory conditions, scientic base and government support favour such companies in the United States (Persidis, 1998). In this respect, the United States are characterised by (Rothwell & Zegveld, 1983; Senker, 1998; Senker, Joly, & Reinhard, 1998): a large domestic market conducive to rapid growth and development; the availability of private wealth as a source of seed capital for the start-up of new ventures; a scal framework which encourages the ow of private risk capital into new ventures; a prevailing attitude (in society at large) which encourages entrepreneurship; high mobility of individuals between academic institutions and private industry;
145
the behavioural and attitudinal character of American scientists, many of whom are willing to establish their own business in order to exploit their technical knowledge; and a large and active government expenditure programme, which provides signicant opportunities for the new high-technology rms, particularly through government procurement programmes. In comparison, the European markets in the past were disadvantaged because (Rothwell & Zegveld, 1983; Senker, 1998; Senker et al., 1998): they were more fragmented; the cultural and attitudinal factors among academics, government, scientists and research institutions have been unfavourable towards technological entrepreneurship; the level of government research and development expenditures was low; the systems of taxation were unfavourable for the development of new high-technology rms; the availability of venture capital was limited; and the geographical and social mobility had low levels. The European market environment has become more favourable for biotechnology rms in the recent years (Ewing, 1996). The national governments have created better facilities for the assistance of SMBEs (Lahteenmaki, Plant, & Michael, 1997), the venture capital market has ex panded and the European Union programmes have offered new opportunities for joint research and development projects (Ernst & Young, 2000; Magnien & de Nettancourt, 1993). For example, the clinical trials and their evaluation are already considered to be quicker and cheaper in Europe than in the USA (Ward, 1995). Among the West European nations, Germany and the UK have the biggest number of biotechnology rms (Ernst & Young, 2000; Green, 1998). In spite of these improvements, the level of biotechnology regulation is still perceived to be stricter than in the USA (Davidson, 1998; Senker et al., 1998) in some industrial areas. It can be reasonably expected that the different market prole of the UK and US biotechnology sector will impact on the internationalisation process of SMBEs. US is presently considered to be the major biotechnology market in the world, in terms of size, diversication and available resources. On the other hand, UK has a successful biotechnology sector, but the market environment is comparatively four times smaller than the US biotech market (see Table 1), which might impact on the capacity of British SMBEs to nd
146
the necessary resources, business partners and customers, exclusively on the domestic market. From this perspective it can be assumed that for the British SMBEs, the international expansion may be a necessary condition for their survival and development, while in the case of US SMBEs, the internationalisation may be just one of the possible alternatives of development. On the basis of these assumptions, three research hypotheses have been formulated: H1. The level of internationalisation of US SMBEs is lower than the level of internationalisation of UK SMBEs. H2. The Born Global phenomenon is more prevalent in the UK biotechnology sector in comparison with the US. H3. The reasons for internationalisation will differ between the US and the UK SMBEs: most UK SMBEs will start international operation to access resources, while most US rms will become international following a strategy of market expansion.
RESEARCH METHODOLOGY
A list of all UK and US biopharmaceutical SMEs was compiled using the Internet (http://bio.com, http://www.bioindustry.org, http://www. biotechanalytics.com, http://www.hum-molgen.de/companies/, http://proles.wisi.com) and hardcopy databases (The UK BioCommerce Data Compendium, 2001). The biotechnology companies have been dened as any UK or US rm involved either directly (product development) or indirectly (platform technology) in the development of new products or processes using biotechnology methods (Daly, 1985). In order to classify the biotechnology companies into small- and mediumsized rms, the number of staff was taken into account. This is also the most common denitional basis used in the OECD countries (OECD, 1997). On the other hand, the extreme variability of SMBEs regarding the market share, the turnover and protability has indicated the choice of the staff number as the most suitable criteria to be used. Owing to the biotechnology sector characteristics, the categories dened by the UK Centre for Exploitation of Science and Technology (CEST) were used to classify the rms with up to 50 employees as small, and the companies employing 51500 people as medium sized (Keown, 1993).
147
A structured questionnaire was sent to 463 UK and 1,334 US SMBEs (which represent all the biotech rms that were identied using the sources presented above), in order to collect information about their business prole, speed and level of internationalisation, main reasons for internationalisation, foreign market entry strategies and presence in global regions. One hundred and thirty-six UK and 464 US completed questionnaires were returned, representing a response rate of 29.37% for the UK survey and 34.78 for the US survey. After the analyses of responses, 122 companies were selected as representing UK-independent biopharmaceutical companies and 444 as independent US SMBEs, which closely corresponds to the proportion between the size of the UK and the US biotech sectors. Only independent companies have been selected for analyses because the integration in a multinational group is likely to considerably affect the international marketing strategy implemented by the SMBEs and the level of resources available at corporate level. The data collected have permitted the identication of the business prole of the selected companies. The SMBEs have been classied into ve distinct sectors: healthcare; environmental biotechnology; agriculture/food; bio-reagents and instruments; and supporting services (such as consulting, bioinformatics, etc.).
This classication provides a rather clear distinction among different biotechnology applications, allowing a delimitation of the specic characteristics of each industry, being consistent with the previous classications used in the specialist literature (Euromonitor, 1988; Mantegazzini, 1986). The level of internationalisation was dened as a composite dimension (Table 2), taking into consideration the criteria established and applied by OECD (1997): (a) the proportion of the SMBEs outputs and inputs (including capital) that are traded across national boundaries, either directly or indirectly; (b) the number of international operations in different regions or countries; and (c) the competitive model faced by the company. The conditions required to consider a rm as Born Global are dened differently by various authors, especially in what concerns the time period
148
Table 2.
Description
The Dimensions of SME Globalisation.

Establishments and Afliations Market Opportunities and Competition No market outside local area, no potential competition from outside local area. Barriers to entry to outside markets and to local market (for competitors) are signicant and amount to more than 50% of costs. Barriers to entry are noticeable, make up to 10% of cost disadvantage, but can be overcome fairly easily.
Traded Inputs and Outputs
No globalisation Domestic
All inputs sourced from local area, all outputs sold in local area. o10% of inputs sourced across borders, and o10% revenue from across borders, usually within a limited span of nations. >10% but o40% of inputs sourced internationally, and >10% but o40% of revenue from across borders, usually across two major international regions. >40% of inputs sourced internationally, >40% of revenue from outputs traded across borders, across all major international regions. Majority of inputs of any establishment sourced across borders, large majority of outputs traded across borders.
Single establishment, no establishments or afliations outside local area. At least one establishment or afliate outside local area or outside national area.
Limited globalisation Mainly domestic
Major globalisation Internationalised
Establishments or close afliates in at least four different nations and in two major international regions.
Extensive globalisation Globalised
Establishments or close afliates in at least one country in all three major international regions.
Barriers to entry to international markets are not signicant impediment for rm or competitors, make up less than 5% cost disadvantage. Markets in all major international regions, competition likely to be present or come from any international region.
Complete globalisation Fully globalised
Multiple establishments or afliates in many countries and in all major international regions.
Source: OECD (1997, p. 23).
149
considered necessary for rms internationalisation, and the extent of rms international involvement (Jolly, Alahuhta, & Jeannet, 1992; Knight & Cavusgil, 1996; Madsen & Servais, 1997; McAuley, 1999). Considering the specic characteristics of the biopharmaceutical sector (highly globalised market), the criteria for dening Born Global companies was the initiation of any form of international operation in the rst three years of rms life, a period of time that is also used by Rasmussan et al. (2001). The main reasons for starting the international activities have been classied into the following categories (Gurau & Ranchhod, 1999): access access access access to to to to sources of funding; new markets; information; and technology,
allowing also an open, direct response, under the category of other reasons, where the respondent was expected to write down any different reasons of the company internationalisation. The data have been processed using the SPSS software package, which allows the quick application of statistical tests to large volumes of data.
PRESENTATION AND INTERPRETATION OF DATA

The Prole of Respondent Firms As it can be seen from Table 3, the distribution of UK and US SMBEs on various sectors of activity is almost similar. The larger differences are in the healthcare sector, where the US companies have the largest percentage (48.6% in comparison with 37.7%). Table 3.
Nationality
Cross-Tabulation between the Nationality of the Firms and Their Sector of Activity.
Health Care Environmental Food and Agribiotech Bioreagents and Instruments 32 (26.2) 94 (21.2) 126 Supporting Services Total
Sector of Activity
UK US Total
N (%) N (%)
46 (37.7) 216 (48.6) 262
12 (9.8) 30 (6.8) 42
20 (16.4) 50 (11.3) 70
12 (9.8) 54 (12.2) 66
122 (100) 444 (100) 566
150
Table 4.
Nationality UK US Total
Cross-Tabulation between the Nationality of the Firms and Their Size.

Size N (%) N (%) Small 64 (52.5) 294 (66.2) 358 Medium 58 (47.5) 150 (33.8) 208 Total 122 (100) 444 (100) 566
Table 5.
Nationality
Cross-Tabulation between the Nationality of the Firms and Their Level of Internationalisation.
Mainly Domestic 17 (13.9) 115 (25.9) 132 Internationalised Globalised Fully Globalised 15 (12.3) 44 (9.9) 59 Total
Level of Internationalisation N (%) N (%)
UK US Total
51 (41.8) 175 (39.4) 226
39 (32) 110 (24.8) 149
122 (100) 444 (100) 566
Note: w 8.416; P 0.038.
The proportion of small- and medium-sized biotechnology rms is almost similar for the UK and US respondent companies (Table 4). However, the US respondent companies are predominantly small (66.2%), while for the UK rms the proportion of small- and medium-sized respondent companies is almost equal. The Level of Internationalisation The results presented in Table 5 show that the level of internationalisation of UK-based and US SMBEs is radically different. 25.9% of the US rms are mainly domestic, while in the same category there are only 17 UK-based rms (13.9%). On the other hand, the percentages of globalised (24.8%) and fully globalised (9.9%) US rms is lower than the percentages of the same categories in UK (32 and 12.3%, respectively). These differences are statistically signicant at a level of P 0.038. These ndings conrm the research Hypothesis 1. This situation can be explained by differences in the market size and structure between the two countries. The US market is very developed for biotechnology products and services, being considered at present as the main global market for this industrial sector. Therefore, the US rms can obtain fairly easily on their local market the necessary resources and a high level of demand for their products/services. On the other hand, the UK
151
Table 6.
Nationality UK US Total
Cross-Tabulation between the Nationality of the Firms and Their Speed of Internationalisation.
Born Global N (%) N (%) Born Global 102 (83.6) 125 (28.2) 227 Not Born Global 20 (16.4) 319 (71.8) 339 Total 122 (100) 444 (100) 566
Note: w 122.515; Po0.0001.
market has, comparatively, a smaller size and cannot offer the same level of business opportunities to the UK-based companies. Owing to these differences, the UK SMBEs are forced to become international and develop overseas operations. Born Global Firms Table 6 shows that Hypothesis 2 is conrmed by the ndings of this study. The UK Born Global SMBEs are the majority among the respondent rms (83.6%), while in the US most of the respondent rms cannot be considered Born Global (71.8%). These differences are logical when considering the prole of the US and UK biotechnology markets and business environments. The limitations of the UK market in terms of size, available resources and level of demand force the UK SMBEs to initiate quickly international operations. This pressure is much lower on the US-based biotechnology rms, who can activate successfully on the large US biotech market. The differences between the UK and the US Born Global SMBEs are statistically relevant to a level of Po0.0001. Main Reasons of Internationalisation Table 7 presents the ndings related with the main reason of SMBEs internationalisation. As it can be seen, the majority of US biotechnology rms are starting international operations to access new overseas markets for their nalised products and services. On the other hand, for the UK SMBEs, the access to new markets is the main reason for a quite small percentage of the respondent rms (18%), the major reasons for the internationalisation of these companies being the access to information and technology. These results conrm Hypothesis 3.
152
Table 7.
Nationality
Cross-Tabulation between the Nationality of the Firms and Their Main Reason for Internationalisation.
Reason for Internationalisation N (%) N (%) Funding New Markets 22 (18) 307 (69.1) 329 Information Technology Total
UK US Total
11 (9)
48 (39.3) 81 (18.2) 129
41 (33.6) 56 (12.6) 97
122 (100) 444 (100) 566
11
2
Note: w 126.353; Po0.0001.
The statistical signicance of the relationship between variables is very strong, at a level of Po0.0001. These ndings complement well the results presented in Table 5. A possible explanation of this situation can be developed considering the corporate lifecycle of the biotech SMEs: starting their internationalisation process later in their corporate life, the US SMBEs have the necessary time to develop and nalise a product or service, that can be offered in overseas markets. These rms do not need to access complementary resources on a global basis, because the US market offers a diversied and complex range of raw materials, products, services and resources, which can be easily accessed by local rms. The situation is not similar in the UK market, which is comparatively smaller and cannot provide all the necessary resources for the UK-based biotechnology rms. The UK SMBEs will initiate international operations to access the necessary resources for their product/process development, and only in a later stage will penetrate overseas market for marketing and commercialisation. However, the validity of this explanation must be veried in future empirical studies.
CONCLUDING REMARKS
The present paper attempted to investigate the differences between the internationalisation process of the US and the UK SMBEs, determined by the specic characteristics of each national market. In order to identify and analyse these differences, the three research hypotheses were formulated. The research results have strongly conrmed these hypotheses. The ndings of this study indicate that the internationalisation process may be inuenced by the characteristics of the domestic market in the case of biotech SMEs. There is an important difference between the level, speed and main reason of internationalisation of the UK- and the US-based biotechnology rms. The size and the level of resources available on each national market
153
represent good explanations for these differences. The US is the worlds major market for biotechnology products and services, offering good opportunities for the development of local rms and attracting many foreign companies. The UK offers good opportunities for the creation of innovative biotech rms, but the business environment cannot sustain these rms through all the stages of growth. In this situation, the UK SMBEs initiate international operations in order to access complementary resources, or to distribute their product overseas. The present study presents importance on three levels: at academic level, it renes the existing internationalisation models, demonstrating that, in a specic industrial sector, the prole of the domestic market signicantly inuences the internationalisation process of SMEs; the management team of the biotech SMEs can use this study to identify the need to internationalise or not in the early stages of the corporate life, depending on the local market prole; and the governmental agencies of various countries can provide specic support for the local SMBEs in order to increase their competitive advantage and provide effective help for their survival and development, depending on their specic need to internationalise. Considering that the focus of this article has been on the biotech SMEs from two countries (UK and US), its results are not directly generalisable to other industries or countries. However, the ndings indicate that the propensity of biotech SMEs to become Born Global and the main reasons of their internationalisation may be inuenced by the characteristics of their domestic market. From this perspective, future studies can be designed to compare the Born Global phenomenon in various countries and industries, in order to verify the validity of these particular research ndings in other industrial and/or national environments. On the other hand, future studies should also consider the differences between high- and low-tech rms in their internationalisation process, a problem already emphasised by some researchers (Boter & Holmquist, 1996; Fillis, 2000).
REFERENCES
Acharya, R. (1998). Conclusions: Biotechnology and European competitiveness. In: J. Senker (Ed.), Biotechnology and competitive advantage (pp. 156163). Cheltenham: Edward Elgar. Acharya, R. (1999). The emergence and growth of biotechnology. Cheltenham: Edward Elgar.
154
Anderson, O. (1993). On the internationalization process of rms: A critical analysis. Journal of International Business Studies, 24(2), 209231. Axinn, C. N., & Matthyssens, P. (2002). Limits of internationalization theories in an unlimited world. International Marketing Review, 19(5), 436449. Bauer, M. (1995). Resistance to new technology and its effects on nuclear power, information technology and biotechnology. In: M. Bauer (Ed.), Resistance to new technology (pp. 1 41). Cambridge: Cambridge University Press. Bilkey, W. J., & Tesar, G. (1977). The export behavior of smaller Wisconsin manufacturing rms. Journal of International Business Studies, 8(1), 9398. Biobox (1998). A primer on biotechnology. http://www.nsta.org/microbix/bioprime.html [accessed March 2000] Boter, H., & Holmquist, C. (1996). Industry characteristics and internationalization processes in small rms. Journal of Business Venturing, 11(6), 471487. Briggs, P. S. (1994). Biotechnology: A commercial evaluation. Richmond: Script Reports. Buckley, P. J., Newbould, G. D., & Thurwell, J. (1979). Going international the foreign direct investment behavior of smaller UK rms. In: L. G. Mattsson & F. Wiedersheim-Paul (Eds), Recent research on the internationalization of business. Proceedings of the Annual Meeting of the European International Business Association, Uppsala. Bullock, W. O., & Dibner, M. D. (1995). The state of the US biotechnology industry. TIBTECH, 13, 463467. Carson, D., Cromie, S., McGowan, P., & Hill, J. (1995). Marketing and entrepreneurship in SMEs: An innovative approach. Hemel Hempstead: Prentice-Hall. Carstairs, R. T., & Welch, L. S. (1982/1983). Licensing and internationalization of smaller companies: Some Australian evidence. Management International Review, 22(3), 3344. Cavusgil, S. T. (1980). On the internationalization process of rms. European Research, 8(6), 273281. Chang, T., & Grub, P. D. (1992). Competitive strategies of Taiwanese PC rms in the internationalization process. Journal of Global Marketing, 6(3), 527. Coviello, N., & Munro, H. (1997). Network relationships and the internationalisation process of the small software rms. International Business Review, 6(4), 361386. Czinkota, M. R. (1982). Export development strategies: U.S. promotion policies. New York: Praeger. Czinkota, M. R., & Ronkainen, I. (1995). International marketing (4th ed.). Forth Worth: The Dryden Press. Daly, P. (1985). The biotechnology business: A strategic analysis. London: Pinter Publishers. Davidson, S. (1998). New European clinical trial regulations are unacceptable to industry. Nature Biotechnology, 16(2), 127. Denis, J., & Depelteau, D. (1985). Market knowledge, diversication, and export expansion. Journal of International Business Studies, 16(3), 7789. Dorabjee, S., Lumley, C. E., & Cartwright, S. (1998). Culture, innovation and successful development of new medicines an exploratory study of the pharmaceutical industry. Leadership & Organization Development Journal, 19(4), 199210. Douglas, S., Craig, S., & Keegan, W. J. (1982). Approaches to assessing international marketing opportunities for small and medium-sized companies. Columbia Journal of World Business, 17(3), 2632. DTI. (2003). UKUS biotech statistics. http://www.dti.gov.uk/biotech03.3.pdf (accessed on March 2006).
155
Ernst and Young. (2000). Evolution. Seventh Annual European Life Sciences Report 2000. Ernst & Young International, London. Euromonitor. (1988). The international biotechnology handbook. London: Euromonitor Publications. Ewing, S. (1996). Biotechnology and the new markets in Europe. Pharmaceutical Form, 6, 1821. Fillis, I. (2001). Small rm internationalisation: An investigative survey and future research directions. Management Decision, 39(9), 767783. Fontes, M., & Coombs, R. (1997). The coincidence of technology and market objectives in the internationalization of new technology-based rms. International Small Business Journal, 15(4), 1435. Gabrielsson, M., Sassi, V., & Darling, J. (2004). Finance strategies of rapid-growing Finnish SMEs: Born Internationals and Born Globals. European Business Review, 16(6), 590604. Garnier, G. (1982). Comparative export behavior of small Canadian rms in the printing and electrical industries. In: R. Czinkota & G. Tesar (Eds), Export management (pp. 113 131). New York: Praeger. Green, D. (1998, April 28). Number of European biotech companies rises. Financial Times, p. 2. Gurau, C., & Ranchhod, A. (1999). The Born Global rms in UK biotechnology. BAM 1999 Conference Proceedings, Manchester. Hakansson, H. (1982). International marketing and purchasing of industrial goods. Chichester: Wiley. Holstein, W. J. (1992, April 13). Little companies, big exports. Business Week, pp. 7072. Johanson, J., & Mattsson, L. G. (1986). International marketing and internationalization processes a network approach. In: P. W. Turnbull & S. J. Paliwoda (Eds), Research in international marketing (pp. 234265). London: Crom Helm. Johanson, J., & Vahlne, J. E. (1977). The internationalisation process of the rm: A model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 80(1), 2332. Johanson, J., & Vahlne, J. E. (1990). The mechanism of internationalisation. International Marketing Review, 7(4), 1124. Johanson, J., & Wiedersheim-Paul, F. (1975). The internationalization of the rm: Four Swedish cases. Journal of Management Studies, 12(3), 305322. Jolly, V. K., Alahuhta, M., & Jeannet, J.-P. (1992). Challenging the incumbents: How high technology start-ups compete globally. Journal of Strategic Change, 1(1), 7182. Jones, M. V. (1999). The internationalization of small high-technology rms. Journal of International Marketing, 7(4), 1541. Keown, B. (1993). European SMEs gain recognition and research funds. Competitive Edge, 1(4), 14. Knight, G., & Cavusgil, S. T. (1996). The Born Global rm: A challenge to traditional internationalization theory. In: S. Tamer Cavusgil (Series Ed.), & Tage Koed Madsen (Volume Ed.), Advances in international marketing (Vol. 8, pp. 1126). London: JAI Press. Knight, G., Madsen, T. K., & Servais, P. (2004). An inquiry into born-global rms in Europe and the USA. International Marketing Review, 21(6), 645665. Kobrin, S. J. (1997). The architecture of globalization. In: J. H. Dunning (Ed.), Governments, globalization, and international business (pp. 146171). New York: Oxford University Press.
156
Lahteenmaki, R., Plant, D., & Michael, A. (1997). Europes enthusiasm for entrepreneurism. Nature Biotechnology, 15(9), 852853. Li, L., Li., D., & Dalgic, T. (2004). The internationalization process of small and medium-sized enterprises: Toward a hybrid model of experiential learning and planning. Management International Review, 44(1), 93116. Madsen, T. K., & Servais, P. (1997). The internationalization of born globals: An evolutionary process? International Business Review, 6(6), 561583. Magnien, E., & de Nettancourt, D. (1993). What drives European biotechnological research? In: E. J. Blakely & K. W. Willoughby (Eds), The management and economic potential of biotechnology, Special Publication of the International Journal of Technology Management, 4758. Mantegazzini, M. C. (1986). The environmental risks from biotechnology. London: Pinter Publishers. McAuley, A. (1999). Entrepreneurial instant exporters in the Scottish arts and crafts sector. Journal of International Marketing, 7(4), 6782. McKelvey, M. D. (1996). Evolutionary innovations. Oxford: Oxford University Press. McKinsey and Co. (1993). Emerging exporters: Australias high value-added manufacturing exporters. Melbourne: Australian Manufacturing Council. Miller, M. M. (1993). Executive insights: The 10 step roadmap to success in foreign markets. Journal of International Marketing, 1(2), 89100. Moen, . (2002). The Born Globals. A new generation of small European exporters. International Marketing Review, 19(2), 156175. Nelkin, D. (1995). Forms of intrusion: Comparing resistance to information technology and biotechnology in the USA. In: M. Bauer (Ed.), Resistance to new technology (pp. 379390). Cambridge: Cambridge University Press. Nordstrom, K. A. (1991). The internationalization process of the rm. Doctoral dissertation. Institute of International Business, Stockholm School of Economics, Stockholm. Norton, R. (1994, August 22). Strategies for the new export boom. Fortune, pp. 124132. Oakey, R., Faulkner, W., Cooper, S., & Walsh, V. (1990). New rms in the biotechnology industry. London: Pinter Publishers. OECD. (1997). Globalization and small and medium-sized enterprises (SMEs). Paris: OECD. Persidis, A. (1996). Building molecular value. Journal of Business Strategy, 17(2), 1821. Persidis, A. (1998). Bioentrepreneurship around the world. Nature Biotechnology, 16(Suppl.), 34. Rasmussan, E. S., Madsen, T. K., & Evangelista, F. (2001). The founding of the Born Global company in Denmark and Australia: Sensemaking and networking. Asia Pacic Journal of Marketing and Logistics, 13(3), 75107. Reid, S. (1981). The decision-maker and export entry and expansion. Journal of International Business Studies, 12(2), 101112. Reid, S. (1983). Firm internationalization, transaction costs and strategic choice. International Marketing Review, 1(2), 4456. Reid, S. (1984). Market expansion and rm internationalisation. In: E. Kaynak (Ed.), International marketing management (pp. 197206). New York: Praeger. Roberts, J. (1999). The internationalization of business service rms: A stages approach. The Service Industries Journal, 19(4), 6888. Robinson, R. (1986). Some new competitive factors in international marketing. In: S. T. Cavusgil (Ed.), Advances in international marketing (Vol. 1., pp. 3045). Greenwich: JAI Press.
157
Root, F. (1987). Entry strategies for international markets. Lexington, MA: Lexington Books. Rothwell, R., & Zegveld, W. (1983). Innovation and small and medium sized rms. London: Pinter Publishing. Roux, E. (1979). The export behavior of small and medium size French rms. In: L. G. Mattsson & F. Wiedersheim-Paul (Eds), Recent research on the internationalization of business. Proceedings of the Annual Meeting of the European International Business Association, Uppsala. Russel, A. M. (1988). The biotechnology revolution: An international perspective. Brighton: Wheatsheaf Books. Saviotti, P. P. (1998). Industrial structure and the dynamics of knowledge generation in biotechnology. In: J. Senker (Ed.), Biotechnology and competitive advantage (pp. 1943). Cheltenham: Edward Elgar. Senker, J. (Ed.). (1998). Biotechnology: The external environment. In: Biotechnology and competitive advantage (pp. 618). Cheltenham: Edward Elgar. Senker, J., Joly, P.-B., & Reinhard, M. (1998). Biotechnology and Europes chemical/pharmaceutical multinationals. In: J. Senker (Ed.), Biotechnology and competitive advantage (pp. 110129). Cheltenham: Edward Elgar. The UK BioCommerce Data Compendium. (2001). The U.K. biotechnology handbook. Richmond: BioCommerce Data. Thorelli, H. (1990). Networks: The gay 90s in international marketing. In: H. Thorelli & S. T. Cavusgil (Eds), International marketing strategy (3rd ed., pp. 7385). New York: Pergamon Press. Turnbull, P. (1987). A challenge to the stages theory of the internationalization process. In: P. Rosson & S. Reid (Eds), Managing export entry and expansion (pp. 2140). New York: Praeger. Varaldo, R. (1987). The internationalization of small- and medium-sized Italian manufacturing rms. In: P. Rosson & S. Reid (Eds), Managing export entry and expansion (pp. 203 222). New York: Praeger. Ward, M. (1995). Should the FDA emulate Europes EMEA? Bio/Technology, 13(7), 636638. Welch, L., & Luostarinen, R. (1988). Internationalization: Evolution of a concept. Journal of General Management, 14(2), 3455. Welch, L., & Luostarinen, R. K. (1993). Inwardoutward connections in internationalization. Journal of International Marketing, 1(1), 4446. Witholt, B. (1999). The European university as a start-up generator. Nature Biotechnology, 17(Suppl.), 78. Yip, G. S., Biscarri, G., & Monti, J. A. (2000). The role of the internationalization process in the performance of newly internationalizing rms. Journal of International Marketing, 8(3), 1035.
MODE OF ENTRY IN SERVICE FIRMS: STRATEGIC VARIABLES AND CHARACTERISTICS OF SERVICES INFLUENCING THE INTERNATIONALIZATION PROCESS
Esther Sanchez Peinado and Jose Pla Barber
ABSTRACT
Despite the importance of the service sector in developed economies and the growth of foreign investments in this sector during the last decade, few studies have undertaken to empirically analyze the factors inuencing entry mode choice. The special characteristics of the service sector increase the complexity of the analysis and, thus, traditional explanations of entry mode choice in manufacturing sectors may need to be complemented by other moderating inuences. Based on 174 entry decisions of service rms, our results suggest the importance of including strategic variables and the specic nature of services to understand a complex phenomenon, which is not always associated just with efciency and valuebased considerations but also with strategic issues and industry characteristics.
159
160
ESTHER SANCHEZ PEINADO AND JOSE PLA BARBER
One of the most critical issues in international market entry strategy is the selection of an appropriate entry mode (Young, Hamill, Wheeler, & Davies, 1989). Much of the literature about entry mode choice is focused exclusively on the process followed by the manufacturing sector (e.g., Hennart, 1991; Kim & Hwang, 1992; Madhok, 1998; Chang & Rosenzweig, 2001, among others) but the question of whether ndings from these studies are applicable to the service sector has not been widely empirically investigated. Nowadays, services have become a driving force in the global economy, representing the most dynamic sector in international commerce. Improvements in technology and reductions in trade barriers have had a unifying effect, making national boundaries less signicant and creating new forms of internationalization. Moreover, rms offering public services (such as energy, telecommunications, y), banks and rms offering information-based services have increased their overseas investments in recent years. Despite the importance of the service sector in world markets and the growth of foreign investments in this area during the last decade, the research, both theoretical and empirical, which analyzes services in an international context is largely fragmented, exploratory and still limited in comparison with research focused on the manufacturing sector (Patterson & Cicic, 1995; Clark, Rajaratnam, & Smith, 1996; Bagchi-Sen & Sen, 1997; Gronroos, 1999; Samiee, 1999; Kennedy, 2004; Lommelen & Matthyssens, 2004). Although there are some important studies that have analyzed entry mode choice in the service context (see, e.g., Goldberg & Saunders, 1981; Ball & Tschoegl, 1982; Agarwal & Ramaswami, 1992; Li & Guisinger, 1992; Erramilli & Rao, 1993; Bouquet, Hebert, & Delios, 2004, among others), they have analyzed some specic sectors and, thus, fail to deal with the heterogeneity problem of the service sector. These facts demand that a more in-depth study of the driving forces behind entry mode choice in the internationalization processes of service rms be undertaken. This paper examines the applicability of the traditional factors determining the entry mode choice of manufacturing rms to service rms. The existing literature has not reached an agreement on which theoretical framework should be used to explain a rms foreign market entry mode. Hill, Hwang, and Kim (1990) have emphasized the need for a unifying theoretical framework within which different factors can be analyzed. We propose an integrative theoretical approach in which not only traditional variables from transaction cost theory (TCT) and organizational capability perspective (OCP) are analyzed, but also strategic variables and some industry characteristics, which have not been considered in conventional studies, are included. In the current competitive environment, entry mode choice is not a
Mode of Entry in Service Firms
161
decision based only on efciency (transaction costs minimization) and valuebased considerations (development of capabilities) and thus existing theory may be insufcient to explain the currently observed behaviors of rms in the international business marketplace (Axinn & Matthyssens, 2002). In dynamic environments, aspects such as strategic motives of internationalization or the rms competitive position in the global environment may also have an impact on the entry mode choice (Hill et al., 1990; Aulakh & Kotabe, 1997; Harzing, 2002). Additionally, the high costs of integration stipulated by economic theories may not be strictly true for many service rms, such as professional services, characterized by low capital intensity (Erramilli & Rao, 1993). For many service rms, the switching costs may be comparatively small because valuable assets rest more in human capital than in physical assets, thus, investment patterns observed in the manufacturing sector could be different in the service sector (Carman & Langeard, 1980). The purpose of this study is to complement prior studies by introducing such new considerations in the entry mode analysis, i.e., the strategic perspective and the specic characteristics of services, in order to contribute to a better understanding of the internationalization processes within the ser vice sector. Additionally, in this paper, the antecedent factors to the entry mode choice of manufacturing rms are analyzed for the service sector context. We aim to provide evidence about which of the underlying principles observed in the manufacturing sector are directly applicable to the service sector and which may be modied by the specic characteristics of services. To achieve these objectives, we have structured the paper as follows. The following sections provide a short overview of the conventional theories explaining the entry mode choice and introduce the need to consider new inuences on the internationalization process of service rms. Next, we present the specic hypotheses of this study, which have been developed from our theoretical framework. Subsequently, we provide detailed information about the research design and methodology used. Finally, we present the empirical results of the study and discuss the most relevant implications for future research.
BACKGROUND
Conventional Theories In explaining the internationalization of business rms, various authors have proposed and used numerous theories, which diverge in their rationale,
162
purpose, and assumptions. For example, some authors have viewed internationalization as a series of static rational decisions (Dunning, 1981, 1988; Teece, 1981; Hennart, 1991). Others see internationalization as an incremental process of increasing international involvement, as rms acquire more experience and gain foreign market knowledge (Johanson & Vahlne, 1977; Johanson & Wiedersheim-Paul, 1975). Although, all these approaches analyze the internationalization as two questions, why do rms internationalize? and how do rms internationalize? (internationalization process), some of them focus predominantly on the rst question, while others focus predominantly on the second question (Weisfelder, 2001). However, the decision about entry mode choice has been widely analyzed from two traditional theories: transaction cost theory (Williamson, 1975; Buckley & Casson, 1976; Anderson & Gatignon, 1986; Anderson & Coughlan, 1987) and organizational capabilities perspective (Kogut & Zander, 1993; Madhok, 1997). In TCT, rms possessing a rent-yielding specic advantage are motivated to enter foreign markets in order to exploit this advantage in the most efcient manner. Under the assumption of opportunistic behavior and bounded rationality by economic agents, market imperfections increase the associated costs of transacting with a partner, which results in a preference for internalizing the transaction within the rms own boundaries (Buckley & Casson, 1976). TCT stresses the efciency of using high control modes when foreign market attractiveness is high, under conditions of high cultural distance and country risk markets and in the presence of intangible assets (Anderson & Gatignon, 1986; Erramilli & Rao, 1993; Luo, 2001). On the other hand, the OCP broadens the focus from minimizing the transaction costs to also incorporate the managing of value, inherent in a rms knowledge base (Kogut & Zander, 1993). The key issue in the entry mode choice is the compatibility between the rms existing capabilities and those needed to be successful in a particular market (Madhok, 1998). As Madhok (1997) proposed, an operation motivated by value-based considerations (the development of capabilities to create future value) will result in a greater proclivity toward collaboration. According to theory, rm-specic capabilities are commonly linked to rm size, international experience, and tacit know-how. Larger and more experienced rms have the condence and competence to manage the uncertainties of operating in international markets, and the tacitness of know-how increases the transaction costs but it also limits the transferability of the know-how to another rm without loss of value (Kogut & Zander, 1993; Luo, 2001). These circumstances increase the efciency of resource utilization and the effectiveness of its transfer in-house (Madhok, 1998).
163
A Framework for Service Firms From both theoretical perspectives, TCT and OCP, empirical studies have tried to analyze the impact that the transaction and rms organizational capability factors have on entry mode choice. These theories consider market-contracting or low control modes as the default choice for situations of low-asset specicity and as an alternative to complement rm capabilities through knowledge gained from the experience of other rms (Gatignon & Anderson, 1988; Madhok, 1998). However, in business practice, rms often have other additional motives to integrate such as to retain control to better coordinate strategies in multinational corporations (Hill et al., 1990), to extend market power by entering new markets (Teece, 1981), to exploit market knowledge following the national clients or competitors to foreign countries, among others (Erramilli & Rao, 1990; Li & Guisinger, 1992). These incentives to integrate should be incorporated into conventional theories in order to reect the current competitive conditions and to manage the multiplicity of pressures that manufacturing and service rms face in their businesses. In fact, Dunning (1993) highlighted the need to introduce the strategic dimension into the analysis of determinants of international production because this has become an important component of competitive success in a world characterized by the growing plurality of forms of globalization and the growth of cross-border collaborative alliances. Nowadays, multinational rms are competing in global rather than national markets and when global competitive interdependence exists, the actions taken by a rm in one market often have repercussions in other national markets (Kim & Hwang, 1992). Moreover, strategic motivations such as setting up a strategic outpost for future expansion, setting up a global sourcing site, achieving economies of scale by concentrating the important activities in a limited number of locations, etc., favor high control types of entry modes (Harzing, 2002). This is even more important in the analysis of the globalization of services because some service industries have shown different internationalization patterns depending on their motives for entry (Dunning, 1993). Therefore, since new competitive conditions can determine entry mode choices due to strategic considerations, in this paper we include the analysis of these variables, which have not been considered in studies focused on conventional theories (TCT and OCP). Moreover, this study analyzes the service sector. Some researchers suggest that service rms differ from manufacturing rms and face unique challenges in their foreign market entry and expansion process (Carman & Langeard, 1980; Erramilli & Rao, 1993; Gronroos, 1999). Alternatively,
164
others contend that many of the underlying principles observed in the manufacturing sector are directly applicable to the service context (Boddewyn, Halbrich, & Perry, 1986; Katrishen & Scordis, 1998). We contend that some of the underlying principles observed in the manufacturing sector are directly applicable to the service sector but some others may be moderated by the specic characteristics of services. In the literature about services, the special characteristics addressed as the factors that distinguish services from goods are intangibility, inseparability, heterogeneity, and perishability. The intangibility means that services are experiences which cannot be touched, seen, transported, or lifted (Dahringer, 1991). Moreover, when a service is performed, no ownership is transferred from the seller to the buyer, so the latter is merely buying the right to a service process (Palmer, 1995). The intangibility of services also presents some challenges in going abroad. For any company trying to market services internationally, this characteristic increases customer risk perceptions and makes more difcult to asses service quality than for manufactured goods (Edgett & Parkinsson, 1993; Winsted & Patterson, 1998). As Sarathy (1994) points out, the subjective nature of service quality perception implies that service process may need some adaptation to attract customers in foreign markets. Another distinctive feature of services is the inseparability of production from consumption, i.e., there is a simultaneous association between the acts of production, exchange, and consumption, whereas, in the case of goods, these are normally separate and discrete activities. Many service rms need to have a local presence in a foreign market because they require direct involvement from the client (Erramilli, 1990) and high levels of customization to meet the needs of local clients. In such situation, the cultural distance between home country and host country becomes a critical aspect that must be analyzed before crossing national boundaries (Lovelock, 1999). Services are also perishable, i.e., they cannot be stored for use after production like manufactured goods. Perishability implies the need of direct contact with the client and presents special challenges for balancing supply and demand (Nicolaud, 1989). Hence, in the international expansion, rms may be forced to operate with excess capacity at each international market (Sarathy, 1994). And nally, heterogeneity means that services are difcult to standardize so they are never the same from one consumption experience to another (Dahringer, 1991). The important role of personnel in providing services produces a high variability in the performance of services. This fact makes harder to control the output of service rms than it is to control the output of those producing tangible products (Edgett & Parkinsson, 1993; Lovelock,
165
1992). This characteristic also increases the difculty of achieving success with new services because variability in performance implies that past performance may not be a valid indicator of future success (Campbell & Verbeke, 1994). Therefore, it is important that all service providers are trained to provide equivalent delivery and rms should take into account that cultural distance affects the nature of the workforce employed and the standardization of the service delivery (Sarathy, 1994). In the literature about entry modes, two characteristics have been acknowledged by researchers as determinants of the different responses in the internationalization process of service rms: the level of capital intensity and the degree of customization (Erramilli & Rao, 1993; Brouthers & Brouthers, 2003). First, the level of capital intensity is different in manufacturing and service sectors. Manufacturing frequently requires large commitments for plants and equipment. However, the level of capital intensity varies among different services within the service sector. Some service rms are people intensive (e.g., consulting, legal services, accounting, advertising, y) and thus high control entry modes involve low expense (Erramilli & Rao, 1993). Given similar risk perceptions, these service rms will respond quite differently from manufacturing ones because differences in switching costs allow service rms to relocate activities more easily if risk conditions become too great. On the other hand, some service rms are capital intensive (e.g., telecommunications, energy, airlines, hotels, y) and they are more likely to show the same investment patterns as the manufacturing sector. The establishment of a new subsidiary abroad entails large xed-investments and, hence, rms will prefer exibility (low control modes) to enter into countries with changing environmental conditions. Second, the heterogeneous nature of services is closely linked to variability in production standards and usually requires that service providers tightly control the quality of the service production process (Sasser, Olsen, & Wyckoff, 1978). Variability depends on the extent to which services can be customized to meet the specic needs of individual customers. Services that are produced for a large number of customers simultaneously may offer little scope for customization. However, services that are closely associated with employees who are embedded in their own cultural and social contexts are difcult to standardize and produce in the same way abroad as in the home market (Palmer, 1995). To offer these services: (i) employees must possess high levels of analytical and technological skills in order to cope with customers needs; and (ii) service providers must control the quality of the service production process (Sasser et al., 1978). The need for local adaptation to customer needs, the importance of controlling the rms quality standards,
166
NEW INFLUENCES
Strategic considerations International strategy Defensive vs. Offensive motives Asset exploration vs. Asset exploitation
Entry mode analysis
Industry characteristics Capital intensity Degree of customization
Control variables Transaction cost variables Organizational capabilities variables
Fig. 1.
Determinant Factors of Entry Mode Analysis.
and the difculty in transferring know-how to local partners to offer services with a high degree of customization will increase the propensity to use highintegration entry modes in culturally distant markets. Both specic characteristics of services, the degree of capital intensity and customization, may moderate the relationship between rms investment risk perception (country risk and cultural distance) and the level of control and resource commitments in the host country. Inherent differences in manufacturing and service rms necessitate a context-specic approach, which extends conventional theories, to understand a service rms internationalization. In this paper, we will focus on the examination of these new inuences whereas traditional variables, analyzed in prior studies focused on transaction cost and resource-based theories, will be considered as control variables in the empirical study. Fig. 1 summarizes the theoretical framework that will be used to develop the hypotheses.
HYPOTHESES DEVELOPMENT
Strategic Factors Type of International Strategy From the integration-responsiveness paradigm (Porter, 1986; Prahalad & Doz, 1987), we can distinguish two types of international strategies: global
167
and multidomestic strategies. A global strategy is based upon the centralization of the main assets, resources, and responsibilities in the head ofce. These companies think in terms of creating products for a world market and manufacturing/servicing them on a global scale in a few highly efcient plants located in countries with cost advantages. Therefore, these plants may not necessarily be located at the corporate center. This approach is focused on efciency and requires central coordination and control. On the other hand, a multidomestic strategy emphasizes the differences between national markets and rms compete predominantly on a domestic level (Harzing, 2002). The subsidiaries become more independent and their managers become more host-country oriented. Products or services are differentiated to meet different local demands and policies are differentiated to conform to different governmental and market demands (Porter, 1986). Hill et al. (1990) and Kim and Hwang (1992) suggested that rms pursuing a global strategy tended to adopt high control modes, due to the necessity of tight coordination, possible synergy effects, and transfers of assets between units. Internalization facilitates centralized decision-making to manage activity interdependencies because a rms competitive position in one market is affected by its competitive position in other national markets (Aulakh & Kotabe, 1997). However, rms pursuing a multidomestic strategy should select low control modes because they are more focused on the costsresources of the relationship and adaptation to local conditions than on the level of control. Since personal selling plays an important part in the offer of services, service providers are generally encouraged to take great care in selecting and training contact personnel. The international delivery of services will require the use of sophisticated recruitment and training techniques to ensure and control the selection and development of service providers whose competence will lie in projecting the right image. This need will be particularly acute when the rm attempts to follow a global approach. Therefore, we hypothesize that H1. Service rms pursuing a global strategy are more likely to select high control and resource commitment modes. Defensive vs. Offensive Strategic Motives of Entry Traditionally, foreign direct investment (FDI) has been explained by market seeking strategic motivations which transfer and exploit rms competitive advantages and therefore overcome the liability of foreignness (Caves, 1971; Hymer, 1976). From this perspective, some authors have distinguished between defensive motives, based on the protection of current
168
markets, and offensive motives, based on the exploitation of new markets (Dunning, 1995). Defensive motives most commonly cited in the literature are following the national client and following the competitors or leaders (Li & Guisinger, 1992; Dunning, 1995). Many service rms follow internationalizing manufacturing companies abroad in order to offer their services to these companies also in overseas markets. Client followers can be expected to select a high integration mode in order to protect their competitive advantage based on the knowledge of specic customer needs. Banking, insurance, consulting, legal services, accounting, and other professional service industries are examples of this type of internationalization motive (Goldberg & Sanders, 1981; Ball & Tschoegl, 1982; Dunning, 1993). They initially followed their goodsproducing clients to the countries where they set up their factories in order to gather necessary information and prepare for and facilitate their home clients expansion in the new market. On the other hand, in oligopolistic markets, competitors react to overseas expansion of service rms in order to proactively achieve a global market power position and head off potential competitors (Terpstra & Yu, 1988). Wholly owned subsidiaries (WOS) allow rms to give a faster response to competitors actions than collaborative modes, so service rms are more likely to select a full-control mode. This international pattern has been empirically demonstrated in some studies focused on service rms (Erramilli & Rao, 1990; Li & Guisinger, 1992). Offensive motives, however, are related to the purpose of serving new foreign clients, seeking growing market potential countries, or locating in specic geographical areas (e.g., banks locating in major nancial districts). Under such circumstances, service rms lack market knowledge and knowledge of local needs. Thus, it is necessary to balance the risks and benets of being the rst mover, so rms are more likely to select collaborative agreements such as contractual agreements or joint ventures in order to obtain external resources, gain access to local customers, and overcome the liability of foreignness (Erramilli & Rao, 1990; Dunning, 1993). According to these statements H2. More defensive entry motives will increase the likelihood of service rms choosing high control and resource commitment modes. Exploitation vs. Exploration Strategic Motives of Entry Recently, researchers have suggested a necessary dynamic balance between exploitation and the development of capabilities for the continued success of
169
the rm (March, 1991; Levinthal & March, 1993; Chang, 1995; Makino, Lau, & Yeh, 2002). Traditionally, FDI is viewed as the transfer of a rms proprietary assets across borders (asset exploitation). However, from the asset-seeking perspective, rms engage in direct investment when they intend to acquire strategic assets (technology, marketing, management expertise y) available in host markets. The asset-seeking motivation consists of experimentation of new alternatives, which allows rms to enhance competitive advantages through complementary assets acquired in host markets (Vermeulen & Barkema, 2001). Some resources, such as capital or labor resources, can easily be obtained through markets. But, other resources such as host market knowledge, local contacts, or technology would be costly to obtain through replication or acquisition and as such rms are more likely to collaborate with local rms (Hennart, 1991). Therefore H3. More asset-seeking entry motives will decrease the likelihood of service rms choosing high control and resource commitment modes. Specic Characteristics of Services and Risk Investment Variables Capital Intensity and Country Risk An important source of uncertainty identied in the international business context is associated with the political instability and economic uctuations in foreign countries. Country risk determines both the need for local knowledge and the exposure of assets. Therefore, rms facing high country risk tend to seek local knowledge, retain exibility, and shift the risk to local rms through agreements or joint ventures (Aulakh & Kotabe, 1997; Erramilli & Rao, 1993; Kim & Hwang, 1992). This inverse relationship between country risk and control proposed by conventional theories has been also supported by studies focusing on service rms (see, e.g., Li & Guisinger, 1992; Agarwal & Ramaswami, 1992; Contractor & Kundu, 1998). However, investment patterns in volatile environments may be moderated by some specic factors related to the specic characteristics of services. Although service rms are generally less capital intensive than manufacturing rms, capital intensity varies across service industries: from low levels in consulting, architecture, engineering rms, etc., to high levels in hotel, airline, telecommunication, electricity, and energy rms. Ownership of high capital-intensive service facilities entails large-scale investments and thus, such service rms will prefer to minimize the risk exposure and resource commitments in high country risk markets. However, the low resource
170
commitments and switching costs for less capital-intensive services may induce rms to invest in unstable countries where competitive conditions are more favorable than in other international markets and where it is easier to overcome local rms advantages and achieve a better market share. Thus, we propose the main effect of country risk, supported by conventional theories and most commonly observed in empirical studies, and the moderating inuence of a services capital intensity over such a relationship to be: H4a. Country risk in the host market will decrease the likelihood of service rms choosing high control and resource commitment modes. H4b. The inverse relationship, between country risk and a service rms propensity for high control and resource commitment modes, will become weaker with the decreasing capital intensity of services. Degree of Customization and Cultural Distance Several studies have considered that the cultural distance between the home country and the intended foreign market is a powerful determinant of entry mode choice. However, there is some controversy over its exact inuence. Traditional TCT arguments consider that rms minimize the high information costs associated with operating in culturally unfamiliar countries by using collaborative modes of entry (Gatignon & Anderson, 1988). Thus, foreign investors share the risk with local partners and avoid costly mistakes in the new market. Empirical evidence supporting a positive relationship between cultural distance and the desire to establish collaborative modes has been found in several studies focusing on the service sector (among others, Erramilli, 1991; Li & Guisinger, 1992; Fladmoe-Lindquist & Jacque, 1995). However, recent research suggests that high cultural distance could provide investors with the incentive to internalize, due to difculties in resource transfer across rm boundaries and the expense of incurring ex ante costs of evaluating the local collaborator (Anand & Delios, 1997). This fact is especially relevant in most services characterized by a low degree of separability and a high degree of customization (Domke-Damonte, 2000). Customized businesses have to conduct face-to-face transactions with their customers, creating signicant risks for the rm. Specically, the performance of employees who deal directly with customers is important for the maintenance of a rms quality standards (and the resulting competitive advantage). If rms use local partners in countries with a high cultural distance, the quality of a service may vary, raising an overall problem of consistency for the foreign investor (Bouquet et al., 2004). Furthermore, the cross-cultural transfer of services requires communicating with foreign
171
employees whose work ethics are embedded in different languages and cultural traditions (Fladmoe-Lindquist & Jacque, 1995). As such, keeping transactions within the context of the rms organizational boundaries could solve this problem. Given the evidence that highly customized service rms favor high control modes, we hypothesize rst the conventional relationship which proposes that information costs in unfamiliar foreign cultures will diminish the rms desire for high control and resource commitment modes. However, the need for local adaptation to customer needs, the importance of controlling the rms quality standards, and the difculty in transferring know-how to local partners to offer services with a high degree of customization will increase the propensity to use high integration entry modes. Hence, H5a. Cultural distance between home and host market will decrease the likelihood of service rms choosing high control and resource commitment modes. H5b. The inverse relationship, between cultural distance and service rms propensity for high control and resource commitment modes, will become weaker with an increasing degree of customization of services.
METHODS
Sample Dun and Bradstreet (2002) compiled an international database providing information on Spanish companies with subsidiaries outside Spain in the year 2002. This amounted to 660 multinational service companies. The survey instrument consisted of an extensive mail questionnaire, which was pretested through personal interviews with Spanish executives responsible for international operations and with academics specialized in international management. The questionnaire was mailed to senior-level managers who were most likely to be involved in the market entry decision process in their rms, including CEOs and directors in charge of international operations. Each respondent was asked to provide data on up to two foreign market entry decisions in which he/she was involved or had enough information about to answer. Two weeks later, each rm received a phone call in order to encourage managers to participate in the study. Finally, a reminder was sent to non-respondents with a copy of the questionnaire. From October 2002 until February 2003, we received 113 questionnaires and rms
172
Table 1.
Sectors Transport Distribution Public services Financial services Commercial services (standardized) Professional services (customized) Total
Service Sectors.
Percentage 17.24 23.56 12.06 12.06 25.28 9.8 100
Table 2.
Firm Characteristics (N 113)
Sample Characteristics.
Mean 1.92 1.27 871.32 14.31 3.50 9.20 4.98 Standard Deviation 0.98 0.64 2,478.41 17.26 4.14 14.43 9.54 Median 2 1 105 8 2 2.50 0
Foreign sales Foreign assetsa Number of employees Number of foreign countries (exports) Number of foreign countries (investments) Export experience (number of years) Investment experience (number of years)
a
Intervals: value 1 (o25%), value 2 (2550%), value 3 (5075%), value 4 (o75%).
provided data on a total of 174 entry decisions that were deemed usable for our analyses, representing a wide variety of service industries (see Table 1). The response rate of 17% compares favorably with rates reported in other surveys involving CEOs (Samiee & Walters, 1991; Chang & Taylor, 1999). Each respondent provided detailed data on at least one decision and some of them provided information about two entry decisions. We did not provide a list of markets, so respondents chose countries in which they were operating. The highest number of entries was in the European Union (48.27%), followed by Latin America (34.48%), OCDE-Non European Union (9.19%), and the rest of the world (8.06%). This investment pattern has also been shown by Spanish statistics. More information about rm characteristics is listed in Table 2. Questionnaires were analyzed using the time trend procedure proposed by Armstrong and Overton (1977). In our paper, the midpoint of the data collection period (October 2002 until February 2003) was used as the cutoff
173
point for distinguishing between early and late respondents. 62% of the responses (70 out of 113) were from early respondents and the remaining 38% were from late respondents (43 out of 113). To ensure that the early and late respondents did not systematically differ, these two groups of respondents were compared on their demographic data, including foreign sales, foreign assets, number of employees, number of countries in terms of exports, number of countries in terms of subsidiaries, international experience in exports, and international experience in investments. We used independent samples t-tests to check for equality of means. Analysis indicated no signicant differences in the variables of interest between late and early respondents. Additionally, responding rms were compared to a random sample of 20 non-respondents regarding size (sales volume) and experience (years since foundation). No signicant differences were found (po0.05), providing no evidence for non-response bias. Finally, those variables from the survey responses were cross-checked against company reports and published data, where possible. A high degree of correspondence between published data and survey responses was found, supporting the veracity of the survey responses. Dependent Variable Measure The hypothesized relationships were examined using two analyses. In the rst one, the study is focused on the inuence of independent variables over ownership control decisions. The dependent variable, degree of control and resource commitments, was represented by a dichotomous variable, coded 1 for the full-ownership control mode (Greeneld investments and full acquisitions) and 0 for the shared-ownership control mode (contractual agreements, partial acquisitions, and joint ventures). Entries were made up of 67 shared-control modes (38.5%) and 107 full-ownership modes (61.5%). In the second analysis, the study focused on the inuence of the same independent variables over specic entry modes. The dependent variable represents the three options included in our study: 1 for contractual agreements (a total of 19.5% of entries), 2 for shared-control subsidiaries (19%), and 3 for WOS (61.5%). The parameters estimated are interpreted in reference to the wholly owned control subsidiaries. A WOS is a 100% investment in which decisions are made solely by the investor. Thus, the parent company has full ownership and sole responsibility for the management of operations. A joint venture entails direct investment in the target country and is dened as a joint partnership in which
174
major decisionmaking is shared with the foreign partner. Partners share the ownership, management, and risk of the new venture. Thus, joint ventures entail a resource commitment that is greater than a licensing arrangement, but less than a WOS. Contractual agreements can be broadly dened as coalignments between two or more rms in which the partners acquire production technologies and marketing skills from each other with little equity ownership. Hence, contractual agreements involve lower resource commitments than joint ventures and WOS.
Independent Variable Measures International Strategy The type of international strategy has been constructed from Harzings (2002) measure. We asked respondents to assess the following statements (1 strongly disagree, 5 strongly agree): (a) rms desire to achieve economies of scale by concentrating the important activities in a limited number of locations, (b) competitive position is dened in worldwide terms and markets are closely linked and interconnected, (c) each subsidiary competes on a domestic level as markets are too different (reverse) and (d) rm responds to national differences by adapting products to the local market (reverse) (Cronbachs alpha 0.6959). We constructed a composite index with these items. High values indicate global strategic approaches. Table 3 provides detailed information about the goodness of t indices. Strategic Motives Strategic motives related to defensive motives and offensive motives was measured using a multiple-item scale based on Weinsteins (1977) scale, Erramilli and Rao (1990), Li and Guisinger (1992) and Kim and Hwang (1992). We asked about managers perceptions about the importance of the following motives (1 very low, 5 very high): (a) three items related to defensive motives: to overcome the rapid overseas expansion of other service rms, to follow the clients, to follow the competitors (Cronbachs alpha 0.7537) and (b) three items related to offensive motives: to take advantage of a market with a large and growing potential, to establish the company name in markets that will be important in the future, and to capture new clients (Cronbachs alpha 0.8539).
175
Table 3.
Goodness of Fit Indices.
Strategic Variables (Independent Variables) International strategy Goodness of t indices: BENTLERBONETT NORMED FIT INDEX 0.956; BENTLERBONETT NONNORMED FIT INDEX 0.937; COMPARATIVE FIT INDEX 0.969; LISREL GFI 0.985; LISREL AGFI 0.950 Motives Market Seeking: defensive vs. offensive motives We conducted a two-factor conrmatory factor analysis. Items were signicant (t-value >1.96) and the analysis conrms the existence of two groups of motives: defensive and offensive motives. Goodness of t indices are: BENTLERBONETT NORMED FIT INDEX 0.948; BENTLERBONETT NONNORMED FIT INDEX 0.906; COMPARATIVE FIT INDEX 0.956; LISREL GFI 0.959; LISREL AGFI 0.878; STANDARDIZED RMR 0.042 Transaction Cost Variables (Control Variables) Marketing intensity Goodness of t indices: BENTLER-BONETT NORMED FIT INDEX 0.940; BENTLER-BONETT NONNORMED FIT INDEX 0.832; COMPARATIVE FIT INDEX 0.944; LISREL GFI 0.970; LISREL AGFI 0.822 Firms Organisational Capabilities (Control Variables) Tacit know-how Goodness of t indices: BENTLERBONETT NORMED FIT INDEX 0.947; BENTLERBONETT NONNORMED FIT INDEX 0.902; COMPARATIVE FIT INDEX 0.951; LISREL GFI 0.952; LISREL AGFI 0.840
Two-factor conrmatory factor analysis conrms the existence of two groups of motives. With an aim to assessing the strategic nature of the entry, offensive motives were reverse coded and then we constructed an index with both defensive and offensive motives (high values indicate more defensive motives). Table 3 provides detailed information about the goodness of t indices of these strategic motives. Asset seeking and asset exploitation motives were captured from Chang (1995) and Makino et al. (2002). We asked about managers perceptions about the importance of the following statements (1 very low, 5 very high): (a) rm orientation toward assets and capabilities seeking in the entry decision and (b) rm orientation toward asset exploitation in the entry decision. The exploitation motive was reverse coded and then we constructed an index with both exploitation and exploration items (high values indicate exploration motives).
176
Risk Investment Variables and the Characteristics of Services Country Risk This variable was derived from the host country risk index published in Euromoney the year before each entry. This publication presents annual ratings of countries based on composite measures of both political and economic risks. The index varies from 0 to 100: the value 0 represents instability of political and economic conditions, while the value 100 represents stability of such conditions. This measure has been used by previous studies (see, e.g., Delios & Beamish, 1999; Shrader, Oviatt, & McDougall, 2000). Cultural Distance This variable was computed according to the composite index used by Kogut and Sing (1988) and data contained in Hofstede (1980). This index measures the deviation along each of the four cultural dimensions identied by Hofstede (1980) (uncertainty avoidance, individuality, power distance, and masculinityfemininity) from the score of a given focal country for each country. The deviations were corrected for differences in the variances of each dimension and arithmetically averaged. Cultural distance between the country of origin (Spain) and host country (j) is hence calculated as: " 2 # 1 X I iSpain I ij CDSpain=j 4 Vi where Iij is the index value for the cultural dimension i of country j, IiSpain the index value for the cultural dimension of Spain, Vi the variance of the index dimension i, and CDSpain/j the cultural difference of country j from Spain. Specic Characteristics of the Service We construct two indices which moderate the inuence of the risk investment variables (country risk and cultural distance) on entry mode decisions: one index related to the capital intensity of the service sector (capital intensity index (CII)), and the other one related to the degree of customization of the service (degree of customization index (DCI)). Both indices were constructed from managers perceptions about the characteristics of the service. Each respondent had to assess (a) Capital intensity: Two items that capture the degree to which the service is provided by machines and the degree to which the service is provided
177
by people (1 very low; 5 very high). The people-intensive service item was reverse coded and then we constructed an index with both items (high values indicate capital-intensive services). This variable was constructed based on Erramilli and Rao (1993), Erramilli and DSouza (1995) and Pan and Tse (2000). (b) Degree of customization: Two items that represent the degree to which rms offer different services adapted to customers preferences or they offer similar services to customers (1 very low; 5 very high). This last item was reverse coded and then we constructed an index with both items (high values indicate customized services). This variable is based on prior studies (see, e.g., Erramilli & Rao, 1993). Control Variables We control the inuence of traditional variables which have been considered in previous studies based on TCT and OCP. From TCT, we control the inuence of market potential and intangible assets. Country risk and cultural distance were examined in the independent variables section because we have hypothesized the main effect of these variables and the moderating inuence of service characteristics. Based on prior studies (see, e.g., Agarwal & Ramaswami, 1992; Erramilli, 1992), market potential was measured as a single-item scale based on managers perceptions about the potential growth of the new market (1 none; 5 high). On the other hand, following Kim and Hwang (1992), we distinguish two variables to capture a rms intangible assets: (a) R+D intensity: A single-item scale based on managers perceptions about R+D expenses. (b) Marketing intensity: A three-item scale based on managers perceptions about: (i) rm reputation with respect to design and quality; (ii) international recognition of brand name; and (iii) advertising investment (1 very low; 5 very high) (Cronbachs alpha 0.7486). Detailed information about goodness of t indices of this variable is presented in Table 3. From OCP, we control the impact of rm size, international experience, and tacit know-how. Firm size was measured as intervals of a rms sales volume the year before the entry. Prior research has also used the sales volume as an indicator of rm size (see, e.g., Erramilli, 1991; Agarwal & Ramaswami, 1992; Li & Guisinger, 1992). Size is a categorical variable that takes three values: 1 (small rms), 2 (medium rms), and 3 (large rms).
178
These categories have been constructed according to managers responses. We asked them to indicate the rms sales volume for the year of the entry. The difculty in remembering the exact sales volume led us to ask this question in terms of intervals, following European Union Commission recommendations: a sales volume lower than h7 million (small), between h7 and 40 million (medium), higher than h40 million (large). Results are interpreted in reference to the last interval. International experience was measured as the number of years from the rst investment abroad, also based on prior studies (see, e.g., Harzing, 2002; Domke-Damonte, 2000; Fladmoe-Lindquist & Jacque, 1995). And, nally, tacit know-how is a four-item scale derived from Kim and Hwang (1992) and Erramilli, Agarwal, and Dev (2002). The construct was based on managers perceptions about (a) the difculty in transferring capabilities and knowledge, (b) the difculty in assessing the proper price of the service, (c) the difculty in copying the skills and knowledge, and (d) the difculty in understanding the know-how. (1 very low, 5 very high) (Cronbachs alpha 0.8396). Detailed information about the goodness of t indices of this variable is presented in Table 3.
RESULTS
Prior to running the statistical analyses, the correlation matrix was examined. Most of the correlations among the variables are small. Further, the variance-ination factor (VIF) reveals that most of these are close to 1. The largest VIF value is 3.058, which is well below the cutoff at 10 (Hair, Anderson, Tatham, & Black, 1999). This evidence reduces concerns about multicollinearity problems. Table 4 shows the correlation matrix and some descriptive statistics. A logistic regression was carried out to analyze the inuence of the independent variables in each of the groups dened by the rst dependent variable (full-ownership control vs. shared-control). The model estimates the probability of choosing the entry modes dened by the dependent variable. The model was specied as follows: expz=1 expz; where z is a linear function of the independent variables. We analyzed the sign of the coefcients that were signicant. A positive coefcient indicates that the independent variable increases the probability of choosing full-ownership control, while a negative sign suggests a preference for shared-control modes. One-tailed tests were used because they are more appropriate to assess directionality.
Table 4.
Mean Standard Deviation 1. Market potential 3.87 2. Country risk 72.42 3. Cultural distance 0.9584 4. R+D intensity 2.26 5. Marketing intensity 3.12 6. International experience 5.92 7. Size 1.5311 8. Tacit know-how 3.02 9. International strategy 3.13 10. Offensivedefensive motives 3.21 11. Exploitationexploration motives 2.54 12. Capital intensity index (CII) 2.84 13. Degree customization index (DCI) 2.94 1.04 24.44 0.7066 1.17 0.97 11.97 0.7492 0.94 1.01 1.14 1.39 1.02 1.41 1 2
Correlation Matrix.
3 4 5 6 7 8 9 10 11 12 13
1.000 0.090 1.000 0.042 0.111 1.000 0.058 0.016 0.027 1.000 0.004 0.214 0.129 0.228 1.000 0.019 0.079 0.020 0.062 0.049 1.000 0.025 0.047 0.089 0.066 0.153 0.258 1.000 0.536 0.010 0.012 1.000 0.027 0.100 0.074 0.224 0.170 0.437 1.000 0.092 0.103 0.139* 0.102 0.524 0.028 0.009 0.095 0.045 0.588 0.547 0.517 1.000 0.080 0.081 0.198 0.016 0.101 0.005 0.090 0.129 0.103 0.165 1.000 0.106 0.219 0.153 0.030 0.399 0.269 0.235 0.133 1.000 0.030 0.167 0.029 0.119 0.255 0.095 0.451 0.316 0.351 0.083 0.376 1.000 0.128 0.088 0.037 0.100 0.286 0.145 0.127
0.01 level of signicance. 0.05 level of signicance.
179
180
Subsequently, we analyzed the impact of the independent variables on the probability that each of the three-entry mode options would be chosen (WOS, shared-control subsidiary, and contractual agreements). We carried out a multinomial regression and took WOS as reference (greeneld investments and full acquisitions). A positive coefcient indicates that the probability of adopting an entry mode rather than WOS increases as increments of the independent variables occur. This analysis allows us to conrm the results obtained from the logistic regression and to assess the consistency of the results by means of different statistical methods, which enhances the robustness of the conclusions. Both regressions introduce two alternative models. In order to examine the contribution of factors traditionally analyzed by TCT (market potential, country risk, cultural distance, and intangible assets) and the contribution of those traditionally analyzed by OCP (size, international experience, and tacit know-how), these variables were entered in the rst regression. The second regression also includes the inuence of strategic variables and the moderating inuences of the characteristics of services on the relationship between risk investment variables (country risk and cultural distance) and entry mode decisions. In general, the models present satisfactory indicators of signicance. However, the addition of the strategic perspective and the characteristics of services to the regression along with variables from conventional theories resulted in a signicant improvement in Chi-square at po0.01 and classies a higher number of cases (see Tables 5 and 6). These results indicate that these new inuences add to the predictive ability of conventional theories. As such, we shall extend our comments on the results of the models to include the strategic variables and the characteristics of services. The inclusion of strategic variables allows us to extend the explanations of entry mode choice to a new research area. Our results from the nal binary logistic regression conrm the high explanatory power of these variables. As Table 5 Model 2 shows, there is empirical support for H1, H2, and H3. As we predicted, a global international strategy in foreign markets (po0.001) and a rms strategic motivations related to the following of clients or competitors (po0.10) and to asset exploitation (po0.10) increase the probability of selecting a full-ownership control mode. Results from the multinomial regression, summarized in Table 6 Model 4, allow us to conrm the relationship between international strategy and entry modes. However, exploration motives are only associated with shared control when equity and non-equity modes are compared, and defensive motives are associated with full control when equity modes are compared.
181
Table 5.
Variables
Logistic Regression.
Model 1 (Control Variables: TCP and OCP) 10.325 (2.037) 0.928 (0.266) 0.004 (0.010) 0.452 (0.348) 0.055 (0.227) 0.857 (0.311) 0.016 (0.020) 0.230 (0.633) 0.284 (0.758) 1.739 (0.347) Model 2 (Complete Model) 7.268 (3.481) 0.988 (0.333) 0.078y (0.044) 2.803 (1.289) 0.155 (0.324) 0.728 (0.518) 0.046 (0.028) 0.801 (1.001) 0.250 (1.130) 1.584 (0.549) 1.478 (0.454) 0.618y (0.369) 0.399y (0.236) 2.199 (0.992) 0.343 (0.491) 0.028y (0.015) 0.766y (0.444) 152.637 72.565 90.5
Constant Market potential Country risk (H4a) Cultural distance (H5a) R+D intensity Marketing intensity International experience Size (1)+ Size (2)+ Tacit know-how International strategy (H1) Offensivedefensive motives (H2) Exploitationexploration motives (H3) Capital intensity Degree of customization Country risk CII (H4b) Cultural distance DCI (H5b) w2 2 log likelihood Overall classication rate (%)
113.160 113.010 85.3
Note: Standard errors in parenthesis; dependent variable: shared control (0); full control (1); CII, capital intensity index; DCI, degree of customization index; + reference category, the highest interval. po0.001; po0.01; po0.05; y po0.10 (one-tailed).
Additionally, the empirical results from the logistic regression show that specic characteristics of services moderate the relationships of investment risk variables (country risk and cultural distance) and entry modes (see Table 5 Model 2). The main effect of country risk is statistically signicant (po0.1) but contrary to the relationship proposed by conventional theories and thus contrary to our H4a. Considering that high values of the country risk index represent stability of political and economic conditions and low values represent instability conditions, the negative sign of the main effect indicates that service rms entering unstable environments (low values of the country risk index) prefer full-control modes (a high value of the dependent variable). The main effect of capital intensity is also signicant
182
Table 6.
Variables
Multinomial Regression.
Model 4: Complete Model Model 4a CA vs. WOS Model 4b SCS vs. WOS
Model 3: TCT and OCP Variables Model 3a CA vs. WOS Model 3b SCS vs. WOS 11.957 (2.261) 1.129 (0.303) 0.006 (0.011) 0.528 (0.422) 0.123 (0.308) 0.706y (0.402) 0.013 (0.024) 0.704 (0.785) 0.901 (0.929) 2.111(0.464)
Intercept Market potential Country risk Cultural distance R+D intensity Marketing intensity International experience Size (1)+ Size (2)+ Tacit know-how International strategy Offensivedefensive motives Exploitationexploration motives Capital intensity Degree of customization Country risk CII Cultural distance DCI w2 2log likelihood Overall classication rate (%)
7.336 (2.261) 0.744 (.228) 0.001 (0.011) 0.420 (0.394) 0.142 (0.257) 0.979 (0.370) 0.016 (0.027) 1.394 (0.902) 0.745 (1.034) 1.451(0.397)
126.459 189.805 74.1
4.851 (3.835) 8.232 (4.106) 0.748 (0.355) 1.314 (0.386) 0.053 (0.049) 0.101 (0.050) 2.921 (1.391) 2.725y (1.486) 0.172 (0.359) 0.315 (0.420) 0.892 (0.569) 0.510 (0.023) 0.056 (0.036) 0.034 (0.032) 1.639 (1.185) 0.077 (1.175) 1.034 (1.349) 0.377 (1.367) 1.042(0.501) 1.483(0.483) 1.335 (0.485) 1.756 (0.546) 0.302 (0.418) 0.975 (0.459) 0.318 (0.266) 0.454y (0.252) 1.788y (1.073) 2.685 (1.136) 0.339 (0.524) 0.382 (0.586) 0.020 (0.016) 0.035 (0.016) 0.866y (0.080) 0.677 (0.528) 170.817 144.478 82.8
Note: Standard errors in parenthesis; CA, contractual agreements, SCS: shared control subsidiary; WOS, wholly owned subsidiary (reference category); CII, capital intensity index; DCI, degree of customization Index; + reference category: the highest interval. po0.001; po0.01; po0.05; y po0.1.
183
(po0.05) and presents a negative relationship with control and resource commitment modes. Hence, propensity to control and commitment is lower when capital intensity increases. The moderating inuence of capital intensity shows a positive sign (po0.1), which indicates that the preference of service rms to use full-control modes in high-risk markets will become weaker with increasing capital intensity, conrming H4b. Results from the multinomial regression (summarized in Table 6 Model 4) allow us to conrm the main effect and the moderating inuence only when equity modes are compared (joint ventures vs. WOS). With regard to cultural distance, the results from the logistic regression (Table 5 Model 2) show that shared-control modes are preferable when service rms enter high cultural distance markets (po0.05), conrming H5a. Additionally, as we have stated in H5b, the moderating inuence of the degree of a services customization makes this relationship weaker (po0.1). Results from the multinomial regression (presented in Table 6 Model 4) conrm the main effect and the moderating inuence only when non-equity and equity modes are compared (contractual agreements vs. WOS). One of the control variables from TCT was statistically signicant (see Table 5 Model 2). As we expected, our results suggest that rms are likely to adopt high control and resource commitment modes when they enter high market potential countries (po0.01). The results are consistent in the four models (see Tables 5 and 6). However, other TCT variables (technological and marketing intensity) do not signicantly affect entry mode decisions. With regard to OCP control variables, our results do not allow us to conrm the inuence of size and international experience on entry mode decisions (see Table 5 Model 2). However, the tacit nature of know-how does signicantly increase the probability of choosing high control and resource commitment modes (po0.01).
DISCUSSION
This paper intends to offer new explanations to entry mode choice in the service sector. Generally, choice of entry mode studies has been developed in manufacturing sectors, and determinant factors have been linked to transaction characteristics or the rms organizational capabilities. The results of this paper provide new insights about international strategy because they show the importance of some variables, not considered in conventional theories, as determinants of the entry decision into foreign markets.
184
First, how rms compete in international markets is a determining factor in control decisions. The efciency requirements of a global approach can easily be achieved if rms choose high control entry modes because they ensure coordination among functional and geographically dispersed units and deal with clients needs on a global level. Moreover, some rms strategic motives are linked to a number of control orientations. Our evidence suggests that rms with asset seeking and offensive motivations are more likely to choose shared-control entry modes as a means of gaining market knowledge and overcoming the liability of foreignness, while rms with exploitation and defensive motivations prefer entry modes with a higher degree of control in order to better preserve their competitive advantage based on their knowledge of customers needs. These results are consistent with other studies focused on the service sector (Weinstein, 1977; Goldberg & Saunders, 1981; Ball & Tschoegl, 1982; Terpstra & Yu, 1988; Li & Guisinger, 1992), showing that following the client abroad to capitalize on pre-established business relationships and following the leader in international investment are important defensive driving forces in developed economies. Second, our study suggests that determinant factors of entry mode choice in manufacturing rms cannot be directly transferred to service rms, and therefore propositions originated in TCT and OCP perspectives are not universally applicable. Some variables which were generally analyzed as determinant factors of control decisions in the manufacturing sector are not signicant or present different results in the service sector. We have observed that, even in areas of high international risks, service rms tend to select more integrated entry modes because control can be acquired with comparatively little expense. Service rms are typically more labor intensive than capital-intensive rms, so they do not require large investments in xed assets as in the manufacturing sector. Lower switching costs allow service providers to more easily relocate operations if the risk becomes too great. As a result, lower resource commitments mitigate service rms perceptions of international risks and enhance the incentives for rms to integrate in countries where opportunities to achieve a better market position are higher than in more stable and competitive locations. However, the capital intensity of services weakens this relationship. The switching costs of capital-intensive service rms are higher and thus the risks may outweigh the benets of proximity. Under such circumstances, the need for exibility becomes more important, favoring lower control and resource commitment modes. On the other hand, our results show that the high information costs associated with high cultural distance markets increase the probability of
185
choosing shared-control entry modes. But this relationship is weakened by the degree of customization of the services. The heterogeneity problem of such services and the necessity for close and personal interaction enhance the importance of cultural distance. To overcome cultural distance in service offers, rms need to place greater emphasis on effective communication skills and workforce training to provide high-quality services according to the preferences and needs of the clients in foreign markets. Therefore, rms show a higher tendency to enter into culturally distant markets through high control entry modes in order to control the quality standards of customized services and the performance of employees. These unique characteristics of services, mainly the capital intensity and the degree of customization of the service, introduce some challenges that should be considered in traditional frameworks, based on transaction cost analysis and the resource-based perspective. However, several other factors proposed by conventional theories present similar results in service rms. As expected according to TCT, our results suggest that rms are likely to adopt high control and resource commitment modes when they enter high market potential countries. When the market potential increases, the benets of internalization will also increase. The xed setup costs of internalization can be spread across wider markets, thus pushing the benets associated with the investments to a higher level. However, other TCT variables, such as technological and marketing intensity, do not signicantly affect entry mode decisions. This fact may reect, as sug gested by other authors (Campa & Guillen, 1996; Lopez-Duarte & Garc aCanal, 2002), that the competitive advantages of Spanish rms in international markets are not based on technological and marketing capabilities. With regard to OCP control variables, our results indicate no signicant effect for size and international experience. Other studies which also focused on service rms have pointed out the non-inuence of service rms size on entry mode decisions. Capital requirements for initial establishment are lower than in manufacturing rms, thus size is not a determining factor in choosing FDI modes (Terpstra & Yu, 1988; Erramilli, 1991). Additionally, according to OCP theory, our results show that the tacit know-how does signicantly increase the probability of choosing high control and resource commitment modes. Service rms are, in general, labor intensive and their main strategic assets lie in employees knowledge. This type of knowledge is difcult to codify and be patented, so tacitness results in difculties in valuing know-how and its transfer through contractual processes. As a result, hierarchical coordination through full-ownership control modes are more efcient given the dissemination risk.
186
The results obtained from this paper provide evidence as to the importance of adopting an integrative approach to obtain the most realistic descriptions of strategic movements. The choice of this approach is essential since the international environment presents multiple pressures that manufacturing and service rms may face in their businesses. In this paper we draw on the entry mode literature, based on transaction cost analysis and the organizational capabilities perspective, and incorporate new determinants derived from the strategic approach and the specicity of the service sector. Collectively, these approaches provide greater explanatory power and provide new insights into the complex phenomenon that entry mode choice is. This study also provides guidelines for management about how to match control and resource requirements not only with host country conditions and rms existing capabilities but also with strategic objectives, rms international approach and the specic characteristics of the sector in which they operate. Specically, managers of service rms should evaluate the characteristics of the service that will be offered in international markets and assess the international potential that they present. Industrial characteristics therefore should be viewed as facilitators or restrictions to international commerce. Additionally, managers should take into account that for many service rms the switching costs may be comparatively small because valuable assets rest more on human capital than in physical assets (Erramilli & Rao, 1993), thus they may invest in high country risk markets where it is easier to achieve a better position than in other international markets with harder competitive conditions.
Limitations and Further Research Areas This study has several limitations. The empirical analysis is based on the general managing directors opinion (a single respondent) but entry mode decisions are usually taken by several decision makers. The use of multiple respondents would have signicantly increased the validity of the data and reduced concerns about potential response bias. To solve this problem, we have selected well-informed respondents by mailing the questionnaire to senior managers who had been involved in the internationalization decision. We should be cautious when generalizing the results because the study is focused on Spanish service rms. Therefore, future research could provide insight into the applicability of our results in different settings. Additionally, although the effective response rate to the survey (17%) is comparable to other surveys of CEOs, this must be acknowledged as a limitation.
187
Consistent with other studies focused on entry mode choice, the unit of study is an individual foreign market entry decision made by the rm. To obtain a representative sample, it would be necessary to identify all foreign market entry decisions made by target rms. However, such action is also linked to measure reliability problems because it is difcult that managers can still remember details of all those entries. This study was conducted post hoc, so some responses are likely to be exposed to the ex post or retrospective rationalizations of managers. Case studies could add more in-depth evidence about the reliability of our measurements and the necessity to assess the impact of some other factors. Moreover, researchers should conduct more detailed examinations of possible interaction effects between industry characteristics and other determinant factors to better understand how the industry sector can facilitate or impede FDI. Therefore, future studies should test specic service-related hypotheses to enhance our understanding about entry mode choice in service rms. Some factors identied in the literature as determinants of entry mode choice have not been considered in this study. We control for the inuence of international experience in the choice between different entry modes, but some researchers have also suggested that prior experience with specic entry modes may also inuence entry mode decisions (Davidson & McFetridge, 1985; Tallman, 1992; Padmanabhan & Cho, 1999). Future studies should include successful and unsuccessful prior experiences with entry types as a determinant of the choice. Furthermore, legislative and scal barriers could inuence the openness of the host country to FDIs and may force rms to choose certain entry modes because of the regulations by host governments. Studies which include entries into different international markets should incorporate this inuence. However, we consider that this variable does not signicantly affect entry mode choice in our study because many entries have been located in European and Latin American countries where legal restrictions to foreign investment are not important barriers. On the other hand, the intangibility, inseparability, perishability, heterogeneity, level of capital intensity, or customization of services present some marketing problems, which are intensied when the international dimension is included. Although there are classications of services according to some of the characteristics mentioned in the paper, all of them are basically theoretical. So, the lack of a general and empirically accepted classication scheme of services is one of the reasons that hinder the study of entry mode choice in the service sector and explains why there are few empirical studies about the internationalization process of service rms, and specially, about
188
entry modes in foreign markets. Future research should measure the dimensions that characterize each service in order to obtain a valid classication of services and to reduce the heterogeneity problem of this sector. Finally, determinant factors and entry mode choice should be linked to rm performance in order to provide insights into whether the proper alignment of entry modes with such factors actually leads to better results.
ACKNOWLEDGMENT
Authors acknowledge the nancial support from the Ministry of Science and Technology (research project: SEC 2003/06466).
REFERENCES
Agarwal, S., & Ramaswami, S. N. (1992). Choice of foreign market entry mode: Impact of ownership, location and internalization factors. Journal of International Business Studies, 23(1), 127. Anand, J., & Delios, A. (1997). Location specicity and the transferability of downstream assets to foreign subsidiaries. Journal of International Business Studies, 28(3), 579603. Anderson, E., & Coughlan, A. T. (1987). International market entry and expansion via independent or integrated channels of distribution. Journal of Marketing, 51(1), 7182. Anderson, E., & Gatignon, H. (1986). Modes of foreign entry: A transaction cost analysis and propositions. Journal of International Business Studies, 17(3), 126. Armstrong, J. S., & Overton, T. (1977). Estimating non-response bias in mail surveys. Journal of Marketing Research, 14, 396402. Aulakh, P. S., & Kotabe, M. (1997). Antecedents and performance implications of channel integration in foreign markets. Journal of International Business Studies, 28(1), 145175. Axinn, C. N., & Matthyssens, P. (2002). Limits of internationalization theories in an unlimited world. International Marketing Review, 19(5), 436449. Bagchi-Sen, S., & Sen, J. (1997). The current state of knowledge in international business in producer services. Environments and Planning, 29, 11531174. Ball, C. A., & Tschoegl, A. E. (1982). The decision to establish a foreign bank branch or subsidiary: An application of binary classication procedures. Journal of Financial and Quantitative Analysis, 17(3), 411425. Boddewyn, J. J., Halbrich, M. B., & Perry, A. C. (1986). Service multinationals: Conceptualization, measurement and theory. Journal of International Business Studies, 17(3), 4157. Bouquet, C., Hebert, L., & Delios, A. (2004). Foreign expansion in service industries: Separability and human capital intensity. Journal of Business Research, 57(1), 3546. Brouthers, K. D., & Brouthers, L. E. (2003). Why service and manufacturing entry mode choices differ: The inuence of transaction cost factors, risk and trust. Journal of Management Studies, 40(5), 11791204.
189
Buckley, P. J., & Casson, M. (1976). The future of multinational enterprises. London: Macmillan. Campa, J. M., & Guillen, M. F. (1996). Evolucion y determinantes de la inversion directa en el extranjero por empresas espanolas. Papeles de Economa Espan ola, 66, 235247. Campbell, A. J., & Verbeke, A. (1994). The globalization of service multinationals. Long Range Planning, 27(2), 95102. Carman, J. M., & Langeard, E. (1980). Growth strategies for service rms. Strategic Management Journal, 1(1), 722. Caves, R. E. (1971). Industrial corporations: The industrial economics of foreign investment. Economica, 38, 127. Chang, E., & Taylor, M. S. (1999). Control in multinational corporations (MNCs): The case of Korean manufacturing subsidiaries. Journal of Management, 25(4), 541565. Chang, S. J. (1995). International expansion strategy of Japanese rms: Capability building through sequential entry. Academy of Management Journal, 38(2), 383407. Chang, S. J., & Rosenzweig, P. M. (2001). The choice of entry mode in sequential foreign direct investment. Strategic Management Journal, 22(8), 747776. Clark, T., Rajaratnam, D., & Smith, T. (1996). Toward a theory of international services: Marketing intangibles in a world of nations. Journal of International Marketing, 4(2), 928. Contractor, F. J., & Kundu, S. K. (1998). Modal choice in a world of alliances: Analyzing organizational forms in the international hotel sector. Journal of International Business Studies, 29(2), 325358. Dahringer, L. D. (1991). Marketing services internationally: Barriers and management strategies. The Journal of Services Marketing, 5(3), 517. Davidson, W. H., & McFetridge, D. G. (1985). Key characteristics in the choice of international technology transfer. Journal of International Business Studies, 16, 521. Delios, A., & Beamish, P. W. (1999). Ownership strategy of Japanese rms: Transactional, institutional and experience inuences. Strategic Management Journal, 20(10), 915933. Domke-Damonte, D. (2000). Interactive effects of international strategy and throughput technology on entry mode for service rms. Management International Review, 40(1), 4159. Dun and Bradstreet (2002): Duns 850.000 Principales Empresas Espanolas. Dunning, J. H. (1981). International production and the multinational enterprise. London: Allen & Urwin. Dunning, J. H. (1988). The eclectic paradigm of international production: A restatement and some possible extensions. Journal of International Business Studies, 19(1), 132. Dunning, J. H. (1993). The internationalization of the production of services: Some general and specic explanations. In: Y. Aharoni (Ed.), Coalitions and competition. The globalization of professional services. London: Routledge. Dunning, J. H. (1995). Multinational enterprises and the global economy. Reading, MA: Addison-Wesley. Edgett, S., & Parkinsson, S. (1993). Marketing for service industries. A review. The Service Industries Journal, 13(3), 1939. Erramilli, M. K. (1990). Entry mode choice in service industries. International Marketing Review, 7(5), 5062. Erramilli, M. K. (1991). The experience factor in foreign market entry behavior of service rms. Journal of International Business Studies, 22(3), 479501.
190
Erramilli, M. K. (1992). Inuence of some external and internal environmental factors on foreign market entry mode choice in service rms. Journal of Business Research, 25(4), 263276. Erramilli, M. K., Agarwal, S., & Dev, C. S. (2002). Choice between non-equity entry modes: An organizational capability perspective. Journal of International Business Studies, 33(2), 223242. Erramilli, M. K., & DSouza, D. E. (1995). Uncertainty and foreign direct investment: The role of moderators. International Marketing Review, 12(3), 4760. Erramilli, M. K., & Rao, C. P. (1990). Entry mode choice in service industries. International Marketing Review, 7(5), 5062. Erramilli, M. K., & Rao, C. P. (1993). Service rms international entry-mode choice: A modied transactioncost analysis approach. Journal of Marketing, 57(3), 1938. Fladmoe-Lindquist, K., & Jacque, L. L. (1995). Control modes in international service o perations: The propensity to franchise. Management Science, 41(7), 12381249. Gatignon, H., & Anderson, E. (1988). The multinational corporations degree of control over foreign subsidiaries: An empirical test of a transaction cost explanation. Journal of Law, Economics, and Organization, 4(2), 305336. Goldberg, L. G., & Saunders, A. (1981). The determinants of foreign banking activity in the United States. Journal of Banking and Finance, 5, 1732. Gronroos, C. (1999). Internationalization strategies for services. Journal of Services Marketing, 13(4/5), 290297. Hair, J., Anderson, R., Tatham, R., & Black, W. (1999). Multivariate data analysis. Madrid: Iberica. Harzing, A. W. (2002). Acquisitions vs. greeneld investments: International strategy and management of entry modes. Strategic Management Journal, 23(3), 211227. Hennart, J. F. (1991). The transaction costs theory of joint ventures: An empirical study of Japanese subsidiaries in the United States. Management Science, 37(4), 483497. Hill, C. W. L., Hwang, P., & Kim, W. C. (1990). An eclectic theory of the choice of international entry mode. Strategic Management Journal, 11(2), 117128. Hofstede, G. (1980). Cultures consequence: International differences in work-related values. Beverly Hills, CA: Sage. Hymer, S. (1976). The international operations of national rms: A study of direct foreign investment. Cambridge, MA: MIT Press. Johanson, J., & Vahlne, J. E. (1977). The internationalization process of the rm: A model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 8(1), 2332. Johanson, J., & Wiedersheim-Paul, F. (1975). The internationalization of the rm: Four Swedish case studies. Journal of Management Studies, 12(3), 305322. Katrishen, F. A., & Scordis, N. A. (1998). Economies of scale in services: A study of multinational insurers. Journal of International Business Studies, 29(2), 305324. Kennedy, A. (2004). Entry mode choice in service rms: A review of contemporary research. Advances in International Marketing, 15, 119145. Kim, W. C., & Hwang, P. (1992). Global strategy and multinationals entry mode choice. Journal of International Business Studies, 23(1), 2953. Kogut, B., & Sing, H. (1988). The effect of national culture on the choice of entry mode. Journal of International Business Studies, 19(3), 411432. Kogut, B., & Zander, U. (1993). Knowledge of the rm and the evolutionary theory of the multinational corporation. Journal of International Business Studies, 24(4), 625645.
191
Levinthal, D. A., & March, J. G. (1993). The myopia of learning. Strategic Management Journal, 14, 95112. Li, J., & Guisinger, S. (1992). The globalization of service multinationals in the Triad regions: Japan, Western Europe and North America. Journal of International Business Studies, 23(4), 675696. Lommelen, T., & Matthyssens, P. (2004). The internationalization process of service providers: A literature review. Advances in International Marketing, 15, 95117. Lopez-Duarte, C., & Garc a-Canal, E. (2002). Adverse selection and the choice between joint ventures and acquisitions: Evidence from Spanish rms. Journal of Institutional and Theoretical Economics, 158(2), 304324. Lovelock, C. H. (1992). Managing services. Marketing, operations, and human resources. Englewood Cliffs, NJ: Prentice-Hall. Lovelock, C. H. (1999). Developing marketing strategies for transnational service operations. Journal of Services Marketing, 13(4/5), 278289. Luo, Y. (2001). Determinants of entry in an emerging economy: A multilevel approach. Journal of Management Studies, 38(3), 443472. Madhok, A. (1997). Cost, value and foreign market entry mode: The transactions and the rm. Strategic Management Journal, 18(1), 3961. Madhok, A. (1998). The nature of multinational rm boundaries: Transaction costs, rm capabilities and foreign market entry mode. International Business Review, 7(3), 259290. Makino, S., Lau, C. M., & Yeh, R. S. (2002). Asset-exploitation versus asset-seeking: Implications for location choice of foreign direct investment from newly industrialized economies. Journal of International Business Studies, 33(3), 403421. March, J. G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2, 7187. Nicolaud, B. (1989). Problems and strategies in the international marketing of services. European Journal of Marketing, 23(6), 5566. Padmanabhan, P., & Cho, K. R. (1999). Decision specic experience in foreign ownership and establishment strategies: Evidence from Japanese rms. Journal of International Business Studies, 30(1), 2543. Palmer, A. (1995). Services marketing principles and practice. Englewood Cliffs, NJ: Prentice-Hall. Pan, Y., & Tse, D. K. (2000). The hierarchical model of market entry modes. Journal of International Business Studies, 31(4), 535554. Patterson, P. G., & Cicic, M. (1995). A typology of service rms in international markets: An empirical investigation. Journal of International Marketing, 3(4), 5783. Porter, M. E. (1986). Competition in global industries. Boston: Harvard Business School Press. Prahalad, C. K., & Doz, Y. (1987). The multinational mission. New York: The Free Press. Samiee, S. (1999). The internationalization of services: Trends, obstacles and issues. Journal of Services Marketing, 13(4/5), 319328. Samiee, S., & Walters, P. G. P. (1991). Segmenting corporate exporting activities: Sporadic versus regular exporters. Journal of the Academy of Marketing Science, 19, 93104. Sarathy, R. (1994). Global strategy in service industries. Long Range Planning, 27(6), 115124. Sasser, W. E., Olsen, R. P., & Wyckoff, D. D. (1978). Management of service operations: Texts, cases, readings. Boston: Allyn & Bacon. Shrader, R. C., Oviatt, B. M., & McDougall, P. P. (2000). How new ventures exploit trade-offs among international risk factors: Lessons for the accelerated internationalization of the 21st century. Academy of Management Journal, 43(6), 12271247.
192
Tallman, S. B. (1992). A strategic management perspective on host country structure of multinational enterprises. Journal of Management, 18(3), 455471. Teece, D. J. (1981). The multinational enterprise: Market failure and market power considerations. Sloan Management Review, 22, 318. Terpstra, V., & Yu, C. M. (1988). Determinants of foreign investments of U.S. advertising agencies. Journal of International Business Studies, 19(1), 3346. Vermeulen, F., & Barkema, H. (2001). Learning through acquisitions. Academy of Management Journal, 44(3), 457476. Weinstein, A. K. (1977). Foreign investments by service rms: The case of multinational advertising agencies. Journal of International Business Studies, 8(1), 825. Weisfelder, C. J. (2001). Internationalization and the multinational enterprise: Development of a research tradition. Advances in International Marketing, 11, 1325. Williamson, O. E. (1975). Markets and hierarchies: Analysis and antitrust implications. New York: The Free Press. Winsted, K. F., & Patterson, P. G. (1998). Internationalization of services: The service exporting decision. The Journal of Services Marketing, 12(4), 294311. Young, S., Hamill, J., Wheeler, C., & Davies, J. R. (1989). International market entry and development: Strategies and management. Hemel Hempstead: Harvester Wheatsheaf.
MANAGING INTERNATIONAL MARKET ENTRY STRATEGY: THE CASE OF RETAILING FIRMS

Kannika Leelapanyalert and Pervez Ghauri
ABSTRACT
Numerous studies have focused on retailing rms and their activities in foreign markets; however, these have not been able to fully identify factors that inuence the process of retail internationalisation. This paper examines the factors that inuence the foreign market entry process in retailing rms and develops a conceptual model. The conceptual model is used to analyse two case studies. The case data were collected through in-depth interviews. N*Vivo was used to encode data and corroborate the analysis. The entry strategies of IKEA in China, and Marks & Spencer (M&S) in Hong Kong are examined. Firms planning to enter foreign markets would greatly benet from our analysis. We provide insights into factors inuencing the foreign market entry process and how rms can manage this process.
INTRODUCTION
Globalisation and the increasing pace of regionalisation, particularly in the European Union, are forcing companies to restructure and seek new
193
194
KANNIKA LEELAPANYALERT AND PERVEZ GHAURI
markets. A number of studies on internationalisation of manufacturing rms are available and there are several established theories on the internationalisation of these rms (Buckley & Casson, 1976; Dunning, 1981; Johanson & Vahlne, 1977). More recently, retailing rms have been extremely active in establishing foreign operations and seeking new markets (Rogers, Ghauri, & George, 2005). The intensity of competition has led retailers to seek new ways of expanding their business (Brown & Burt, 1992). The retailer has become the superior party in business, while the manufacturer has become the subordinate who has to bid for contracts with the retailer (PBS Frontline, 2004). Several studies focus on retailing rms (Alexander & Myers, 2000; Dawson, 2003; McGoldrick & Davies, 1995) and motives for their international expansion (Alexander, 1990; Dawson, 1994; Quinn, 1999; Williams, 1992). However, these have not been able to fully identify factors that inuence the market entry strategies and process. The existing literature is mainly concerned with how major retailing rms achieve geographic expansion and which strategies they use while entering foreign markets (Whitehead, 1992). Nevertheless, it is important to address the organisational dimension of the processes as well, which should include the nature of the decision-making process for the companies intending to enter the international arena and how the internationalisation process is managed. The changing nature of environment political forces, economic climate and cultural issues amplify uncertainty in international retailing expansion (Salmon & Tordjman, 1989; Treadgold, 1988). The global market development and increasing sophistication of customer demand, especially in the last decade, have made the world a more complex place (Brown & Burt, 1992). Consequently, rms have to develop strategies to respond quickly to different customers needs and become market-orientated. This can be achieved through market research and corresponding design of decisionmaking processes (Craig & Douglas, 2001). Market information helps a company identify factors that may inuence its entry strategy, market positioning and the level of adaptation to the local market. Market orientation is thus considered extremely important while designing marketing strategies (Deshpande & Farley, 1998; Deshpande, Farley, & Webster Jr., 1993; Kohli & Jaworski, 1990; Narver & Slater, 1990) but no attempt has been made to study market orientation and its impact on international activities of retailing rms. A company faces the uncertainty of uncontrollable factors in both home and foreign markets during internationalisation process (Ghauri & Cateora, 2006). The knowledge of operating environment in the target country
Managing International Market Entry Strategy
195
diminishes uncertainty. Hence, it is crucial to identify factors that inuence rms international activities and performance. In this study thus, we aim to answer the following research questions: 1. What are the factors that inuence the foreign market entry process of retailing rms? 2. Do retailing rms use market orientation while entering foreign markets? 3. How does networking and relationship building impact on international retailing operations? We have chosen to study retailing rms in China mainly because China is considered to be the most promising market for retailing rms. Moreover, the retailing sector has been recently opened for foreign rms in China, and a number of foreign retailers have entered this market. The success rate of foreign retailers in this market is, however, not very high.
EARLIER RESEARCH AND THE CONCEPTUAL MODEL

Based on the literature on internationalisation process and entry strategies (Buckley & Casson, 1976; Dunning, 1981; Gatignon & Anderson, 1986; Root, 1987), matching (Ghauri & Holstius, 1996; Holstius, 1991) and market orientation (Elg, 2003; Kohli & Jaworski, 1990; Narver & Slater, 1990; Rogers et al., 2005), a model is developed to study the above questions. Table 1 summarises the literature used in this study. Firm Characteristics Firm characteristics play an important role during the retail internationalisation process. These include factors such as commitment and learning orientation (Ghauri, Elg, & Sinkovics, 2004). It can be observed that previous research on internationalisation often emphasises the result of the internationalisation strategy, but fails to address the management of the process (Whitehead, 1992), that is the nature of decision-making process and the relationship between company behaviour and entry strategies. By exploring in this area further, our study will lead to a better understanding of internationalisation of retailing rms. Commitment of a company to internationalisation is dened by how much the rm is prepared to invest and how much risk to accept to achieve a
196
Table 1.
Theory and Concept Matching
Literature Review Summary.

Main Aspect Example of Studies
Logic
Understanding of how to facilitate the development of successful business relationships abroad
Network theory
Relationship of two or more independent companies that take advantage of mutual resources and sharing of information
Processes that facilitate the development of business relationship between rms in dissimilar countries, at a global, macro and micro level A network approach involves three factors: actors, resources and activities
(Ghauri et al., 2004; Ghauri & Holstius, 1996; Holstius, 1991; Rogers et al., 2005)
Firm characteristic
Firm characteristic includes rms commitment and learning orientation
Firm characteristic has impact on decision-making process and entry strategies These comprise all activities in obtaining customers and competitors information in the target market and disseminate it throughout the organization. The strategies are then formulated in response to this information
Market orientation
It involves customer orientation, competitor orientation, and inter-functional coordination, aiming to create value for customers and maximise companys longrun profits.
(Anderson, Hakansson, & Johanson, 1994; Coviello & McAuley, 1999; Ford, Gadde, & Hakansson, 1998; Ghauri & Prasad, 1995; Hakansson & Snehota, 1989; Johanson & Mattsson, 1987; Meyer & Skak, 2002) (Ghauri et al., 2004; Hofstede, 1994; Matsuno, Mentzer, & Ozsomer, 2002; Senge, 1990; Vida & Fairhurst, 1998) (Clark, 2002; Gilbert, 1999; Gummesson, 1987; Han, Kim, & Srivastava, 1998; Harris, 2001; Houston, 1986; Jaworski & Kohli, 1993; Kohli & Jaworski, 1990; Lings, 2000; Narver & Slater, 1990; Rogers et al., 2005; Slater & Narver, 1994)
197
Table 1. (Continued )
Theory and Concept Performance Logic Main Aspect Example of Studies
Performance involves three key elds: annual revenue growth, annual prot growth and average return on total assets. It is an evaluation of the perceived performance
Business performance can be measured by two different approaches: judgmental/ subjective and objective measurement
(Clark, 2002; Day, 1990; Dunne, Lusch, & Gable, 1995; Han et al., 1998; Jaworski & Kohli, 1993; Jennings & Young, 1990; Katsikeas & Morgan, 1998; Kohli & Jaworski, 1990; Lings, 2000; March & Sutton, 1997)
successful entry into a foreign market (Anderson & Weitz, 1992; Ghauri & Cateora, 2006). There are various types of commitment. For example, resource commitment is something that a company builds gradually as the market knowledge increases (Johanson & Vahlne 1977; Pedersen & Petersen, 1998). Previous studies suggest that the more a company is committed to the host market, the higher the level of relationship commitment to local parties (OReilly & Chatman, 1986). An increase in the level of commitment to a foreign market, accompanied by more local experience, will enhance the level of investment in a foreign market (Johanson & Vahlne, 1990). For the same reasons, companies with scarce resources tend to form alliances with local partners while entering overseas markets, instead of investing in these markets (Cavusgil, Ghauri, & Agarwal, 2002). Learning orientation is a concept describing an organisations learning capability (Sinkula, Baker, & Noordewier, 1997). It consists of three core values: commitment to learning, open-mindedness and shared vision (Kandemir, Ghauri, & Cavusgil, 2002; Sinkula et al., 1997). Commitment to learning enhances market understanding as a rm proactively collects information on a particular market. Effectively, retailing companies can be more effective and efcient if they gain more knowledge about the market and disseminate this knowledge throughout the organisation to formulate and adapt their strategies. Thus, the accumulation of knowledge and experiences facilitates further internationalisation as companies become more condent and commit more resources (Williams, 1992). Organisational learning and information ows within the organisation could thus enhance
198
market orientation and performance (Kohli & Jaworski, 1990). Therefore, Sinkula (1994) suggested that successful learning orientation and market orientation can lead to significant competitive advantage. Matching Matching is a fundamental concept which allows an understanding of how the development of successful business relationships can be facilitated (Ghauri & Holstius, 1996). Matching is a process that facilitates the development of business relationships between companies in dissimilar countries at global, macro and micro levels (Holstius, 1991). Matching at the global level relates to global-level actors such as international organisations involved in trade agreements, relationships and trade facilitation in certain countries. It includes international activities that could assist the international operations in different countries, e.g., by providing credit guarantees, subsidised loans and other nancial support (Ghauri & Holstius, 1996). Institutions establish relationships through membership in global organisation such as World Trade Organisation (WTO) and International Bank of Reconstruction and Development (IBRD) and by working together to develop mutual benets for member countries. At the macro level, matching relates to the level of support provided by governments (home and host) for business activities. Matching at these two levels, although seeming to be beyond the control of the organisation, can be initiated and planned/arranged by rms, e.g., through trade delegations and involvement of Ministers and Embassies in foreign countries. At the micro level, matching refers to the activities the company needs to carry out in order to achieve a successful market entry. Matching at the micro level is closely related to networking with competitors, customers and other actors in the environment. This level of matching is thus quite similar to the network approach. In this paper, we focus on matching at macro and micro level, as these are considered more relevant for retailing rms entering China. It has been emphasised that building, developing and managing business relationships are crucial issues for western companies investing in emerging markets (Cavusgil et al., 2002). It is enormously important for western companies to develop and manage relationships with the host countrys government, local partners, suppliers, local customers and even local communities. Local government plays an important role during internationalisation process and has an immense inuence on the way businesses are run.
199
Hence managing relationships with the local government is a crucial task. Managing relationships with employees (locals as well as expatriate) has proved to be crucial for retaining valuable workforce and ensuring successful relocation. Cavusgil et al. (2002) suggest approaches such as training both employees and their families about the culture and background of the foreign country, communicating effectively between the head ofce and subsidiary, and motivating people through appropriate rewards. Companies need to build and maintain relationships with local partners, such as joint venture partners, franchisees, distributors and suppliers. Relationship with local communities could be managed through investing in human resources, supporting local sports and cultural activities, having environmentally friendly policies and showing respect towards the local communities culture. The benets from such a relationship could improve the company image and reputation as well as attract more support from actors involved. It is also necessary to create and manage good relationship with local customers and create a positive image in a foreign market. Market Orientation Narver and Slater (1990) explain market orientation as all activities related to obtaining customers and competitors information from the target market as well as disseminating it throughout the organisation in order to create value for customers by developing responsive strategies. It has been previously dened as understanding customer orientation, competitor orientation and coordinating information within the organisation (Narver & Slater, 1990; Slater & Narver, 1994). Kohli and Jaworski (1990) explain the method that companies use to achieve market orientation as following: companies create intelligence by conducting market research to understand what consumers need. Dissemination means communicating the information throughout the organisation by using information exchange systems between departments. Responsiveness means designing an implementation plan that corresponds to the customers needs in a particular market. Different studies provide different definitions of market orientation (Clark, 2002; Deshpande et al., 1993; Kohli & Jaworski, 1990; Narver & Slater, 1990), however, they generally agree that it includes collection, analysis and communication of market information throughout the organisation and adapting products and strategies accordingly. Many studies found that profitability is an outcome of market orientation (Deshpande et al., 1993; Narver & Slater, 1990). The above discussion helped us to develop the conceptual model used in this study (see Fig. 1).
200
Corporate
Host Market
Matching of actors, activities, resources
Firm Characteristics
Macro Level Micro Level

Host Market Performance
Commitment to host
market
Learning orientation
Market Orientation

Fig. 1.
Intelligent Generation Intelligent Dissemination Responsiveness
The Role of Market Orientation and Matching in International Retailing. Source: Based on Ghauri et al. (2004).
201
The model includes the matching concept (Ghauri & Holstius, 1996) which claries how successful business relationships could be facilitated at global, macro and micro levels. Furthermore, the model includes market orientation in the retail industry. It is assumed that marketing orientation and matching factors can be affected by rm characteristics and allow retailing companies to better understand consumers and facilitate business activities, leading to better performance in a foreign market. The model allows us to explore the impact of matching and market orientation on the internationalisation process of retailing rms.
RESEARCH METHODOLOGY
Our research approach is mainly based on case studies and semi-structured interviews. Case studies are intended to point out detailed and valuable insights and understanding of the retail internationalisation process, which could not have been achieved by a survey method (Ghauri & Grnhaug, 2005; Yin, 1994). We try to understand how rms handle different factors that inuence their activities while entering a foreign market. As we want to answer how and why questions and are conducting inductive research, case study method is considered most appropriate. This approach is also suitable for theory development, which suits the objective of this study. Case studies examine similarities and differences in the strategies used by M&S and IKEA to enter China. Data is collected through a series of indepth, semi-structured interviews with head ofce managers as well as company representatives in the host country. This includes frontline employees, middle-level and senior managers. This paper is a part of a bigger project where we are studying international activities of IKEA and M&S in several markets. We have full access to managers and documents of these two companies. In this paper, however, we present their entry into China. For this study, a total of twenty-four interviews were conducted at the M&S head ofce and their Hong Kong subsidiary. Twelve interviews were conducted in IKEA head ofce and their subsidiaries and suppliers in China. We also collected data through various sources such as follow-up telephone interviews, document analysis and observations. Data has been analysed using pattern matching method (Miles & Huberman, 1994; Yin, 1994). Validity of the research is reinforced by the fact that various in-depth interviews have been conducted. The same questions have been posed to different managers at the different levels. Information was also drawn from
202
secondary sources such as articles and news. These approaches allowed us to cross-check the accuracy of data and ensure internal validity and reliability of data. Similar questions were raised from each case and common outcomes led us to draw conclusions (Miles & Huberman, 1994). N*Vivo software was used to code the data and corroborate the analysis. It is a useful instrument for the management of enormous amount of interview transcripts, encoding of data and management of themes in order to analyse qualitative data (Richards, 2000; Sinkovics, Penz, & Ghauri 2005).
MARKS & SPENCER IN HONG KONG

M&S is one of the leading clothing retail companies in the UK. The company was founded as a partnership between Michael Marks and Tom Spencer in 1894. In 1926, Marks and Spencer Limited became a public company. M&S strategy is based on its vision, mission and values (Company Report, 2005), which are: Vision: The standard against which all others are measured Mission: Making inspirational quality accessible to all Values: Quality, service, innovation and trust M&S has expanded to food, beauty products, home furnishing and nancial services. It is a large organisation, which comprises 70,000 employees, with 3,000 employees in head ofce. The company operates more than 400 stores in 29 countries across Asia, Europe and the Middle East (Company Report, 2005). The focus of this research is on its subsidiary in Hong Kong. M&S expanded to Hong Kong in 1980s by franchise, which was later turned into a wholly owned subsidiary. Firm Characteristics After a number of bad years in several markets and the Asian crisis, M&S announced a restructuring strategy on March 29, 2001. One of the outcomes of the restructuring plan was a focus on UK retail market and improvement of capital structure to leverage benets through return of value to shareholders. Company decided to focus on its home market rather than on the international markets as part of the getting t for growth concept. As a result, M&S closed its operations in a number of European markets in order to rationalise its operations.
203
In 1999, it became apparent that M&Ss global ambitions had failed so the company decided to return to its home market. By 2002, M&S had failed to sell its 10 Hong Kong stores to a local franchisee and acquired direct ownership of the stores. Matching We make all of our profits and revenues from all three stores located only in prime locations. We need to be in the main shopping malls in Hong Kong (Managing Director, M&S Hong Kong). M&S recognises the importance of using and building micro-level matching to gather knowledge of the local market. It recognises the importance of having the right contacts in retail business. International consulting companies such as PriceWaterHouseCoopers (PWC) and Ernst & Young also provide local information on trading environment. Landlords are very important actors to have relationships with in Hong Kong because having the prime location is crucial. Major retailers need to be in the prime location Tsim Sha Tsui, Central, Admiralty and Time Square in order to guarantee success and reasonable exposure to the market. Severe acute respiratory syndrome (SARS) outbreak, which happened in March 2003, had an adverse effect on Hong Kongs economic conditions and performance of all retailers in the market. M&S is a member of the Hong Kong Retail Association that negotiated a decrease in rent with landlords due to the SARS outbreak. Market Orientation The company does not systematically gather customer, competitor and market information. Some managers follow their intuition based on their own experience and entrepreneurship skills in doing business. They also use secondary data and country visits to identify potential new markets. However, some systematic market research has recently been introduced in some foreign markets. Generating Information M&S uses a global strategy consulting company which works mainly on M&S brand strategy. Research consists of analysing M&S international brand strategy, positioning M&S brand in country clusters, developing a brand strategy in each country and suggesting marketing tools in order to increase sales and profitability. Hong Kong subsidiary recently conducted
204
their own market research to understand Mainland Chinese people and their buying behaviour. The Mainland Chinese tourists represent as much as 2530% of M&S customers. Disseminating Information M&S communicate through regular meetings within departments as well as between departments. The sales performance and important issues are brought up to the weekly meetings amongst management teams in order to discuss potential solutions and improvements. There are no systematic communication tools and there is little evidence of sharing best practices and experiences. The Hong Kong ofce and head ofce communicate regularly via email and country visits. Hong Kong ofce nally turned loss into prot in 2004. Responsiveness The company made small adaptations to the local market in sizes, colours and styles of products. We have two inches shorter sleeves shirt for Asian men. We also have unique products for proper climate, e.g., we developed a range of short sleeves T-shirts to ll in the longer summer period in Hong Kong (Merchandising Manager, M&S Hong Kong). Major attention is paid to logistics as efciency is imperative in retail operations. M&S aims to shift a greater volume of products by cutting costs through improvement of their logistics systems. They now deliver products directly from far-east manufacturers to the east Asia shipping hub in order to reduce costs and time of transport. Host Market Performance M&S focuses more on the UK market and has less commitment to its Hong Kong market. In the past, the Hong Kong subsidiary worked mainly independently from head ofce at the beginning, as operations were run by UK expatriates using somewhat standardised strategies from the head ofce, which resulted in limited local adaptation and less successful performance. The Asian crisis in 1998 affected the nancial performance dramatically. Hierarchy structure in Hong Kong had slowed down decisionmaking process, leading M&S to cut unnecessary positions and become a lean organisation. The company turned loss into prot in 2004 by cutting cost. However, the weak market orientation which is reected in unattractive products and high prices is the main concern to securing the brand reputation in the market.
205
IKEA IN CHINA
IKEA is a Swedish furniture company, which was founded by Ingvar Kamprad. The company has 224 IKEA stores in 33 countries across Europe, North America, the Middle East and Asia-Pacic. The company expands mainly by wholly owned subsidiaries. The IKEA Group itself owns 200 stores in 24 countries. The rest operate as franchises in 15 countries such as Hong Kong, Taiwan and Australia (Company Corporate Website, 2005). China is an increasingly attractive market because of a large population, the loosening up of Chinese regulation, cheap labour and the GDP growth rate of 9.1% indicating the growth of consumer demand (Central Intelligence Agency, 2005). International retailers also expand into China to take advantage of low cost production suppliers and increasing purchasing power of local consumers. IKEA opened a store in Beijing in 1998 and a agship store, with 32,000 square metres, in Shanghai in 2003. China is the top purchasing country for IKEA as 19% of IKEA products are sourced from China (Company Corporate Website, 2005). Firm Characteristic Ingvar Kamprad, the founder of IKEA, is particularly interested in Chinese market and shows long-term commitment to it. In terms of learning orientation, IKEA transfers the experience and knowledge to the local ofces. Experienced expatriates work on establishing an IKEA store, which includes training in infrastructure, shop, management structure and how to transfer their knowledge to the local market. Company needs those people to set up more IKEA stores for further expansion in other cities. However, the company expects the local people to move up and take management roles in the near future. There is a competence centre set up in Shanghai for training new employees and there are regular training workshops for existing workers. IKEA also learns lessons from the local market and uses its experience in new stores. It is important to establish relationships with local stakeholders as property and human resource managers. Matching Macro matching is very important to IKEAs Chinese operations. Company needs to build relationships with central government and local authorities. Networking with the right people could reduce legal and regulation barriers. Company also needs to be up to date on government policies towards retail
206
and foreign direct investment. IKEA thus works with the government on all levels. We rely very heavily on government in China when it comes to our retail operations: District governments, municipal government and obviously also the central government (President of IKEA China). In the early stages of entry process, Mr. Kamprad travelled to China and had several meetings with regional and local government ofcials. IKEA also uses micro matching and strong network relationships with media and suppliers. There is a two-way supporting relationship between IKEA and its suppliers. IKEA educates suppliers about production techniques and factory standards and provides nancial assistance. Suppliers need to follow I-way standard, which includes environmental responsibility, no child labour, and acceptable working environment in order to become IKEAs supplier. IKEA helps suppliers to adopt higher factory standards, while IKEA relies on low prices and assistance with distribution. Suppliers also give IKEA recommendations about new product ideas and local customer needs. Market Orientation Conducting market research has not only allowed an understanding of the local market, but it has also enabled IKEA to test the market before deciding to enter fully. Steed (1985) explains the successes of IKEA internationalisation through their unique entry strategy. The companys strategy is to enter initially through small investment and gain experience before entering fully in the market. The evidence of this unique entry strategy is also found in China. IKEA opened small stores to test the market before making commitment to building big stores in Beijing and Shanghai. We had stores in Shanghai before we opened this one. That was smaller, what we call small shops to test the market (Expansion manager, IKEA China). IKEA emphasises generating market intelligence. Main market research and compilation of customer satisfaction index have been conducted at the corporate level in order to measure brand recognition and customer satisfaction. Management also makes store visits and cashier surveys at the store level. The competitive shopping and competitor visits have been conducted in order to gain information about copying products and benchmark prices. The customer information generated through keeping records of customers most frequent questions, sales records, customer requirements, day-to-day operational issues and solutions. Suppliers also provide information about local product needs such as Chinese wok and knife that were introduced on their suggestion. We hire customers best friend customer service. So we are getting information, transforming that and addressing the issues. Example of issues
207
could be difculties to shop, customer do not understand the system and concept, or long hallway to walk. We are close to customers and aim to get information from customers and we immediately worked on it to make necessary adaptations and changes. We have new local marketing manager in each store working with me (Marketing manager, IKEA China). Information is disseminated through department meetings, store meetings, management meetings, email and intranet. Intranet is often used at management level. At the store level, there is a language barrier to using intranet and managers always spend time in the store to get the required information. There is, however, little cooperation between IKEA China and IKEA in other countries. Contact and discussion happens only in training sessions, although some managers feel that managers in other countries might face similar problems and are desperate to nd solutions. However, some managers do communicate with others because of previous contact and it is easier for them to communicate and share experiences. In terms of responsiveness, IKEA uses the concept of standardisation of the brand all over the world. It is accepted that the brand image must be consistent. The store look must be the same in every country. The company, however, does not neglect the importance of local customers and has some local solutions. Several IKEA products and prices are adapted to local market and local customers needs. The company takes advantage of being market-orientated to understand local consumers and especially local market environment. It makes use of its market research and market information in order to educate people in the market about IKEA concept. Company introduced the do-it-yourself concept to Chinese market and explained to customers how and why IKEA maintains its low prices, through store posters. IKEA made small adjustments to the local market, such as having small space living showroom, local brochure written in Chinese characters and use of local language in advertising campaigns and media. Children and living with kids was a key priority for global IKEA 2003 campaign. It will not work in china because Chinese only have one child according to one child policy. So we focus on low price offer commercial theme in china because the IKEA products are still perceived as expensive (Head of bedroom department, IKEA Shanghai). Host Market Performance The company expects aggressive growth in Asia, particularly in China, and plans to open several stores in the near future. Although Europe is still the main market, we see enormous growth in Asia, North America, Japan and
208
particularly in China (Business Development Manager, IKEA China). IKEA currently searches for new store locations in 14 cities including Chengdu, Dalian, Nanjing and Qingdao (China Daily, 2004). Price is a priority issue that IKEA is working on. IKEA has also increased the number of products sourced from local suppliers. IKEA needs to focus more on local sourcing in order to increase volume and revenue, to change its image to an affordable price brand, to be different from its competitors and to expand to lower-income cities in China in the future. Marketing manager points out that IKEAs key success factor is to maintain IKEA position and present what IKEA is to the market in a consistent manner. In Chinese market, IKEA would like to be the leading home furnishing company which offers affordable products and affordable solutions for better living at home. It is the same concept as globally but adds affordable solution.
DISCUSSION
Our cases reveal that rm characteristics, including host market commitment, learning orientation and organisational culture impact host market performance. IKEA shows long-term commitment through investment and future expansion and its ability to learn from the local market. Also, the combination of a at structure and its modern organisational culture has lead IKEA to become market orientated. IKEA management style also encourages the company to use matching concept and build the strong relationships with local authorities and suppliers. IKEA has internationally experienced people who move around to help set up new operations in different countries. For example, the expansion manager has experience in introducing IKEA in Poland, France and Germany before coming to China. This is not the case for M&S, which has only one subsidiary ofce in Hong Kong, while stores in other countries are run as franchise operations. The other subsidiaries, particularly in Europe, were not successful and were closed down. The similarity in the two cases is the role played by the expatriates. Companies need expatriates to build up the organisation in new markets. The problem for M&S was that expatriates stayed in Hong Kong for too long and did not let local people operate the business. When the local people took over, performance improved. This is consistent with Jackson and Sparks (2005) study that concluded that expatriates are normally less inclined to adapt as they normally follow the head ofces standardised strategies.
209
Matching is a crucial element for doing business in China. At the macro level, the central government and local government played an important role because they are decision makers in policy and regulation and issue permits and permissions. Moreover, they can open doors to other useful organisations. At the micro level, landlord was the most important actor for M&S because of the limited access to prime retail locations. Suppliers play a significant role in reducing the production cost and maintaining good quality in China as in case of IKEA. As a result, those factors have inuenced IKEA operations to become one of the most successful foreign companies in the Chinese market. This conrms Ghauri and Holstius (1996) and Rogers et al. (2005) ndings that matching activities with local governments can facilitate the market entry by a foreign rm. Market Orientation This research points out that IKEA was more market orientated. The market research was conducted systematically and the company made appropriate use of it. Communication within organisation was effectively used among all levels of the management. IKEA has responded to the local needs through adaptation of its products, prices and local sourcing. This was reected in their market performance and was consistent with our conceptual model and earlier studies on market orientation (Slater & Narver, 1994; Kohli & Jaworski, 1990). M&S recently began conducting market research but there is still no systematic research and dissemination of information in the Hong Kong subsidiary. Managers from different markets meet occasionally to share experience and best practice. M&S started as a successful British company using common sense and entrepreneurship in business operations. It worked in the past and they assume that it will always work. As suggested by our conceptual model and earlier studies (Rogers et al., 2005), it has therefore led to a less successful market performance. M&S wanted to sell its operations in Hong Kong but it did not succeed. Both IKEA and M&S still face the price issue in China. The products prices are still too high for Chinese market. IKEA, however, is more consistent in focusing on lower prices and is using more and more local suppliers to achieve that, thus being more responsive to local market conditions. As far as performance is concerned, IKEA in the Chinese market is a huge success mainly because of its efcient matching and market-orientation activities. The company plans to open several more stores by 2010. In case of
210
Table 2. The Inuencing Factors in International Retailing: The Cases of IKEA China vs. M&S Hong Kong.
IKEA China Firm characteristic Long-term commitment in China by investing in supplier training Flat structure helps two-way communication and share knowledge Motivating people to share knowledge by training and education Central and district government are crucial actors in retail operations Long-term relationship with suppliers help IKEA lowprice purchasing Suppliers provide local and product knowledge Extensive market research at headquarter level, country level and store level Customer satisfaction index Cashier survey Competitor visits Store visits Information system, i.e., database and intranet used at all management levels to share information on local markets E-mail Regular meeting at all levels Adaptation of product and prices Product is too expensive for the Chinese market to handle price issue more and more local production Creating Chinese brochure to respond to Chinese life style Successful entry to China Plan for further expansion to open 14 stores by 2010 Challenge on lower the price to respond to further expansion M&S Hong Kong Company wanted to sell Hong Kong subsidiary in 2001 but did not succeed Hierarchy structure leads to slow decision-making process Company tried to get rid of hierarchy and give power to local people Government support tourism, Mainland Chinese entry to Hong Kong boost sales Landlord money talk, improve brand image Retail Association rarely share information due to business competition Feedback from sales assistant Competitor shopping Store visits No systematic market research
Macro matching
Micro matching
Market orientation: intelligence generation
Market orientation: intelligence dissemination
Department meetings and management meetings but no information from local markets is exchanged E-mail used in management
Market orientation: responsiveness
Host market performance
Some adaptation of products style, size and colour to local needs Cherry-picking standard product by experience local buying and merchandising staff Direct delivery not based on market research Turn loss into prot in 2004 Well establish in Hong Kong but brand damage from UK situation Challenge to stay in prime location and build up M&S brand
211
M&S, there is no evidence of market orientation and thus its performance is not so good. It showed a prot rst in 2004, rst time since 1998. Matching activities with landlords in Hong Kong are crucial because of the competition for prime locations, but there is hardly any interaction with local customers or understanding of local competitors and other conditions. The company needs to rebuild the brand image to make the brand attractive to customers and the landlords in order to regain its market share and maintain stores in prime locations, which can only be achieved through market orientation and matching. Table 2 summarises our ndings.
CONCLUSION
To answer the research question, our study reveals that market orientation plays a significant role in international operations of retailing rms. Western retailers by applying market orientation in the Chinese market and by using matching through developing relationship with local government and other actors can improve their performance. In case of market orientation, intelligence generation, regular market research and intelligence dissemination results in achieving local responsiveness and adaptation of products and strategies to the local markets. Different companies however apply different levels of market orientation depending on rm characteristics. It has been found from both cases that rm characteristics play an important role in driving market orientation strategy. Compared with IKEA, M&S is still slow to react to the environment, customer needs and competition. Therefore the company does not have the ability to respond quickly to local needs and thus loses competitiveness. Our studies agree with Ghauri and Holstius (1996) that matching facilitates market entry and operational efciency. It has been found that market-orientated company has ability to better understand market environment, its competitors, customers and to be more responsive to local customer needs, therefore leading to significant success in the host market. Hence, the market orientation host market performance link appears to be strong while comparing the M&S and IKEA operation, which is in line with earlier studies (e.g., Slater & Narver, 1994). From theoretical perspective, this paper takes the matching and market orientation concepts further by applying it in international retailing context. The study provides a conceptual framework for further development of theory within an international retailing context. From the managerial point of view, this research is useful for practitioners and retailers in terms of highlighting the key factors, which could inuence the successful
212
international operations. Businesses planning to enter foreign market would greatly benet from processes and models, which deliver insights into local markets and trends as shown in our study. It has been assumed that marketing orientation and matching could be affected by rm characteristics, allowing retailing companies to better understand consumers and facilitate business activities, leading to a better performance. Moreover, it is clear that a rms learning orientation and commitment to a particular market are the main characteristics that enable it to be more market oriented. Further studies of this research would be to test the model in other company cases and other countries to establish its external validity.
REFERENCES
Alexander, N. (1990). Retailers and international markets: Motives for expansion. International Marketing Review, 7(4), 7585. Alexander, N., & Myers, H. (2000). The retail internationalisation process. International Marketing Review, 17(4/5), 334353. Anderson, E., & Weitz, B. (1992). The use of pledges to build and sustain commitment in distribution channels. Journal of Marketing Research, 29(1), 1843. Anderson, J., Hakansson, H., & Johanson, J. (1994). Dyadic business relationships within a business network context. Journal of Marketing, 58(4), 115. Brown, S., & Burt, S. (1992). Conclusion-retail internationalization: Past imperfect, future imperative. European Journal of Marketing, 26(8,9), 8084. Buckley, P. J., & Casson, M. (1976). The future of the multinational enterprise. New York: The Macmillan Company. Cavusgil, T. S., Ghauri, P. N., & Agarwal, M. R. (2002). Doing business in emerging markets: Entry and negotiation strategies. Thousand Oaks, CA: Sage. Central Intelligence Agency. (2005). The world factbook 2005: China economy overview. Available at http://www.cia.gov/cia/publications/factbook/geos/ch.html, 6/10/05. China Daily. (2004). IKEA outlines mainland expansion plan. Available at http://www. china.org.cn/english/Life/92256.htm, China Daily, 6/04/04. Clark, B. (2002). Measuring performance: The marketing perspective. In: A. Neely (Ed.), Business performance measurement: Theory and practice. Cambridge: Cambridge University Press. Company Corporate Website. (2005). About IKEA: Timeline full story. Available at http:// www.IKEA.com/ms/en_GB/about_IKEA/timeline/full_story.html, 6/7/05. Company Report. (2005). Marks & Spencer: International franchise. Available at http:// www2.marksandspencer.com/thecompany/index.shtml, 5/6/05. Coviello, N., & McAuley, A. (1999). Internationalization of the smaller rm: A review of contemporary empirical research. Management International Review, 39, 223256. Craig, S., & Douglas, S. (2001). Conducting international marketing research in the twenty-rst century. International Marketing Review, 18(1), 8090.
213
Dawson, J. (1994). Internationalization of retailing operations. Journal of Marketing Management, 10, 267282. Dawson, J. (2003). Towards a model of the impacts of retail internationalisation. In: J. A. Dawson, M. Mukoyama, S. C. Choi, & R. Larke (Eds), The internationalisation of retailing in Asia (Chapter 11, pp. 189209). London: Routledge Curzon. Day, G. (1990). Market driven strategy. New York: The Free Press. Deshpande, R., & Farley, J. (1998). Measuring market orientation: A generalization and synthesis. Journal of Market Focused Management, 2(3), 213232. Deshpande, R., Farley, J., & Webster, F., Jr. (1993). Corporate culture, customer orientation, and innovativeness in Japanese rms: A quadrad analysis. Journal of Marketing, 57, 2357. Dunne, P., Lusch, R., & Gable, M. (1995). Retailing (2nd ed.). USA: International Thomson Publishing Company. Dunning, J. H. (1981). International production and the multinational enterprise. London: Allen and Unwin. Elg, U. (2003). Retail market orientation: A preliminary framework. International Journal of Retail & Distribution Management, 31(2), 107117. Ford, D., Gadde, L.-E., & Hakansson, H. (1998). Managing business relationships. Chichester: Wiley. Gatignon, H., & Anderson, E. (1986). Modes of foreign entry: A transaction cost analysis and propositions. Journal of International Business Studies, 17(3), 126. Ghauri, P., & Cateora, P. (2006). International marketing (2nd ed.). London: McGraw-Hill. Ghauri, P., & Grnhaug, K. (2005). Research methods in business studies: A practical guide (3rd ed.). London: Prentice-Hall. Ghauri, P. N., Elg, U., & Sinkovics, R. (2004). Foreign direct investment location attractiveness for retailing rms in the European Union. In: L. Oxelheim & P. Ghauri (Eds), European Union and the race for foreign direct investment in Europe (pp. 407428). UK: Elsevier. Ghauri, P. N., & Holstius, K. (1996). The role of matching in the foreign market entry process in the Baltic states. European Journal of Marketing, 30(2), 7588. Ghauri, P. N., & Prasad, B. S. (1995). Probing Asias inter-rm linkages: A network approach. Advance in International Comparative Management, 10, 6378. Gilbert, D. (1999). Retail marketing management. Essex: Pearson Education Limited. Gummesson, E. (1987). Using internal marketing to create a new culture: The case of Ericsson quality. Journal of Business and Industrial Marketing, 2(3), 2328. Hakansson, H., & Snehota, I. (1989). No business is an island: The network concept of business strategy. Scandinavian Journal of Management, 5(3), 189200. Han, J. K., Kim, N., & Srivastava, R. K. (1998). Market orientation and organizational performance: Is innovation a missing links? Journal of Marketing, 62(4), 3045. Harris, L. C. (2001). Market orientation and performance: Objective and subjective empirical evidence from UK companies. Journal of Management Studies, 38(1), 1743. Hofstede, G. (1994). The business of international business is culture. International Business Review, 3(1), 114. Holstius, K. (1991). A matching concept for international business: The case of Finland and China. Advance in Chinese Industrial Studies, 2, 177185. Houston, F. S. (1986). The marketing concept: What it is and what it is not. Journal of Marketing, 50(April), 8187.
214
Jackson, P., & Sparks, L. (2005). Retail internationalisation: Marks and Spencer in Hong Kong. International Journal of Retail & Distribution Management, 33(10), 766783. Jaworski, B. J., & Kohli, A. K. (1993). Market orientation: Antecedents and consequences. Journal of Marketing, 57(3), 5370. Jennings, D. F., & Young, D. M. (1990). An empirical comparison between objective and subjective measures of the product innovation domain of corporate entrepreneurship. Entrepreneurship: Theory & Practice, 15(1), 301344. Johanson, J., & Mattsson, L.-G. (1987). Internationalization in industrial systems a network approach compared with the transaction-cost approach. International Studies of Management and Organization, 17(1), 3448. Johanson, J., & Vahlne, J.-E. (1977). The internationalization process of the rm a model of knowledge development and increasing foreign market commitment. Journal of International Business Studies, 8(1), 2332. Johanson, J., & Vahlne, J.-E. (1990). The mechanism of internationalization. International Marketing Review, 7(4), 1124. Kandemir, D., Ghauri, P., & Cavusgil, T. (2002). The strategies role of organizational learning in relationship quality in strategic alliances. In: F. Contractor & P. Lorange (Eds), Corporative strategies and alliances (pp. 799828). London: Elsevier. Katsikeas, C. S., & Morgan, N. (1998). Assessing rms export performance: A literature review and guidelines for future research. Working paper. Cardiff: Cardiff Business School, University of Wales. Kohli, A. K., & Jaworski, B. J. (1990). Market orientation: The construct, research propositions and management implications. Journal of Marketing, 54(2), 118. Lings, I. (2000). Developing the domain of internal market orientation & its consequences for market orientation & performance, research paper series. Birmingham: Aston Business School, Aston University. March, J. G., & Sutton, R. I. (1997). Organizational performance as a dependent variable. Organization Science, 8(6), 698706. Matsuno, K., Mentzer, J. T., & Ozsomer, A. (2002). The effects of entrepreneurial proclivity and market orientation on business performance. Journal of Marketing, 66(3), 1832. McGoldrick, P. J., & Davies, G. (1995). International retailing: Trends and strategies. London: Pitman Publishing. Meyer, K., & Skak, A. (2002). Networks, serendipity and SME entry into Eastern Europe. European Management Journal, 20(2), 179188. Miles, M. B., & Huberman, M. A. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). London: Sage. Narver, J. C., & Slater, S. F. (1990). The effect of a market orientation on business profitability. Journal of Marketing, 54(4), 2035. OReilly, C., & Chatman, J. (1986). Organizational commitment and psychological attachment: The effects of compliance, identication, and internalization on prosocial behavior. Journal of Applied Psychology, 71(3), 492499. PBS Frontline. (2004). Is Wal-Mart good for America? Available at http://www.pbs.org/wgbh/ pages/frontline/shows/walmart/etc/synopsis.html, 14/11/05. Pedersen, T., & Petersen, B. (1998). Explaining gradually increasing resource commitment to a foreign market. International Business Review, 7(5), 483501. Quinn, B. (1999). The temporal context of UK retailers motives for international expansion. The Service Industries Journal, 19(2), 101116.
215
Richards, L. (2000). Using NVivo in qualitative research (2nd ed.). Australia: QSR International. Rogers, H., Ghauri, P. N., & George, K. L. (2005). The impact of market orientation on the internationalization of retailing rms: Tesco in Eastern Europe. The International Review of Retail, Distribution and Consumer Research, 15(1), 5374. Root, F. R. (1987). Entry strategies for international markets. Massachusetts: D.C. Health and Company. Salmon, W. J., & Tordjman, A. (1989). The internationalization of retailing. International Journal of Retailing, 4(2), 216. Senge, P. M. (1990). The fth discipline: The art and practice of the learning organization. New York: Doubleday. Sinkovics, R., Penz, E., & Ghauri, P. N. (2005). Analysing textual data in international marketing research. Qualitative Market Research: An International Journal, 8(1), 938. Sinkula, J. M. (1994). Market information processing and organizational learning. Journal of Marketing, 58(January), 3545. Sinkula, J. M., Baker, W. E., & Noordewier, T. (1997). A framework for market-based organizational learning: Linking values, knowledge, and behaviour. Journal of the Academy of Marketing Science, 24(4), 305318. Slater, S. F., & Narver, J. C. (1994). Does competitive environment moderate the market orientation-performance relationships? Journal of Marketing, 58(1), 4655. Steed, J. (1985). Selling Swedish style. In: H. B. Thorelli & T. S. Cavusgil (Eds), International marketing strategy. Oxford: Pergamon Press. Treadgold, A. D. (1988). Retailing without frontiers. International Journal of Retail & Distribution Management, 16(6), 812. Vida, I., & Fairhurst, A. (1998). International expansion of retail rms: A theoretical approach for future investigations. Journal of Retailing and Customer Services, 5(3), 143151. Whitehead, M. B. (1992). Internationalization of retailing: Developing new perspectives. European Journal of Marketing, 26(8/9), 269285. Williams, D. E. (1992). Retailer internationalization: An empirical inquiry. European Journal of Marketing, 26(8/9), 824. Yin, R. K. (1994). Case study research: Design and methods (2nd ed.). London: Sage.
EXTENDING AND COMPARING CAVUSGILS OVERALL MARKET OPPORTUNITY INDEXES

Michael R. Mullen and Shirley Ye Sheng
ABSTRACT
The imperatives of globalization are clear across many industries: rms must look to expand into international markets to survive and thrive. This study complements and extends a growing body of work developing and using overall market opportunity indexes (OMOIs) based on Cavusgil (1997) to rank the attractiveness of potential foreign markets. The index developed in this paper assesses countries market potential beyond the traditional measures of market size and economic development by also including political risk, economic freedom, telecommunications as well as physical infrastructure and geographic distance. We provide a current analysis of market attractiveness and opportunity for the largest set of countries indexed and ranked to date, including 24 countries not in previous OMOI studies. The validity of the OMOI is also so assessed for the rst time by comparing the ranking of market opportunity to actual subsequent trade ows from the US. Furthermore, we compare the dimensions, variables, samples and results from three of Cavusgil and colleagues previous studies and the two conducted herein. The choice of sample and, to a lesser degree, weights are shown to directly affect the OMOIs and rankings. The modied OMOI is shown to be a exible, valid and fairly stable tool for preliminary analysis of foreign market opportunity.
219
220
MICHAEL R. MULLEN AND SHIRLEY YE SHENG
INTRODUCTION
Global, political and economic liberalization have created tremendous business opportunities and challenges for international marketers (Kotabe, 2001). Those foreign opportunities and intense competition in domestic markets have caused rms to consider entering international markets. The imperatives of globalization are clear across many industries; rms must look to expand into international markets to survive and thrive. However, seizing international opportunities require careful research of foreign markets and their potential (Czinkota & Ronkainen, 2004). One of the most important strategic determinations when deciding to export or to establish operations in foreign countries is the selection of the right market (Papadopoulos & Denis, 1988; Douglas & Craig, 1992; Kumar, Antonie, & Joachimsthaler, 1994; Cavusgil, 1997). Entering new markets require a major commitment of nancial and managerial resources. Choosing the wrong markets can lead to substandard nancial performance or failure. Therefore, the most frequent objective of international market research is that of foreign market opportunity analysis, (Czinkota & Ronkainen, 2004, p. 191). Developing research methods to facilitate the choice of markets is, then, an important area of international market research. Papadopoulos and Denis (1988) stress the importance and necessity for systematically evaluating the potential of foreign market opportunities. With more than 200 potential markets, a business executive can be overwhelmed with the diversity and complexity of alternative market opportunities, (Cavusgil, Kiyak & Yeniyurt, 2004, p. 607). For small and medium-size businesses that have not yet entered international markets, entry decisions constitute an especially critical rst step on the path to internalization. Many scholars agree that a rms initial market selection is the rst building block on which a rms entire internationalization strategy depends. Nonetheless, evidence indicates that many rms, especially small and medium rms, rarely or never do any research in this area (Kothari, 1978). Businesses just starting international marketing typically choose their rst markets based on reaction to unsolicited orders or other stimulation (Bilkey & Tesar, 1997; Bilkey, 1978; Cavusgil, 1980) rather than through a proactive, strategic research analysis. A multi-stage selection process is an appropriate approach to selecting a market(s) and/or making mode of entry decisions (Papadopoulos & Denis, 1988; Kumar et al., 1994; Russow & Okoroafo, 1996; Cavusgil, 1997). Cavusgil (1997) proposes a three-step process for choosing overseas markets. He suggests a preliminary screening to determine the overall attractiveness of
Extending and Comparing Cavusgils Overall Market Opportunity Indexes
221
potential markets for further investigation, to be followed by an assessment of industry market potential to estimate aggregate demand, and nally, an analysis of sales potential particularly unique to a companys products. The rst step in the market selection process is a preliminary market assessment aimed at reducing the total number of available countries in the world to a more manageable set of high-potential markets (Czinkota & Ronkainen, 2004). Regardless of the number of stages in the market selection process, it is important to emphasize that in-depth market analysis must follow the preliminary screening before making a nal choice of market(s) to enter (Green & Allaway, 1985; Papadopoulos & Denis, 1988) and the subsequent decision on the mode of entry (Root, 1984). Cavusgil (1997) provides insights into measuring the potential of emerging markets through development of an OMOI. Cavusgil uses 13 indicators of market potential to form his OMOI which he uses to rank the attractiveness of 23 emerging markets. Most recently, Cavusgil et al. (2004) updated the OMOI and applied it to 89 countries. The purpose of this study is to modify and extend Cavusgils OMOI into a modied OMOI covering a majority of the countries in the world representing 96 percent of world gross domestic product (GDP). We also seek to validate our modied OMOI and to compare the measures, samples and results across studies, and to examine the sensitivity of the linear compensatory models used in these analyses. While this approach to preliminary screening and ranking does not identify the nal right market, it does reduce the world to a set of high-potential markets for further in-depth evaluation, saving rms time and money in the nal selection process. The following section provides a selective literature review followed by a discussion of our modied index including the data, samples and the analysis. The results of our modied index and resultant rankings for 108 countries are presented and compared to Cavusgils previous studies. To better understand the sensitivity of this modeling technique, we conducted a second study, holding constant the variables, dimensions and sample, while varying the weights. Then, we discuss those results compared with the others, the limitations and directions for future research opportunities before concluding.
LITERATURE REVIEW
Papadopoulos and Denis (1988) reviewed the literature on the international market selection process identifying qualitative and quantitative as two
222
general approaches to analyzing international markets. They argue that qualitative approaches are more subjective (e.g., Vogel, 1976) and may lead to discrepancies between perceived and objective reality (Johansson & Moinpour, 1977). The issue of quantifying foreign market opportunity has always been a primary concern for decision makers (e.g., Douglas & Craig, 1983; Cavusgil et al., 2004). Papadopoulos and Denis note that quantitative methods have three main advantages: (1) they reduce subjectivity, (2) they allow the rm to consider markets beyond its immediate neighbors, and (3) they make it possible to screen a large number of markets (p. 45). Papadopoulos and Denis divide quantitative approaches into market grouping and market estimation methods. Market grouping methods cluster countries based on their similarity while estimation methods aim to differentiate countries based on market potential. Cavusgils (1997) OMOI and its extension in this study, fall into the quantitative category of market estimation. Boddewyn and Falco (1986) study the size of the market sector around the world. They reexamine research by an economic geographer who concluded that in 1960 perhaps only 10 percent of the worlds population lived in fully commercial countries. Using better data available in the late 1980s, Boddewyn and Falco replicated the survey covering 124 countries and classied them into mutually exclusive categories along a spectrum of market versus non market economies and classied countries into go/no-go categories thereby reducing the number of countries in need of further evaluation. Also, Root (1994) recommends conducting preliminary screening to identify prospective target countries without regard to entry mode, and points out that the purpose of preliminary screening is to identify country markets whose size warrants further investigation. Preliminary screening tries to minimize two errors: (1) ignoring countries that offer good prospects for a companys generic product and (2) spending too much time investigating countries that are poor prospects. He suggests that to minimize the rst error, which is by far the most common, preliminary screening should be applied to all countries. According to Root, one of the problems is that too often managers start with assumptions or prejudices that rule out certain countries (or certain regions) as possible target markets. During the preliminary screening of all countries, Root (1994) recommends using direct estimates of market size and market-size indicators. This preliminary screening/analysis allows rms to identify prospective target countries worthy of further investigation. Russow and Okoroafo (1996) propose a method to screen global markets based on theoretical considerations. Their study addresses the questions of
223
how to order screening relative to the assessment process, what criteria to use to screen countries, and the relative importance of each criterion. They present an initial screening technique to identify markets that are suitable for further assessment, leaving the selection of entry method to a subsequent, in-depth investigation. The technique the authors propose measures demand and selected macro-environmental conditions. The screening criteria used are the costs and availability of the factors of production, the level of economic development, and product-specic market size and growth. Russow and Okoroafos analysis tries to identify the relative importance of the criteria and uses them in screening for potential markets of ve randomly selected products. The study uses principal component and cluster analysis along with multiple discriminant analysis to identify clusters of countries that have high market potential by product category. Their study falls into the category of market grouping methods with its strength being its product specicity. However, a limitation of their grouping techniques is their failure to rank the markets (i.e., which of the 39 countries in the cluster pumps for liquids offer the best opportunity). Moreover, Ivanova et al. (1998) developed a technique to evaluate the potential of Latin American countries, with the principal advantage of the study being its coverage of a large set of social and economic indicators. However, its major disadvantage resides in its limited coverage of countries, as it only measures a set of Latin American countries. Rahman (2000) sought to understand what market indicators, Australian marketing managers used to evaluated markets. He used surveys across industries to arrive at a factor-analyzed set of macro-economic indicators useful for Australian rms. Cavusgil (1997) rst developed the OMOI to measure and rank the market potential for the 23 countries identied by The Economist as emerging markets. As industrialized countries continue to mature, these emerging markets have tremendous growth potential such that sufcient consideration must be given to evaluating their potential. While acknowledging that each step in foreign market expansion is critical, Cavusgil argues that a formal and systematic analysis of aggregate market potential can be particularly fruitful. Cavusgil selected 13 economic, infrastructure, political and social variables to characterize seven dimensions of a markets attractiveness from the Western management point of view (see Table 1 for details). The raw values of these 13 variables were standardized, putting them in a scale of 1 to 100. The standardized data were then used to form the seven dimensions. Weights for each dimension were chosen through a Delphi decision process. A liner compensatory model was used to create the OMOI as the
224
basis for ranking the emerging markets. The indexing approach proposed by Cavusgil provides a systematic tool for reducing the complexity of evaluating emerging markets. Since the publication of his index in 1997, globalEdge at Michigan State University has continued to annually update and publish the OMOI on its web site. The globalEdge (2004) OMOI identies Hong Kong and Singapore as the current rst and second-ranked emerging markets. Table 4 lists the original (Cavusgil, 1997) and more recent (globalEdge, 2004) emerging markets and their OMOI based rankings. Additionally, Cavusgil et al. (2004) combine country clustering and country ranking into a preliminary foreign market assessment and selection process. They reduce 29 macro level variables using principal component analysis to seven factors for input into a cluster analysis. This grouping process identied 10 clusters of markets in terms of macro similarities, (p. 614). Next, Cavusgil et al. (2004) modied Cavusgils (1997) OMOI using 16 variables to form seven dimensions1 as outlined in Table 1. Eighty-nine countries were indexed and ranked (see Table 4 for the list of countries and their rank). They demonstrate the potential for using these two methods in combination. The ranking of the potential markets within each cluster are compared to help identify the best market(s) within each cluster. The major drawback of country clustering for market segmentation has been repeatedly identied as an exclusive reliance on aggregate, macro indicators (Cavusgil & Nevin, 1981; Douglas & Craig, 1983; Papadopoulos & Denis, 1988) at the neglect of specic-product/service market indicators (Cavusgil et al., 2004, p. 608). Product, situation, and consumer related market indicators are typically gathered through primary research that is both expensive and time consuming. Luqmani, Yavas, and Quaraeshi (1994) argue that international markets should be seen as a continuum rather than distinct and mutually exclusive clusters. Therefore, we suggest that it is more appropriate to conduct a market opportunity ranking at the preliminary market assessment stage. Nonetheless, market clustering can be an extremely useful method for segmentation of markets at a later stage of market opportunity assessment when marketing executives can afford to use primary research to attain product/service specic information. Such primary research is more feasible, from time and money perspectives, after the potential set of countries and territories has been reduced from a total topping 208 to a smaller number of prospective markets for further investigation in a later stage of analysis. Since the analysis of Boddewyn and Falco (1986), the world economy has witnessed tremendous change. The transaction of Communist countries to market economies has opened new market opportunities. For instance,
Table 1.
Market indexc Dimensions. Market size
Dimensions and Indicators of Market Attractiveness.

Cavusgil et al. (2004) Studies One and Two One Two
Cavusgil (1997)
Total population 20% GNP/capita PPP Private consumption/capita 15% Growth rate industry (annual average) 15% Telephones/capita (mainline only) Homes with color TV Paved road density Trucks and busses/capita Population/retail outlet
Market intensityeconomic intensity
Urban population Electricity production kwh 24% GDP/capita Urbanization, % of population 12% GDP real growth rate Growth commercial energy use 16% Telephones/capita (mainline only) TVs/capita Radios/capita Internet Hosts/capita Paved road density
Total urban population 15% GDP/capita PPP Energy consumption/capita Electric consumption/capita 10% GDP annual growth rate Secondary school enrollment ratio 10% Telephones/capita (includes wireless) TVs/capita PCs/capita Internet connections/capita Paved road density Merchant shipping eet Railway freight net ton km 15% Economic freedom index 10% Political rights and civil liberties Total imports of goods and services % of GDP Total merchandise imports Geographic distance 10% 15% 5% 10% 5% 10% 10% 15% 25%
Market growthfuture market potential
Commercial infrastructure
Freedom (economic and political) and risk
10% Economic freedom index
Market receptivity market accessibility
10% US imports/capita (40%) Growth rate of US imports 20% Size of the middle class 10% 100%
12% Economic freedom index Survey of political freedom 12% Euromoney country risk ratings 12% US imports/capita Trade as % GDP 12%
15%
20%
Market construction capacity Weights
225
100%
100%
100%
226
Chinas economy is becoming one of the worlds largest markets, and its entry into the World Trade Organization (WTO) will help cut import barriers for foreign products, which will give Western rms better access to over 1.3 billion potential consumers. While the need for market opportunity assessment is great, there are very few studies in the marketing literature that quantitatively analyze and rank the attractiveness of markets around the world. This study seeks to compliment and extend those studies, specically Cavusgils (1997) OMOI, by updating and validating the index and using it to assess and rank the largest sample of countries to date. Given that Cavusgil et al. (2004)2 and this study both seek to amend and extend Cavusgils (1997) OMOI, we clarify the similarities and differences between the three studies in Tables 1, 3 and 4 as well as in the discussion. Dimensions of Market Attractiveness and Opportunity First, Cavusgil (1997) developed his OMOI to quantify and rank the market potential of emerging markets. Most of the 13 variables and the data used in Cavusgils (1997) OMOI were from 1994, more than a decade ago. We suggest some modications in the number and choice of variables to update the dimensions of market attractiveness. The variety of free, quality data available over the Internet has increased markedly since 1994 making it easier to nd appropriate indicators for a large sample of countries. Our dimensions and indicators along with those used in Cavusgil (1997) and Cavusgil et al. (2004) are in Table 1: Secondly, we conducted a primary (Study One) study ranking all the countries on each dimension developing our modied OMOI to use as the basis for an overall ranking of market attractiveness. In a linear compensatory model, a weight is assigned to each dimension reecting its relative importance compared to the other dimensions. Those weights must total 100 percent in order to calculate the overall index (as discussed below). Study Two explores the sensitivity of this market modeling approach to the choice of weights. The different weights used in the liner compensatory models to form the modied OMOIs in Study One and Study Two are in the last two columns of Table 1. The seven dimensions and indicators we use have considerable overlap with Cavusgil and colleagues. Nonetheless, the differences in dimensions, indicators and weights as shown in Table 1 are worth noting. For instance, we broaden the dimension of commercial infrastructure by emphasizing two aspects: telecommunications and physical infrastructure. We added wireless phones and Internet connections per capita as indicators of
227
telecommunications infrastructure, and we also added the merchant shipping eet and railway freight as indicators of physical infrastructure. We also use a secondary school enrollment ratio to indicate future market potential rather than relying on current GDP growth as an indicator of the market growth rate. In addition, we use geographic distance as an indicator of market access for the rst time in an OMOI. Table 1 outlines the dimensions and indicators we use in our studies as discussed next. Market Size The total population of a country gives us important information about the size of a nations markets; however, a countrys urban population is most accessible for marketing (i.e., advertising, sales, distribution, after sale service, etc.). Therefore the percent of the population living in urban areas, called urbanization, is multiplied times the total population to indicate the accessible or market population. For instance, India has a total population of just over 1 billion people but only 28 percent live in urban areas. The effective market population is only 289.5 million. We differ from Cavusgil et al. (2004) in that they combined the size of the urban population with electricity production to estimate market size. Electricity production has most often been used as a measure of industrialization and is more appropriate as an indicator of market intensity, which combines several indicators of production per capita to indicate a nations economic output per person. Market size has a relative weight of 15 percent in our total index for Study One and 25 percent in Study Two. Economic Intensity We use three variables as indicators of the intensity of economic output. This dimension parallels Cavusgils (2004) market intensity but is different enough to label it as economic rather than market intensity. Both studies use GDP per capita adjusted for purchasing power parity (PPP) as an indicator. It represents the per capita GDP in terms of international dollars in order to account for real buying power. Cavusgil and colleagues use urbanization (i.e., the percent of the population that lives in urban areas) as their second indicator of market intensity. However, we focused on economic activity rather than urban population (used in Cavusgils studies to calculate market size). Instead, we chose two additional indicators of economic activity that are free of potential distortion due to concerns over comparability of dollar based measures of GDP and concerns over the strict comparability and the theoretical nature of international dollars (i.e., PPP). However, we do utilize both energy and electric consumption as additional indicators of economic
228
activity, which per capita have been used as reliable indicators of economic development (Sofranko, Nolan, & Bealer, 1975; London, 1987; Bollen, 1979; Mullen, 1993) and as indicators of the relative industrialization of nations (Bornschier & Chase-Dunn, 1985). Economic output has a relative weight of 10 percent in our total index for Study One and 15 percent in Study Two. Future Market Potential Cavusgil et al. (2004) and our study both use GDP annual growth rates to reect a markets recent economic growth. We differed from their use of growth in commercial energy use as a second indicator and chose the secondary school enrollment ratio as an indicator of the longer term future of potential and opportunity. Education has often been shown to be a cause of economic growth (e.g., Jaffee, 1985). The secondary school enrollment ratio (i.e., the percentage of school age children completing secondary education) is used in a number of studies (Douglas, 1971; Stokes & Anderson, 1990; etc.) and has been shown to be a reliable and valid indicator of economic development (Mullen, 1993). Education also provides an indicator of market readiness for advanced products and services. The GDP annual growth from 1998 to 2002 and secondary school enrollment ratio are combined as equal indicators of future market potential with relative weights of 10 percent in our total index for Study One and 10 percent in Study Two. Commercial Infrastructure As noted above, we broaden the dimension of commercial infrastructure by emphasizing two aspects: telecommunications and physical infrastructure, both of which are critical for modern commerce. While Cavusgil et al. (2004) use telephone mainlines/capita, we added cell phones (i.e., wireless) to our list of indicators since wireless communication is becoming increasingly popular and even necessary for modern commerce. There has been substantial evidence that the Internet has a growing positive impact on international trade. Freund and Weinhold (2000) found only weak evidence of the effect of the Internet on trade in 19951996 but found a signicant and increasing impact of the Internet on trade ows from 1997 to 1999. We use the number of Internet users per 10,000 and Cavusgil uses Internet hosts per capita to capture this effect on trade. In addition, we added the number of computers per 100 as a third measure of telecommunications infrastructure that enables Internet usage and embodies infrastructure necessary for modern commerce. Both studies also use televisions per capita. Cavusgil et al. (2004) use one indicator of physical infrastructure paved road density. To that we added measures of the gross registered tonnage of the countries merchant shipping
229
eet and railway freight trafc, measured by net ton kilometers, as three indicators of physical infrastructure. These six measurements are combined to assess commercial infrastructure and have a relative weight of 15 percent in our total index for Study One and 10 percent in Study Two. Freedom and Risk Economic and political freedoms are anatomical to, and necessary for, the smooth functioning of free markets. Political freedom and civil rights increase the prospects for stability and reduce the long-terms risks of operating in a market. Economic Freedom. According to Gwartney and Lawson (2003), The key ingredients of economic freedom are personal choice, voluntary exchange, freedom to compete, and protection of person and property. Economic freedom is reduced when taxes, government expenditures, and regulations are substituted for personal choice, voluntary exchange, and market coordination. The 2003 Fraser Institutes annual report ranks 123 countries representing according to the extent to which they allow their citizens economic freedom. Economic freedom has a relative weight of 10 percent in our total index for Study One and 12 percent in Study Two. Political Freedom and Risk. Political Freedom is included as an important factor in ranking potential foreign markets because it reects the risk of doing business in a particular country. By focusing only on national wealth, consumer income, and peoples propensity to consume, companies ignore the costs and risks of doing business in foreign countries. Karatnyckys (2002) Freedom in the World survey calculates political freedom as an average of each countrys ratings for political rights and civil liberties. According to the 2002 Freedom in the World survey there are 85 free countries in which basic political rights and civil liberties are recognized, and 59 partly free countries with limited respect for political rights and civil liberties. Thirty-ve percent of the global population is living in 48 not free countries in which a broad range of freedoms are systematically denied. Political freedom and risk has a relative weight of 10 percent in our total index for Study One and 12 percent in Study Two. Market Accessibility In line with Boddewyn and Falco (1998), Rahman (2000) and Cavusgil et al. (2004), our focus is from a US rms perspective. However, we deviate from
230
Cavusgil (1997) and Cavusgil et al. (2004) by not including US imports as an indicator of a markets receptivity, reserving that measure to validate the study as a whole nor do we use total trade as an indicator of market openness.3 Total Imports The OECD uses total imports of goods and services as a percent of GDP to indicate open markets. We combine that measure of import intensity with total merchandise imports as indicators of market openness/access from an exporting rms perspective, regardless of where those imports come from. Total imports have a relative weight of 15 percent in our modied OMOI index for Study One and 10 percent for Study Two. Geographic Distance4 Distance still matters and rms must explicitly consider the effects of distance when they make decisions about entering international markets (Ghemawat, 2001). The amount of trade that takes place between countries 5,000 miles apart is only 20 percent of the amount of trade that would be predicted to take place if the same countries were 1,000 miles apart (p. 138). They argue that the further the physical distance, the harder it will be to conduct business due to aspects like increased transportation and communications costs. Bradner and Mark (2002) demonstrated that geographic distance negatively affects a persons willingness to be persuaded by others, reducing the opportunity for new business. Their results further show that greater physical distance reduces peoples willingness to cooperate with others and also makes people more willing to deceive business partners; both of which will lead to increased transaction costs (Williamson, 1981). Therefore, we add geographic distance5 as an additional dimension/indicator of market access. Geographic distance has a relative weight of 15 percent in our total index for Study One and 20 percent in Study Two. Cavusgils studies use seven dimensions to calculate OMOIs while our analysis subdivides market accessibility into two separate dimensions, imports and geographic distance. These eight dimensions of market attractiveness are used as the basis for forming our modied OMOI to rank the worldwide market opportunities in the next section, followed by a validity analysis. Model Validity The model and modied index we develop are created from the perspective of a rm operating in the US and looking for foreign market opportunities. Therefore, we address the face validity of our modied OMOI index by
231
comparing it with the trade ows from the US.6 Three years of trailing imports from the US were averaged and used as the basis to compare our ranking of market attractiveness with actual subsequent trade from the US to those countries.
RESEARCH METHODS AND RESULTS

Sample One hundred and eight countries are included in our sample because their full, current data sets are available. Other countries are unfortunately not included because of insufcient data because of aspects like political unrest (e.g., Afghanistan, Burma, Cuba, Haiti, Iraq and North Korea) or size or lack of development and insufcient resources (e.g., Andorra, Equatorial Guinea, Rwanda, Somalia). Because we are doing our study from a US rms perspective, we explicitly exclude the US from the sample. While this is admittedly a convenience sample, the countries account for 92 percent of the balance of the worlds population and 96 percent of the balance of the worlds GDP.7 The sample contains 24, primarily African nations, that have not been previously indexed and ranked. Also, the 108 countries in the sample are listed alphabetically in the rst column of Table 2. The Data Data for analysis in this research are collected from three international organizations: the IMF, United Nations statistics division and the World Bank statistics division. These sources are considered among the most respectable and reliable with uniform terminology and methodology. They also provide wide coverage of countries and are frequently used. Importantly, the data are updated annually and cover a large variety of economic and social variables. Most of the data are from 2001 to 2003, although some indicators of market intensity are as old as 1998. The rst line of Table 2, in bold, shows the year of the data. Quantitative Analysis Following Cavusgil (1997), the index was created from the raw values of the 18 variables by standardizing the items and placing them into a scale of 1 to 100. As Cavusgil explains in his study, standardization is a statistical process
232
Table 2.
Countries Market Rank Size Market Econom intensity
Extended Overall Market Opportunity Index and Rank: Study One.

Rank Market Growth Rate Rank Commercial Rank infrastructure Freedom and Risk market accessibility Mod. OMOI Overall index rank
Economic Rank Political Rank imports Rank geographic Rank distance 20002002 12 62 14 57 18 47 21 33 54 6 54 6 3 97 46 12 5 88 6 84 17 50 31 19 16 53 4 92 4 92 56 2 4 92 22 31 40 16 56 2 12 62 20 39 35 18 41 15 68 1 11 67 10 72 11 67 36 17 3 97 56 2 53 8 25 26 2003 31 27 23 42 58 54 18 51 24 43 31 32 27 36 20 56 22 57 20 70 31 40 31 47 60 20 27 50 61 26 56 42 29 61 72 87 32 9 17 98 18 83 30 61 57 72 49 90 13 90 12 90 2 61 36 61 26 7 90 72 20 5 77 13 32 69 20012002 49 59 21 88 57 53 42 70 85 3 85 3 49 59 78 15 64 42 71 34 57 53 71 34 42 70 21 88 14 99 85 3 21 88 71 34 7 103 99 1 42 70 78 15 71 34 78 15 85 3 57 53 14 99 64 42 78 15 28 82 85 3 78 15 42 70 20012003 12 55 7 89 4 107 14 49 12 62 23 22 6 100 44 4 7 89 7 93 8 82 20 28 6 100 21 26 7 89 34 10 17 39 10 70 32 11 69 1 6 98 14 49 18 36 24 19 15 45 9 80 7 93 13 54 27 14 9 78 11 64 39 6 12 55
Year(s) of Datac Albania Algeria Argentina Armenia Australia Austria Bangladesh Belgium Benin Bolivia Brazil Bulgaria Burkina Cambodia Cameroon Canada Chad Chile China China, HK Colombia Costa Rica Croatia Czech Republic Denmark Ecuador Egypt El Salvador Estonia Ethiopia Finland France Georgia
2001 1 5 8 1 5 2 8 3 1 2 31 2 1 1 2 6 1 4 100 2 7 1 1 2 2 3 7 2 1 3 2 10 1 97 31 19 86 32 54 18 41 72 54 3 59 86 86 45 24 86 35 1 47 23 86 81 45 63 44 21 67 103 41 72 13 81
19982001 8 107 40 55 31 90 49 42 90 3 70 17 32 85 79 9 33 81 27 100 36 67 48 43 30 96 36 69 33 81 101 2 41 53 44 49 44 50 61 27 32 87 37 65 46 46 64 24 72 15 38 59 34 77 30 93 60 29 39 58 84 5 75 13 37 63
19982002 68 27 59 52 47 79 73 6 72 7 64 42 50 68 65 37 38 95 55 60 57 55 71 12 30 103 40 91 41 89 71 16 42 88 62 47 83 1 60 51 49 72 45 81 65 34 72 9 66 30 54 62 65 37 41 89 72 7 42 87 67 29 69 22 66 31
51 59 58 41 24 58 20 62 46 69 66 51 51 18 40 100 42 62 32 23 87 89 53 59 62 81 42 90 57 28 59 63 44
53 31 35 68 86 35 93 24 60 14 16 53 53 98 72 1 65 24 79 10 7 50 31 24 12 65 6 40 81 31 21 64 28
27 27 29 32 45 48 20 50 25 32 36 38 24 20 19 61 22 38 46 51 32 39 37 45 50 32 24 36 45 20 48 51 30
74 72 62 53 19 12 97 6 78 52 41 37 79 100 102 1 90 36 15 4 51 35 38 18 7 50 80 42 16 99 10 5 60
Germany Ghana Greece Guatemala Guinea Honduras Hungary India Indonesia Iran Ireland Israel Italy Jamaica Japan Jordan Kenya Korea, South Kuwait Laos Republic Latvia Lebanon Lithuania Madagascar Malawi Malaysia Mali Mauritius Mexico Mongolia Morocco Mozambique Nepal Netherlands New Zealand Nicaragua Niger Nigeria Norway Oman Pakistan Panama
16 2 2 2 1 2 2 60 21 10 1 2 9 1 21 2 3 9 1 1 1 2 1 2 1 4 2 1 17 1 4 2 2 4 2 1 1 13 2 1 11 1
8 47 53 54 81 67 51 2 6 15 86 54 17 97 5 63 40 16 86 103 97 72 81 59 97 35 67 106 7 97 32 54 72 34 72 81 86 9 72 86 12 97
70 32 62 30 28 30 54 36 32 46 83 57 66 35 72 37 23 60 88 35 51 39 51 34 16 41 23 47 35 28 40 51 33 77 77 31 34 31 112 73 33 33
19 88 25 95 98 94 37 66 86 46 6 32 21 72 16 63 103 28 4 72 38 57 40 74 106 53 102 44 70 99 55 38 81 12 11 91 74 92 1 14 78 81
63 43 72 34 30 49 71 57 49 53 77 60 68 59 70 64 31 32 40 45 79 57 76 72 28 31 56 64 49 50 62 53 48 66 71 71 50 29 70 70 57 50
44 86 9 99 104 72 16 55 72 63 3 50 28 53 19 40 101 100 92 80 2 55 4 9 106 102 59 40 75 70 47 64 76 31 12 12 70 105 18 19 54 68
50 8 42 9 4 7 30 14 12 15 48 46 51 28 55 23 4 8 20 8 19 18 21 3 3 15 3 19 10 2 11 3 5 23 18 3 2 6 27 14 9 19
10 76 14 74 92 79 21 57 62 55 11 12 9 23 5 29 92 76 39 76 44 47 33 97 97 55 97 44 72 104 67 97 89 29 47 97 104 84 24 57 74 44
55 25 38 30 28 22 42 22 17 6 61 48 50 38 44 39 28 99 40 2 48 31 51 31 24 30 26 34 36 36 35 27 22 55 62 35 24 13 48 38 25 37
15 79 41 67 70 90 32 90 100 105 5 23 20 41 29 39 70 1 36 106 23 61 18 61 83 67 77 55 49 49 52 72 90 15 4 52 83 102 23 41 79 46
78 64 71 49 21 57 78 64 49 14 85 71 78 64 78 28 21 99 35 7 78 21 78 56 49 28 64 78 64 64 28 49 49 85 85 57 42 35 85 21 21 78
15 42 34 59 88 53 15 42 59 99 3 34 15 42 15 82 88 1 78 103 15 88 15 58 59 82 42 15 42 42 82 59 59 3 3 53 70 78 3 88 88 15
58 18 10 8 7 17 26 8 11 7 30 17 30 18 36 22 9 24 11 8 18 12 18 8 12 38 10 18 25 26 11 11 8 39 10 15 6 12 10 10 6 20
3 32 70 82 89 40 16 82 64 93 12 40 12 36 9 25 80 21 64 86 36 55 32 86 55 8 70 32 18 17 64 69 82 6 70 48 98 55 70 70 100 28
60 49 50 91 57 92 55 18 6 37 69 41 58 95 46 41 22 38 35 18 57 42 57 11 19 12 55 1 100 36 63 15 23 63 32 89 50 46 64 26 28 89
58 56 5 40 87 4 44 98 107 74 14 68 35 3 60 68 90 73 77 98 40 65 40 104 95 103 44 108 1 75 21 102 87 21 79 7 56 60 19 85 81 7
54 28 40 31 21 34 41 33 22 22 53 40 48 41 50 30 16 41 30 14 40 26 40 23 17 23 27 28 41 27 30 23 21 48 39 36 24 22 47 28 22 39
2 67 30 56 95 46 22 48 89 88 3 32 11 25 8 58 105 24 57 106 31 76 27 85 104 83 73 65 23 71 59 86 96 13 34 43 81 87 14 68 92 33
Extending and Comparing Cavusgils Overall Market Opportunity Indexes 233
Countries Market Rank Size Market Econom intensity Rank Market Growth Rate Rank Commercial Rank infrastructure Freedom and Risk market accessibility Mod. OMOI Overall index rank
234
Economic Rank Political Rank imports Rank geographic Rank distance 78 76 67 31 37 24 24 49 83 55 44 22 65 12 61 65 24 42 81 83 107 96 34 98 94 5 44 19 83 97 34 93 108 11 5 7 17 20 14 17 29 6 21 20 21 7 25 16 12 27 21 5 2 20 6 13 11 2 24 21 31 22 7 7 12 67 89 79 50 39 57 50 22 84 33 39 33 79 26 53 62 24 33 89 104 39 84 61 67 104 28 33 19 32 79 79 62 25 37 32 38 47 19 24 32 34 64 46 39 38 49 32 58 59 14 27 37 18 35 25 40 23 42 60 43 8 13 23 79 46 57 41 26 97 83 57 55 3 28 39 41 22 57 9 9 101 72 46 98 52 79 36 87 32 7 30 104 102 87 49 71 64 78 85 71 28 0 49 28 78 78 78 78 49 85 85 0 42 64 28 21 35 21 42 21 78 85 42 7 35 14 59 34 42 15 3 34 82 106 59 82 15 15 15 15 59 3 3 106 70 42 82 88 78 88 70 88 15 3 70 103 78 99 12 5 18 13 15 13 11 9 12 62 27 19 10 23 13 2 17 20 10 7 24 15 16 12 7 17 22 41 6 5 20 8 5 55 106 32 51 45 51 64 78 62 2 15 31 70 22 51 108 40 28 77 93 19 45 44 55 93 40 24 5 100 104 27 86 104 62 77 23 58 66 53 54 35 60 7 55 58 20 64 7 36 60 59 41 21 16 46 54 48 27 53 28 66 54 86 19 22 19 24 13 87 35 16 50 47 77 28 105 44 35 93 19 105 75 28 31 68 92 101 60 47 59 84 50 81 16 47 11 95 90 95 29 32 27 37 41 33 34 27 28 34 40 40 28 43 21 45 43 18 23 26 21 29 24 20 37 32 49 33 28 22 20 63 54 75 40 26 47 45 70 66 44 29 28 64 20 94 17 21 103 84 77 93 61 82 98 39 55 9 48 69 91 101
Paraguay Peru Philippines Poland Portugal Romania Russia Saudi Senegal Singapore Slovak Republic Slovenia South Africa Spain SriLanka Sudan Sweden Switzerland Syrian Arab Republic Tanzania Thailand Togo Tunisia Turkey Uganda Ukraine United Arab Emirat United Kingdom Uruguay Venezuela Vietnam Zambia Zimbabwe
2 5 11 6 2 3 23 5 2 2 2 1 6 7 2 4 2 2 3 3 4 1 2 10 2 8 1 12 2 5 5 2 2
72 28 11 26 50 37 4 28 59 63 72 103 25 22 63 63 47 59 43 39 37 86 52 14 67 20 86 10 72 27 30 67 67
36 27 33 43 54 47 62 65 34 57 56 65 50 54 20 38 80 67 35 70 38 29 43 21 46 78 81 58 31 38 59 33 6
67 101 80 51 36 45 26 23 74 33 34 22 41 35 105 61 8 20 70 17 59 97 52 104 48 10 7 31 89 61 30 78 108
47 48 51 66 65 69 69 61 44 57 63 69 52 71 54 52 69 64 45 44 24 36 65 35 38 74 63 70 44 35 65 39 20
235
enabling us to directly compare variables with very different measures and distributions. The original data are rst transformed into standardized data, or z-scores, that are used to form an index for each dimension. This index is sorted from largest to smallest in order to rank the 108 countries on each dimension as summarized in Table 2. The dimensions (i.e., market size, market/economic intensity, etc.) are listed across the top of the table. There are two columns beneath each dimension. The rst is the standardized index for each country. The next column is the rank order of the markets attractiveness on that dimension. The rankings of the eight separate dimensions in Table 2 indicate that China is ranked number one, as expected, for size and market potential/ growth but drops to 90th and 10third for economic freedom and political risk. Canada and Mexico are naturally closest geographically to the US followed by nations from the Caribbean basin (i.e., Jamaica and Venezuela) and from Central America (i.e., Honduras, Guatemala and Costa Rica). Denmark is shown to have the best ranking for commercial infrastructure and Hong Kong has the highest ranking for imports. This is undoubtedly due to its strategic position as the gateway to mainland China. Norway is the top ranked market for economic intensity, reecting its high economic output per capita. Study One Finally, a linear compensatory model is used to calculate our modied OMOI. We multiply the weights by the indexes for each of the eight dimensions and add the results together to gure the OMOI for each country. Sudan and Zimbabwe had data to form indexes for six of the eight dimensions and are ranked on those dimensions, but they are not included in the calculations for the OMOI and overall rankings because they are missing data on the other two dimensions. Our modied OMOI from Study One is then used to rank the other 106 countrys overall market attractiveness, relative to the other countries in the study. Table 2 shows the OMOI for the 106 countries in the second to last column and the rank order of market attractiveness in the last column. For instance, Table 2 shows that Canada is ranked second in economic intensity and commercial infrastructure and third in political risk taken together with no physical distance to yield an overall rst place in the OMOI assessment of market attractiveness for US rms, corresponding to its role as Americas leading trading partner. A Comparative Analysis: Study Two In order to sort out the sensitivity of the linear compensatory model to changes in the weights used to calculate OMOIs, we conducted a second
236
study using the same indicators, dimensions and countries as in Study One. The only variable that is changed is the weight assigned to each dimension. We increased the weights for market size to 25 percent and for market/ economic intensity to 15 percent following Cavusgil (1997) and colleagues (2004). Further we increased the weight for geographic distance from 15 to 20 percent to improve the t of the model to the validity coefcient. Consequently, the weights for commercial infrastructure, both aspects of freedom and risk, and imports are reduced to 5 percent each to keep the total equal to 100 percent. The weights used to calculate the modied OMOI for Study Two are shown in the last column of Table 1 above. Because the weight is the only variable in the analysis that is changed, it is possible to isolate differences due strictly to changes in the weights. The results provide some insight into the sensitivity of liner compensatory models to different weights. The resulting rankings from Study Two are reported in Table 3 along with Cavusgil (1997), globalEdge (2004), Cavusgil et al. (2004) and Study One. As can be seen in Table 3, 12 countries showed no change in ranking between Studies One and Two. Just over two-thirds (68 of 106) moved up or down by ve or less ranks in the market attractiveness ratings. Eleven countries improved their rankings by 10 or more led by Russia, Brazil, Venezuela, Iran, China and India, all beneting by the increased emphasis on size, and Venezuela by its geographic proximity to the US. China improved from 15th to 2nd overall with the increased weight for size given that almost a half billion of their 1.3 billion population live in urban areas. Seven countries (and Hong Kong) fell by 10 or more ranks lead by Mauritius, Singapore and South Korea due to increased emphasis on size and distance and a reduced weight for imports. Validity It is important to address the face validity of the modied OMOI by comparing it to subsequent trade ows. The modied OMOI is used to evaluate and rank the attractiveness of 106 foreign markets from a US perspective (i.e., the US is not in the sample). The rank order of the dollar volume of three years of subsequent US exports to the 106 countries is compared to the rankings derived from OMOI values. For data sets in which both rows and columns contain ordered values, Spearmans correlation coefcient, rho, is most appropriate as a measure of association between rank orders and is proposed as a validity coefcient for OMOIs. For Study One, Spearmans rho is 0.61 and 0.63 for Study Two; both statistically signicant at the 0.000 level (a 0.05) demonstrating reasonable validity for our modied OMOIs.8
237
Table 3. Countries and Rankings by Study.

Countries Cavusgil (1997) globaledge (2004) Cavusgil et al. (2004) 80 80 39 61 6 22 70 85 76 9 49 29 65 Study One Study Two
Albania Algeria Argentina Armenia Australia Austria Azerbaijan Bangladesh Belarus Belgium Benin Bolivia Brazil Bulgaria Burkina Cambodia Cameroon Canada Chad Chile China China, HK Colombia Costa Rica Croatia Czech Republic Denmark Dominican Republic Ecuador Egypt El Salvador Estonia Ethiopia Finland France Georgia Germany Ghana Greece Guatemala Guinea Honduras Hungary
10
19
74 72 62 53 19 12 97 6 78 52 41 37 79 100 102 1 90 36 15 4 51 35 38 18 7 50 80 42 16 99 10 5 60 2 67 30 56 95 46 22
75 61 62 57 21 14 99 8 83 56 25 42 84 102 97 1 86 38 2 18 47 32 46 17 10 48 73 41 19 98 11 5 67 3 70 30 50 88 45 28
15
20
1 13 2 3 10 4 1 23 25 2 12 61 38 44 34 16 74 65 55 28 9 14 49 7 76 32 65 65 34
18
7 54 16
14
238
Countries Cavusgil (1997) 12 20 globaledge (2004) 8 17 Cavusgil et al. (2004) 21 76 5 24 22 3 58 88 17 43 39 44 Study One Study Two
India Indonesia Iran Ireland Israel Italy Jamaica Japan Jordan Kenya Korea, South Kuwait Lao Republic Latvia Lebanon Lithuania Madagascar Malawi Malaysia Mali Mauritius Mexico Moldavia Mongolia Morocco Mozambique Nepal Netherlands New Zealand Nicaragua Niger Nigeria Norway Oman Pakistan Panama Paraguay Peru Philippines Poland Portugal Romania Russia
14
33
11
11
34 73 81 65 59 88 11 20
48 89 88 3 32 11 25 8 58 105 24 57 106 31 76 27 85 104 83 73 65 23 71 59 86 96 13 34 43 81 87 14 68 92 33 63 54 75 40 26 47 45
36 89 74 4 40 12 22 7 66 105 42 53 104 34 71 33 96 106 94 69 85 15 78 55 92 100 13 39 37 77 80 6 65 87 31 64 51 76 44 26 49 29
84 15 86 46 70 55 52 34 25 74 27
23 16 9 22
21 18 9
12
239
Countries Cavusgil (1997) globaledge (2004) Cavusgil et al. (2004) 53 81 4 41 29 47 17 70 12 17 103 47 57 49 61 55 9 41 60 76 87 Study One Study Two
Saudi Arabia Senegal Singapore Slovak, Republic Slovenia South Africa Spain SriLanka Sudan Sweden Switzerland Syrian Arab Republic Tanzania Thailand Togo Tunisia Turkey Uganda Ukraine United Arab Emerates United Kingdom Uruguay Venezuela Vietnam Yemen Zambia Zimbabwe Sample size
19
22
83 6 13
21
15
31 7 17
70 66 44 29 28 64 20 94 6 of 8 17 21 93 84 77 93 61 82 98 39 52 9 48 69 91 101 6 of 8
60 68 63 35 26 72 20 103 16 23 79 91 90 58 82 95 24
59 54 81 101 108
23
23
89
108
Country changes from Cavusgil (1997) to globalEdge (2004): dropped Greece, Portugal and Venezuela and added Egypt, Peru and Columbia.
Visual examination of the results indicated several unusual cases, so we ran a simple regression of OMOI on US imports and examined DfBeta, Standardized DfBeta and covariance ratio statistics to identify inuential outliers (for a full discussion of outlier detection and analysis (see Belsley, Kuh, & Welsch, 1980; Mullen, Milne, & Doney, 1995). Ethiopia and Malaysia were identied as especially inuential outliers followed by Thailand and Slovenia. For Study One, the validity coefcient improved from 0.61 to 0.65 without the rst two outliers, and further to 0.67 by holding out all four from the correlation analysis. The validity coefcients
240
for Study Two improved from 0.63 to 0.67 and from 0.65 to 0.70, without the two and then four outliers. The modied OMOIs developed in Studies One and Two demonstrate reasonable validity, especially Study Two without the four outliers. For the remaining 102 countries, the 0.70 validity coefcient provides some evidence of the external (face) validity of these modied OMOIs and the resultant rankings of attractive markets. Please note that we do not drop the four countries from the rankings but caution that their status as inuential outliers indicates the need for further analysis before accepting their rankings at face value. Comparative Results The 30 best prospect markets for American rms (based on Study Two because of the high 0.70 validity coefcient) are listed in Table 4. For comparison purposes, the ranking of those markets from Study One and Cavusgil et al. (2004) are also shown. All three studies rank Canada as #1 in overall attractiveness for US rms in line with its status as the number one market for US exports. However, they are somewhat off the mark for Americas second largest export market, Mexico, with Study Two ranking it at 15, Study One at 23 and Cavusgil et al. at 34. Nonetheless, there is substantial correspondence across the three studies on the top 30 prospective markets. With 90 (30 3) potential top 30 rankings, only eight do not match and four of those mismatches are very close (i.e., Cavusgil et al. rank Greece at 32 and tie Russia, the Check Republic and Hungary at 34). Furthermore, all three studies include the following six countries in the list of the top 10 most attractive markets for US rms: Canada, Germany, Japan, Belgium, UK and Ireland. The Ukraine has the greatest deviation from its 24th rank in Study Two, compared to 39 in Study One and a distant 61st place in Cavusgil et al. It is worth noting that among our top 30, Estonia and Slovenia are outliers indicating that there is a substantial difference between their ranks and their actual imports from the US. Furthermore, it appears that the rankings of the Ukraine are less than reliable. Therefore closer evaluation of those markets is necessary before concluding that they are or are not high prospect export markets.
DISCUSSION
The modied OMOIs developed, validated, and compared here add to the growing literature on ranking the best prospect foreign markets in order to
241
Table 4.
Country Canada China Germany Ireland France Norway Japan Belgium United Kingdom Denmark Finland Italy Netherlands Austria Mexico Sweden Czech Republic China, HK Estonia Spain Australia Jamaica Switzerland Ukraine Brazil Portugal Slovenia Hungary Russia Greece
The 30 Most Attractive Markets for US Firms.

Study One 1 15 2 3 5 14 8 6 9 7 10 11 13 12 23 17 18 4 16 20 19 25 21 39 41 26 28 22 45 30 Study Two 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Cavusgil et al. (2004)* 1 2 7 5 14 15 3 9 7 16 9 22 11 22 34 12 34 12 28 17 6 17 61 29 25 29 34 27 32
help rms in the process of identifying the right markets to enter. Given the reasonable external validity and the overlap in the top 30 rankings from all three studies, it behooves managers considering entering new markets to focus scarce resources to further analyze the top 10 or top 30 markets, depending on rm resources and constraints. For rms already in foreign markets, comparison with the list of the top 30 markets may give insights to those that warrant further investigation for future expansion. Cavusgil et al. (2004) analyzed 84 of the same countries as those in our studies making comparisons possible. Nonetheless, there are ve countries (e.g., the Dominican Republic) in Cavusgil et al. that are not in the current
242
analyses and there are 24, primarily African countries, in our studies that are not in Cavusgil et al. making direct comparison of the results problematic. While there is a substantial overlap in variables and the weights for some dimensions such as economic freedom, there are substantial differences in indicators and weights for other dimensions making it difcult to identify the cause of the variation in the rankings (see Table 1 for details). For instance, we used total urban population as the indicator of market size and Cavusgil et al. used urban population plus electricity production. The most pronounced differences in variables are those used to assess market receptivity or accessibility. Cavusgil et al. used imports from the US per capita and total trade as a percent of GDP as their indicators. Because we used imports from the US to validate the studies, we rely on total imports for goods and services as a percent of GDP rather than just those form the US. We also used geographic distance as an additional (inverse) indicator of market accessibility. In addition, the dimensions are somewhat different (Cavusgil: 7 with 16 indicators versus ours: 8 with 18 indicators) with different weights in the three studies. Further, most of the data are from different years. So while the comparisons between different studies are interesting, it is hard to know whether the source of variation in country rankings is due to the differences in samples, variables or weights. Nonetheless, it is interesting to note in Table 3 that emerging markets near the top in Cavusgil (1997) and globalEdge (2004) are moderately positioned in our research. For instance, Cavusgil (1997) and globalEdge (2004) rank Singapore 2nd and 1st while our Studies One and Two rank it 44th and 63rd. Also they rank South Korea fourth and third compared to our rankings of 24th and 42nd. There are very substantial differences in the make up and numbers of countries in the emerging market samples from Cavusgil (1997) and globalEdge (2004) compared to Cavusgil et al. (2004) and Studies One and Two (see Table 3). As a result, there are large differences for countries such as Venezuela that ranked 17th in Cavusgil (1997), 60th in Cavusgil et al. (2004) and 69th and 54th in Studies One and Two. Furthermore, Egypt (16, 65, 8073), Indonesia (2017, 76, 8989) and Malaysia (814, 33, 8394) show similar patterns in Table 3 to that of Venezuela where the rankings suffer when included in the larger sample of countries representing 96 percent of the worlds net GDP. As this discussion of the results demonstrates, substantial differences in the samples render direct comparison of the results subjective, at best. Nonetheless, an important implication is that no matter how important the emerging markets are for the future, rms cannot afford to overlook the market opportunities in neighboring countries and other wealthy OECD member countries.
243
The geographic distance measure was added as an important indicator of market accessibility for the rst time in an OMOI improving the face and predicative validity of the index. We use straight-line distance from the closest of one of three major US port cities to measure distance in our analysis. Other researchers have used straight-line sea and land distances but multiply the land distances by a factor of two (e.g., Bergstrand, 1985). Regardless, studies that include geographic distance (Bergstrand, 1985; Ghemawat, 2001; Bradner & Mark, 2002), including this one, have found it to be an important predictor of trade ows. The face or external validity of OMOIs has not been adequately addressed in the literature. We compared the average of three years of subsequent US imports by country to the OMOIs and their rankings with Spearmans rho to assess the validity of our modied OMOIs. The validity coefcients showed a reasonable t of the model to the market and provide a baseline for future research. Where there are signicant differences between the OMOI rankings and the actual US imports, additional thought and research is necessary to sort out the cause before rms make market entry decisions. For instance, if the ranking indicates much higher potential than actual US imports, there may be untapped market opportunity or markets closed to trade, possibly requiring FDI to enter. Where trade ows are signicantly higher than predicted, the model may need to be modied to account for the discrepancy, unless research demonstrates idiosyncratic cause(s). Canada, Mexico and the UK deserve such further consideration from a US perspective. Colonial ties, preferential trade arrangements (i.e., FTAs) and land borders have all been shown to be positively related to bi-lateral trade ows (Bergstrand, 1985; Frankel & Rose, 2000; Ghemawat, 2001). Canada and Mexico, Americas number one and two trading partners share long land borders with the US and preferential trade arrangements through NAFTA. In addition, Canada, Britain and the US share common colonial ties. The UK is the USAs fourth largest export market while our OMOIs ranked it ninth. The UKs imports are almost twice (1.93) those of the number nine US importer, France. As noted above, Mexico is ranked 23rd and 15th most attractive markets by our studies and 34th by Cavusgil et al. (2004) but is clearly the number two importer of US goods and services. While Canada is ranked number one in all studies, its OMOIs understate its importance as a US market (see Table 1). Canadas imports of US products and services are about equal to Japanese (#3), German (#5), Chinese (#6), South Korean (#7), Dutch (#8) and French (#9) US imports combined. Future international marketing models need to address preferential trade arrangements,
244
land borders and colonial ties along with the other issues raised in these analyses. Limitation and Future Research The quantitative analyses exemplied in these studies provide a valuable tool to reduce the overall uncertainty associated with the preliminary identication of high-potential foreign markets. Nonetheless, the rank and the index of the countries are not permanent and are expected to vary over time with changes in the national world economies. These and other limitations are discussed next. Limitations There are several limitations of the modied OMOI as applied here. First, data are not available on many nations. The World Bank currently lists 208 countries and/or territories and we only evaluated 108. While an advantage of this study is that it looks at world markets from the perspective of rms from one particular country, a weakness of the analysis for rms located in other domestic markets is that it excludes the US from the analysis. Further, this modeling approach assumes that one size ts all, so to speak. Product and service specic analysis would likely identify alternative solutions (see Russow & Okoroafo, 1996). For instance, rms selling consumer durables to households may need to adjust population gures by the number of people per household. As the size of households varies widely by country (i.e., from just under two to more than eight people per household, United Nations, 2004), it cannot be assumed that relative population size is a fair proxy for the relative number of households. In both of Cavusgils OMOIs and the modied version developed here, GDP numbers are adjusted for PPP to convert them to international dollars. The intention is to try to eliminate some of the well-known distortions of money-based GDP gures caused by uctuating exchange rates and underreported national income due to non-cash and black markets, tax evasion, etc. Nonetheless, there are very substantial problems with the lack of uniform estimation procedures and time frames used to develop PPP data. For instance, PPP data for China relies on a study done more than 20 years ago by two academics with a different methodology then currently in use. Further, there is no attempt to measure the differences in quality and no consistent standard to compare product features (i.e., the question raised is whether a car in Brazil is the same as one in Germany or the US). Lastly,
245
international dollars are theoretical and do not exist to actually buy products or services from any countries rms. These limitations, and the other limitations discussed above, provide ample opportunity for future research into this most important area of international marketing enquiry. Future Research Economy development is a dynamic process that may result in a necessary periodic updating of appropriate market indicators. Therefore, the indicators used to measure the dimensions of the OMOI should be improved and modied over time. These OMOI analyses can be reproduced and or extended in the future with new data, indicators, samples and/or weights to reect trends and observed changes in international markets. The number of factors and indicators varied considerably across the 1997, 2004 and current studies as outlined Table 1. While we tried to be clear about our motivations for variable selection, greater attention might be paid in future research to theoretical denitions of constructs/dimensions leading to more consensus on appropriate indicators. Only one country ranking study, Rahman (2000), has done quantitative measurement analysis using CFA and it was only for one construct/dimension. Clear denitions of constructs could lead to improved measurement analysis and standards that might provide a common language for marketing scientists studying foreign market opportunity analysis. Industry specic foreign market opportunity analysis leading to industry by IMOI and rankings could make a valuable contribution to international marketing research as well. While primary or expensive secondary data may be required, an industry focus might facilitate the necessary funding. Firms within an industry could easily tailor such an industry index into one tting its own circumstances. Industry and eventually rm-specic models/indexes/ rankings would be very useful to marketing managers seeking practical international marketing research techniques for making a critical international marketing decision. The sample excluded the US, so rms in other nations may want to modify this approach to add the US back in and to drop their domestic market from the analysis. While it is always appropriate for rms based in another market to add the US to the analysis, it is sometimes more helpful to do the analysis with and without the US. Given that the US has about one-third of the worlds GDP and a high level of economic and technical development, it tends to swamp other markets when it is included in an analysis. This results in making it more difcult to realize
246
important differences in the indexes among the other markets under consideration. While we introduced geographic distance, various measures (i.e., from economic activity centers) and methods may improve the t of the model. For example, we have to differentiate between having a geographically common land border compared to measuring a physical distance separating countries. Together with more precise distance measures, models that include FTAs and colonial ties may help future international market researchers attain better predictive validity of market attractiveness with fewer outliers.
CONCLUSION
The imperatives of globalization are clear. Firms must look to expand into international markets to survive and thrive. Indeed, failure to internationalize may eventually threaten a rms domestic dominance since not going international incurs severe competitive risks (Tallman & Yip, 2001). Choosing the right market(s) has been identied as a critical decision, rendering the approach to that choice an important area of international market research. This research complements and extends a growing body of work developing OMOIs and using them to rank the attractiveness of potential markets. Those rankings help rms choose the right markets to enter by allowing them to focus scarce resources on a limited set of best prospect countries. Our analyses provides a current analysis of market attractiveness and opportunity for the largest set of countries indexed and ranked to date, including 24 countries not in previous OMOI studies. Furthermore, it compares the dimensions, variables, samples and results from three of Cavusgil and colleagues previous studies and the two conducted herein. The market attractiveness rankings in our analyses demonstrate reasonable external validity based on the validity coefcient and on the fact that the top 10 and top 30 ranked markets overlap considerably across our studies and Cavusgil et al. (2004). Nonetheless, there are important differences between the outcomes. The choice of sample, variables and weights are shown to directly affect the OMOIs and rankings. OMOIs are shown to be a exible, dynamic, efcient and reasonably valid tool for preliminary analysis of foreign market opportunity. The short list of most attractive countries will help rms focus scarce resources and avoid the common errors of wasting time and money on low prospect markets, while or overlooking less visible high potential markets thereby helping rms efciently identify the right foreign markets to enter.
247
NOTES
1. Seven factors with 19 variables are used in the cluster analysis, six of these variable and 10 are to form seven different dimensions for the OMOI rankings. Only six of those variables overlap with the 19 used to form the PCA factors. Cavusgil et al. (2004) differentiate by calling one set factors and the other dimensions. 2. We independently completed our data collection and analysis in 2004 before we became aware of Cavusgil et al. (2004). 3. While total trade as a percentage of GDP has been used by Cavusgil and others as an indicator of market openness, it is the sum of exports and imports over GDP, but in this case we are only interested in a nations receptivity to imports rather than its ability to export. 4. We gratefully acknowledge the reviewer that recommended that we consider distance as an important dimension in identifying foreign markets. 5. Geographic distance is measured from the closer of Los Angles, New York City or Miami with the exception of Canada and Mexico, where long land borders eliminate physical distance between countries. 6. We thank two anonymous reviewers for suggesting that we validate the results by comparing actual trade ows from the US to our modied OMOI. 7. The USAs population and GDP are not included in the totals. 8. We did not assess the validity of Cavusgil et al.s (2004) rankings by correlating them with US imports because that is one of their measures of market attractiveness. The argument would be tautological.
ACKNOWLEDGMENT
The authors wish to thank two anonymous reviewers and the participants of the 13th Annual CIMaR Seminar, Universitat Automoma de Barcelona, Spain for their helpful comments and suggestions and Cynthia Lyles-Scott for her professional editing. Michael Mullen is the 2005 Fulbright Scholar in International Business at the Dublin Institute of Technology, Ireland, and an Associate Professor of Marketing and International Business at Florida Atlantic University. Shirley Ye Sheng is a marketing doctoral student at Florida Atlantic University.
REFERENCES
Belsley, D. A., Kuh, E., & Welsch, R. E. (1980). Regression diagnostics: Identifying data and sources of collinearity. New York: Wiley. Bergstrand, J. H. (1985). The gravity equation in international trade: Some microeconomic foundations and empirical evidence. The Review of Economics and Statistics, 67(3), 474481.
248
Bilkey, W. J. (1978). An attempted integration of the literature on the export behavior of rms. Journal of International Business Studies, 9(1), 3346. Bilkey, W. J., & Tesar, G. (1997). The export behavior of small sized Wisconsin Manufacturing Firms. Journal of International Business Studies, 8(1), 9398. Boddewyn, J. J., & Falco, J. J. (1998). The size of the market sector around the world. Journal of Macromarketing, 8(1), 3242. Bollen, K. J. (1979). Political democracy and the timing of development. American Sociological Review, 44, 572587. Bornschier, V., & Chase-Dunn, C. (1985). Transnational corporations and underdevelopment. New York: Preager. Bradner, E., & Mark, G. (2002). Why distance still matters: Effects on cooperation, persuasion and deception. Proceeding on the ACM 2002 conference on computer supported cooperative work (pp. 226235). New Orleans, LA, US. Cavusgil, S. T. (1980). On the internationalization process of rms. European Research, 8(6), 273280. Cavusgil, S. T. (1997). Measuring the potential of emerging markets: An indexing approach. Business Horizons, 40(1), 8791. Cavusgil, S. T., Kiyak, T., & Yeniyurt, S. (2004). Complementary approaches to preliminary foreign market assessment: Country clustering and country ranking. Industrial Marketing Management, 33, 607617. Cavusgil, S. T., & Nevin, J. R. (1981). State of the art in international marketing: An assessment. In: B. M. Enis & K. J. Roering (Eds), Review of Marketing (pp. 195216). Chicago: AMA. Czinkota, M. R., & Ronkainen, I. A. (2004). International marketing (7th ed.). Mason, Ohio: Thomson-Southwest. Douglas, S. P. (1971). Patterns and parallels of marketing structure in several countries. MSU Business Topics, 19, 3848. Douglas, S. P., & Craig, S. C. (1992). Advances in international marketing. International Journal of Research in Marketing, 9, 291318. Douglas, S. P., & Craig, S. C. (1983). International marketing research. Englewood Cliffs, NJ: Prentice-Hall. Frankel, J. & Rose, A. (2000). An estimation of the effects of currency unions on growth. Unpublished working paper as presented and discussed in Ghemawat (2001). Freund, C. & Weinhold, D. (2000). On the effects of the Internet on International Trade. Board of Governors of the Federal Reserve System, International Finance Discussion Papers, Number 693. Ghemawat, P. (2001). Distance still matters: The hard reality of global expansion. Harvard Business Review 79, 8(September), 137147. globalEdge (2004) accessed at http://globaledge.mus.edu/irid/marketpot.asp. Green, R. T., & Allaway, A. W. (1985). Identication of export opportunities: A shift share approach. Journal of Marketing, 49, 8388. Gwartney, J. D. & Lawson, R. A. (2003). Economic Freedom of the World: 2003 Annual Report. The Fraser Institute. Jaffee, D. (1985). Export dependence and economic growth: A reformulation and respecication. Social Forces, 64, 102118. IMF statistics division: http://www.imf.org/external/data.htm
249
Ivanova, I. M., Arcelus, F. J., & Srinivasan, G. (1998). Assessment of the competitiveness position of the Latin American countries. International Journal of Commerce and Management, 8(2), 732. Johansson, J., & Moinpour, R. (1977). Objective and perceived similarity of Pacic Rim Countries. Columbia Journal of World Business, 12(4), 6576. Karatnycky, A. (2002). Freedom in the world: The annual survey of political rights and civil liberties 20012002. New York: Freedom House. Kotabe, M. (2001). Contemporary research trends in international marketing: The 1990s. In: A. Rugman & T. Brewer (Eds), The Oxford handbook of international business (pp. 457 502). Boston, MA: Oxford University Press. Kothari, V. (1978). Strategic approaches of small US manufacturers in international markets. Proceeding of the Academy of Management, August, 8. Kumar, V., Antonie, S., & Joachimsthaler, E. (1994). An interactive multicriteria approach to identifying potential foreign markets. Journal of International Marketing, 2(1), 2952. London, B. (1987). Structural determinants of Third World urban change: An ecological and political economic analysis. American Sociological Review, 52, 2843. Luqmani, M., Yavas, U., & Quaraeshi, Z. (1994). A convenience-oriented approach to country segmentation. Journal of Consumer Marketing, 11(4), 2940 (as reported in Cavusgil et al. (2004)). Mullen, M. R. (1993). An empirical analysis of the effects of exporting and importing on two dimensions of economic development. Journal of Macromarketing, 13(1), 319. Mullen, M. R., Milne, G., & Doney, P. (1995). An international marketing application of outlier analysis for structural equations: A methodological note. Journal of International Marketing, 3(1), 4562. Papadopoulos, N., & Denis, J. E. (1988). Inventory, taxonomy and assessment of methods for international market selection. International Marketing Review, 5, 3851. Rahman, S. (2000). Towards developing an international market selection decision. ANZMAC 2000 Conference: Visionary Marketing for the 21 Century: Facing the Challenge, 15. Root, F. R. (1984). Some trends in the World-economy and their implications for international business strategy. Journal of International Business Studies, 15(3), 1923. Root, F. R. (1994). Entry strategies for international markets. New York: Lexington Books. Russow, L., & Okoroafo, S. C. (1996). On the way toward developing a global screening model. International Marketing Review, 13(1), 4664. Sofranko, A. J., Nolan, M. F., & Bealer, R. C. (1975). Energy use and alternative measures of societal modernity. Sociology and Social Research, 59, 301317. Stokes, R. G., & Anderson, A. B. (1990). Disarticulation and human welfare in less developed countries. American Sociological Review, 55, 6374. Tallman, S. B., & Yip, G. (2001). Strategy and the multinational enterprise. In: A. M. Rugman & T. L. Brewer (Eds), The Oxford handbook of international business (pp. 317348). Oxford University Press. United Nations statistics division: http://unstats.un.org/unsd/databases.htm Vogel, R. H. (1976). Uses of managerial perceptions in clustering countries. Journal of International Business Studies, 7(1), 9199. Williamson, O. E. (1981). The economics of organization: The transaction cost approach. American Journal of Sociology, 87, 548577. World Bank statistics division: http://econ.worldbank.org
CITY-OF-ORIGIN EFFECTS IN THE GERMAN BEER MARKET: TRANSFERRING AN INTERNATIONAL CONSTRUCT TO A LOCAL CONTEXT
Patrick Lentz, Hartmut H. Holzmuller and Eric Schirrmann
ABSTRACT
Irrespective of the popularity of country-of-origin research in international marketing, no attention has been paid to effects which stem from the declaration of a products local origin, like Made in City X. In this study, insights from country-of-origin research as well as exploratory qualitative studies are used to model determinants of preference for local products. Conjoint analysis results based on a sample of consumers from three neighboring cities in Germany show the importance of local origin for product preference. Structural equation analysis sheds rst light on the mechanism of city-of-origin effects.
International Marketing Research: Opportunities and Challenges in the 21st Century Advances in International Marketing, Volume 17, 251274 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1474-7979/doi:10.1016/S1474-7979(06)17009-X
251
252
PATRICK LENTZ ET AL.
INTRODUCTION
Traditionally, the places of production of a good as well as the location where a service is provided are important attributes by which customers assess their quality. Consequently, the country of origin of goods is a wellresearched topic in the international marketing literature. In their seminal review of the eld, Heslop and Papadopoulos (1993) identied more than 400 studies in the respective eld and the phenomenon still attracts prominent scholars (Agrawal & Kamakura, 1999; Nijssen & Douglas, 2004; Verlegh, Steenkamp, & Meulenberg, 2005). However, origin effects that stem from other geographical units like trade zones, intra-national areas, and cities have been largely neglected. Specifically, from a managerial perspective a deeper understanding of different kinds of origin effects would help to make use of or deal with these market-relevant phenomena better. On the one hand, a better knowledge of city-of-origin effects would help local marketers to rene their strategies against national or international competitors. International marketers, on the other hand, would benet because they are better able to assess mechanisms export or international markets which shape the competitive environment. Strong city-of-origin effects in specic industries would for instance lead to higher entrance barriers in those markets, and corresponding strategies allowing for a successful entry have to be developed. Acquisition of existing companies or brands instead of exporting or starting greeneld operations as well as using more localized promotional messages are examples of more suitable and appropriate approaches, which might nally lead to a quicker penetration of the target market by avoiding or using city-of-origin effects, respectively. Although the country is only one of many possible geographical entities (see Fig. 1), it was the only category that has received significant attention in origin research. On a more aggregate level a group of countries (e.g., EU, NAFTA) or even a continent could be used to denominate a products or a services heritage. The same applies to intra-national geographic units like federal states or specic sub-regions (e.g., mountain range, costal area) within a country. Specifically, intra-national units have traditionally been used for agricultural products like wine, meat, and vegetables to signify product quality (Muller & Kesselmann, 1996; van Ittersum, Candel, & Meulenberg, 2003; Trognon, Lagrange, & Janin, 1999; van Ittersum, 2002; Verlegh et al., 2005). In addition to that, large cosmopolitan cities such as New York, Paris, Tokyo, or Rome, as well as smaller cities known for a certain specic expertise such as Solingen for knives, Ceski Budejovice for beer, and Edam for cheese, are commonly used to signal a certain quality of
City-of-Origin Effects in the German Beer Market
253
Geographic Focus International Continent Trade Zones
Examples NAFTA EU USA France Germany Rockies Champagne Bavaria New York City Paris Munich SoHo Montmartre Schwabing
Customer National International
National
Countries
States/ Sub-Regions Intranational Cities
Neighborhoods
Fig. 1.
Typology of Origin Effects.
products designed, manufactured or offered there. And even individual neighborhoods in bigger cities such as Montmartre in Paris, SoHo in New York City, or Schwabing in Munich are widely known for ne art products and thus are used to carry a distinct image which reects a certain level of product quality. From a managerial perspective it is also important to take the specic target group into consideration. Generally speaking, goods made or services offered at a certain location could be marketed to national or international customers. In the case, customers from abroad are the target group, the frame of reference from the customers perspective is quite different and typically reects less rsthand experience, higher psychic distance (Dow, 2000; Stottinger & Schlegelmilch, 2000) and a different set of affective re actions. When products are aimed at customers in the home country or a sub-region thereof, the origin effect is characterized by more rsthand experience, more or less personal involvement, psychic closeness, and emotional attachment. If we consider all the origin elds provided in Fig. 1, it is interesting to see that the country level is well researched. The extensive literature covers the investigation of customers reactions to country of origin in a national setting, quite typically the assessment of different imported products, as well as
254
cross-cultural comparisons. Some research deals with regional origin with respect to national target groups. But to the best of our knowledge, no work has been done on city and neighborhood origin effects, neither on national nor on international audiences. From a conceptual as well as a managerial perspective, these research gaps need more attention. It seems worthwhile to extend the current research on country of origin to geographic entities with a smaller focus like cities and neighborhoods. The core question when moving research practices, underlying concepts, and ndings from country-of-origin research to different geographic domains is, whether these approaches are suitable. Take for example, a positive productcountry image which enhances the perceived quality for a product originating from this country. Can we assume that the same mechanisms are at work when we focus on a region as carrier for product image? Would the antecedents of and consequences from region of origin be identical or not? From a conceptual point of view, origin mechanisms presumably work the same way irrespective of the geographic entity under investigation. More or less positive connotations related to its origin lead to more or less positive preferences for respective products or services. The core problem of shifting the focus in research from one geographic level to another lies in the fact that the major origin mechanisms are placed in very different sets of contingencies and antecedents. Consider, for example, the different environmental forces that determine country- or city-related origin effects. The institutional background, political and legal aspects, and different players that are at work to promote certain geographic units as well as economic relevance might be significantly different. With respect to antecedents of attitudes to and preferences for local products the psychological frame might also vary significantly. Chances are high that cognitive structures of customers (e.g. evoked sets, schemas), level and type of expertise, as well as affective patterns such as personal attachment to geographic entities, emotional and motivational drivers, and attitudinal reactions, are not the same. Thus the transferability of concepts from country-of-origin research to cityof-origin is in itself a valuable research question, which needs theoretical as well as methodological caution and rigor when being approached. In addition to conceptual and theoretical issues, using localization as a unique selling proposition seems to be a valid strategy especially for smalland medium-sized companies to ght global competition in their immediate home markets. The use of geographic units for specic propositions cannot be easily copied by competitors from outside the respective geographic unit and thus seems to be an attractive pathway to ght national or even international players. From a managerial perspective it pays off to know how
255
origin mechanisms work with respect to possible geographic entities, how they increase brand equity, and thus make possible premium prices for goods with a certain origin (Agrawal & Kamakura, 1999). If we are able to better understand how city-of-origin mechanisms work, local companies could develop stronger competitive advantages using more rened localized branding strategies. Following this approach might in a next step also lead to a distinct positioning outside the local market and could even carry the potential to use city of origin on an international scale when exporting products. In similar ways, international marketers might use city-of-origin insights to better assess the competitive environment in foreign markets, customer preferences related entry barriers into international markets as well as intranational differences with respect to the affective attachment of customers to local brands. Information of that kind is highly relevant for the development of promising entry strategies in foreign markets. Our research is a rst step in the investigation of city-of-origin effects which deals with two central questions. First, we want to establish the existence of origin effects on the city level. Without providing evidence for its existence, the remainder of our work would be highly questionable. Second, we investigate how a city-of-origin effect is able to inuence consumers buying behavior. In other words, assuming that an origin effect on a city level exists, we specifically analyze how and through which mechanisms it is able to affect consumers preferences with respect to buying products that originate from their own city of residence. To accomplish that, we rst review the country-of-origin literature in order to provide a theoretical basis. Next, we present results from qualitative research that identies relevant aspects of city-of-origin effects, with a focus on the German beer market. Both the results from the literature review and the qualitative eldwork are combined in a conceptual model which aims at explaining preferences for local products. This model as well as the measurement employed is empirically tested with data from residents in three large cities in the Ruhr area in the north-west of Germany, using both descriptive and conrmatory methods. Finally, we discuss our results and give implications for future research.
APPROACHES TO COUNTRY OF ORIGIN

Drawing from previous literature on consumers attitudes and preferences toward foreign products, two main research streams can be identied.
256
While one stream focuses on the impact of a products country of origin on making inferences about or evaluating foreign products, the other stream investigates factors that might inuence consumers attitudes toward foreign products as well as possible effects on repurchase intentions and corresponding behaviors. Next, we will briey address each of these two streams, analyze shortcomings, and discuss limitations in transferring these approaches to a more local level, i.e., when focusing no longer on the country, but on the region or city of origin. An extensive body of more than 600 studies exists that focuses on country-of-origin effects (Papadopoulos & Heslop, 2003). Within this research stream, it is of central interest whether, and, if so, how the products country of origin is taken into consideration for a products evaluation. From an information theory perspective, products are perceived as consisting of a large number of information cues, which can relate to both intrinsic and extrinsic cues. While intrinsic cues contain product characteristics such as taste or design, the extrinsic cues relate to a products brand, price, or the country of origin (Bilkey & Nes, 1982). However, the term country of origin has been found to be more complex and more ambiguous than initially assumed. Country of origin may consist of different components such as country of manufacture, country of assembly, or country of design (Han & Terpstra, 1988). In his work, Han (1989) developed and empirically investigated two different working mechanisms of country-of-origin effects, i.e., a halo effect and a summary effect. In the presence of a halo effect, customers use information about a products origin (and thus about the image of the corresponding country of origin) and transfer these to relevant product attributes. In other words, information about the image or similar characteristics of the products country of origin is used to infer about product attributes that are not directly available to the consumer. Consequently, the product-specic image of a country inuences product evaluation based on certain product attributes, which nally determines the attribution toward country-specic brands and products. This causal chain has been both theoretically and empirically examined by several studies (Erickson, Johansson, & Chao, 1984; Johansson, Douglas, & Nonaka, 1985; Johansson & Thorelli, 1985). When conceptualizing country-of-origin effects by a summary effect, on the other hand, consumers summarize their experiences with and perceptions of corresponding products from a particular country to form a country image, which is then used to facilitate product evaluation (Han, 1989). These experiences can also be based on word of mouth through other consumers as well as through mass media. In other words, the countrys
257
image inuences consumers attitude toward this particular product/brand (Crawford & Garland, 1988; Hong & Wyer, 1989; Howard, 1989) so that an evaluation of a product based on this summary effect will be preferred if the consumer has the strong wish to facilitate buying decision. Hence, consumers perceptions about country-specic product attributes directly inuence their attitude toward country-specic brands or products. In his empirical study, Han (1989) found that the mechanism underlying country-of-origin effects depends on the consumers product familiarity. When consumers possess only a low level of familiarity with the respective product, they use a products origin as a basis for their reasoning (i.e., halo effect), while consumers with a high level of product familiarity use information about a products origin in summary for the overall evaluation of a product. However, while several studies exist that found a positive correlation between product familiarity and usage of information about a products country of origin (Johansson, 1989; Johansson et al., 1985; Johansson & Nebenzahl, 1986), other studies report only very weak and marginally significant correlations (Heimbach, Johansson, & MacLachlan, 1989; Jaffe & Nebenzahl, 2001). Thus, Knight and Calanton (2000) recommend using a combination of both summary and halo effects in a exible model, which was found to be superior over either model used separately (Jaffe & Nebenzahl, 2001). However, and that is a key issue for all preceding approaches, if the country of a products origin is assumed to play a central role in the evaluation process, a consumer necessarily needs to possess sufcient knowledge about a products country of origin. In many cases, this cannot be assumed and may lead to incorrect assumptions about a products respective origin (Ettenson & Gaeth, 1991). Thus, several studies have concluded that information about a products country of origin was only used in the absence of other information on which to evaluate products (Nijssen & Douglas, 2004, p. 25). The second stream of research relevant for our investigation focuses on consumer attitudes toward foreign or imported products and brands. While most studies have used negative attitudes toward foreign products in general, a limited number of studies have also investigated attitudes toward specic countries (Balabantis, Diamantopoulos, Mueller, & Melewar, 2000; Javalgi, Khare, Gross, & Scherer, 2005; Nijssen & Douglas, 2004). Within this type of research, a widely studied construct focuses on consumer ethnocentrism, a concept which is closely related to LeVine and Campbells (1972) conceptualization of ethnocentrism, the latter dened as when the symbols and values of ones own ethnic or national group become objects of pride and attachment, whereas symbols of other groups may become objects
258
of contempt. Shimp and Sharma (1987) extended this concept to the domain of consumer behavior and conceptualized it as beliefs held by consumers about the appropriateness and/or morality of purchasing products made in or imported from a foreign country. Ethnocentric consumers share a common belief that buying products which have been manufactured in foreign countries is unpatriotic and wrong since it has the potential to harm ones own domestic economy and, consequently, might cause loss of jobs (Nijssen & Douglas, 2004). In addition to consumer ethnocentrism, research exists that specifically investigates the inuence of negative attitudes toward a particular country on consumers attitudes toward products from that country (Klein, Ettenson, & Morris, 1998), a stream of research which already begins to focus on an intra-domestic level (Hinck, 2004). However, based on the preceding discussion, research in both streams has so far been limited to larger geographical/political entities. Drawing from Papadopoulos (1993), further geographical units appear plausible and need to be taken into consideration, e.g., regions, cities, or even neighborhoods. These smaller units may lead to an increased level of information diagnosticity, since the geographical focus is much narrower than when focusing on countries. It is highly plausible to assume that in certain product categories (e.g., those with strong local and historical ties), the analysis of effects stemming from the fact that services or products have their origin near the consumers place of residence, can contribute to a better understanding of processes which lead to product preferences. Those aspects have not been investigated with respect to a products city of origin. Looking at the theoretical concepts that evolved in country-of-origin research, namely halo and summary effects in the product evaluation process and ethnocentrism/ animosity aspects in product evaluation, they seem suited to be used in analogy in research targeted to the inuence of smaller geographical/political units. Consequently, we take up both streams of research and include them in our conceptual model. First, as highlighted by previous studies (e.g., Jaffe & Nebenzahl, 2001; Knight & Calanton, 2000), we use different types of origin conceptualizations, i.e., halo and summary effects, as antecedents to attitudes and preferences toward local products. While attitude toward city of residence focuses on the summary effect component, knowledge about product origin represents the halo effect facet of product origin concepts. Both facets are expected to inuence customers attitude and corresponding preference toward local products. The second relevant stream of research discussed previously (i.e., consumers attitudes toward foreign products) is captured through the concept of customers attitude toward buying local.
259
Here, previous research has also stressed the importance of including consumers ethnocentric tendencies as well as their negative attitudes toward particular countries or regions in explaining consumers preferences toward specic types of products. Since we agree on the arguments brought forward regarding the importance of this facet, we capture relevant aspects through a conceptualization of consumers attitudes toward buying local in inuencing their attitude toward products from certain cities. However, because of the completely different setting, a simple transfer of existing theories and models will not necessarily be appropriate. We therefore undertake extensive exploratory research in order to assess the viability of using country-of-origin research insights in an intra-country context.
Qualitative Studies To the best of our knowledge, no studies exist that investigate the cityof-origin effect on consumers buying decision. In order to generate new insights and to assess the suitability of certain concepts from country-of-origin research for our eld of investigation, we embark on exploratory qualitative research. Herein, we concentrate on identifying attributes that are considered relevant for purchase decisions about beer in particular, but also more general, nonetheless inuential attributes about, e.g., the city of origin or consumer ethnocentrism. To accomplish that, we used a two-step procedure which consists of a repertory-grid test followed by two separate focus groups. The advantage of this two-step procedure is that we can include the results from the repertory-grid test within the discussion during our focus groups so that we are able to generate a deeper insight of the corresponding topics. The major aim of a repertory-grid test is to identify relevant, cognitive dimensions that are employed during the process of buying goods. The test procedure is based on an advancement of the Role Construct Repertory Test by Kelly (1955) and basically involves the task that consumers should distinguish certain elements, with the distinctions being collected in a data matrix (Bannister & Mair, 1968). For our test procedure, we carried out 21 personal interviews in a supermarket. We focused on consumers who had just purchased beer, so they did not have to report on an imaginary or previous buying decision. The test used as reference products 10 German beer brands that can be easily distinguished through origin, market position, and price. More than 90% of the consumers identied good taste as the most important product characteristic, followed by local origin (87%) and
260
Table 1.
Construct Good taste Local origin Brown bottle Appealing label design Bottle design High brand awareness Bottle cap Good reputation Reasonable price
Most Important Product Cues.

Percentage of Choice 90.5 86.7 76.2 76.2 76.2 71.4 57.1 47.6 38.1 Sum of Ratings 106 79 52 47 39 66 33 40 39
No. of Choice 19 18 16 16 16 15 12 10 8
brand awareness (71%). Only less than half of all consumers think that a reasonable price and a good brand image represent salient characteristics (see Table 1). These results already show that the beers origin is an important factor in a customers buying decision. Next, we use these very preliminary insights as key discussion points in our focus groups. Focus groups are able to identify relevant topics and aspects of buying decisions related to beer and thus they can complement to the results from both the literature review as well as the repertory-grid test. A focus group consists of interviews that are carried out by a moderator with a small group of interviewees with largely different backgrounds (Morgan, 1998). In contrast to single interviews, focus groups embody the advantages that spontaneous reactions can be triggered more easily and that salient characteristics and opinions can be made transparent. This allows for an investigation of different dynamics within the ongoing discussion, such as a change of opinion or rather weak judgments within the discussion (Bellenger, Bernhardt, & Goldstucker, 1976). Through this exploratory character, focus groups are well suitable as pre-analytical techniques for subsequent quantitative studies, since they allow for generation of a rather large number of relevant attributes/factors (Gordon & Langmaid, 1988). We carried out two separate and independent focus groups to identify further information, opinions, and motives as well as attitudes toward the process of buying beer. Therefore, we invited 16 beer-consuming men, aged 2560 and living in the Ruhr-Rhine area, to participate in one of the discussion groups. With respect to demographic criteria like, for example, marital status, income, level of education, and professional position, we generated a mixture so as to achieve a focus group as heterogeneous as possible. Also, the consumers did not know each other in advance, a characteristic that facilitates and intensies an ongoing discussion within the
261
group (Morgan, 1998). Since the aim of this focus group is to identify relevant attributes and characteristics that are related to the decision of buying beer, we additionally developed a protocol, that is, an ordered set of questions already formulated in complete sentences. This protocol helps us to compare the outcome of both focus groups and thus facilitates the later analysis. Since both focus groups have been audio taped and thus could be transcribed completely, we used content analysis to identify relevant topics, arguments, and typical expressions of the participants (Krueger, 1998). Results from both focus groups reveal that the beers taste is a dominant criterion related to buying beer, that is, the different categories of bitter and mild beers. Additionally, participants identied the quality of different beers as a salient criterion, although they had major problems in objectively rating a beers quality. However, the most important criterion in both focus groups has been identied as the beers origin. All participants agreed that the region or city in which a beer is brewed has major implications on both taste and quality and thus is used as a major determinant of a customers preference for different brands of beer. Importantly, all participants agreed that a focus on the region or city of origin is necessary since the country level is much too general when it comes to the assessment of a beers quality with respect to its origin. Further, most of the participants revealed a high level of local orientation and patriotism, that is, they buy products, if possible, from their own city of residence in order to support local industries. They demonstrated pride in living in that particular region or city, and thus pride in their local products, which in last consequence leads to a positive attitude toward these local products. Only if no local brands exist, most participants report to buy brands from preferably rural areas. This additionally supports our arguments that the beers city of origin is an important factor in a consumers buying decision. Also, participants revealed a strong relation between their attitude toward local products and their corresponding attitude toward their city of residence. There seems to exist an effect that carries over both positive and negative attitudes which exist toward the city on products having their origin in this particular city. Significantly, in both focus groups, participants revealed difculties in naming different beers cities or regions of origin as long as they originate from cities or regions different from the participants residence. Interestingly, those participants who showed a high level of knowledge about different brands regions or cities of origin place less importance on a beers origin for their buying decision. In other words, the more our participants know about the origin of different products within different categories, the less important this knowledge becomes for their decision process.
262
Altogether, we obtained some interesting insights into consumers buying decisions related to local products. Overall, we nd that the same mechanisms seem to be present as with country-of-origin effects, but both the concepts as well as the operationalizations have to be adopted when focusing on city-of-origin effects. Next, we combine the results from both the literature review as well as from our qualitative studies so as to develop research propositions and a conceptual model.
DEVELOPMENT OF A CONCEPTUAL MODEL

On the basis of country-of-origin research as well as our qualitative studies, a theoretical model to investigate and explain preferences for products produced in the city of residence is developed and shown in Fig. 2. Preference for local products is conceptualized as a construct with behavioral character. It is dened as local products rank order of preference according to the perceived attractiveness for the buyer. A key construct in explaining preference for local products, which played a major role in the preliminary research undertaken for this project, is labeled attitude toward local products. This construct is dened as subjective perceptions of local brewery products and captures beers quality, taste, and freshness as well as intentions to offer this product to friends and family (e.g., Schaefer, 1997). Another concept, which came up in both previous studies of countryof-origin effects (Forgas & ODriscoll, 1984; Or & Lehmann, 1986; Papadopoulos, Heslop, & Bamossy, 1990; Parameswaran & Pisharodi, 1994; Pisharodi & Parameswaran, 1992) as well as our qualitative research,
Attitude towards city of residence
+ + Attitude towards local products Preference for local products
Attitude towards Buy Local Knowledge of product origin
+ +
Fig. 2. Conceptual Model.
263
is related to the attitude toward the city of residence. This concept captures subjective opinions and views about the city of residence, and depicts several characteristics of this city, e.g., friendliness, cleanliness, modernity, and trustworthiness. Attitude toward Buy Local is very closely related to consumer ethnocentric tendencies and thus conceptualized as a representation of consumers beliefs and attitudes about the appropriateness of purchasing nonlocal products (Nijssen & Douglas, 2004; Shimp & Sharma, 1987). A nal concept identied in previous investigations as well as in our qualitative study, is knowledge of product origin. Consumer knowledge about the origin of a product should be regarded as a multidimensional construct, i.e., it should capture both objective and subjective components (Schaefer, 1997). Based on country-of-origin research with a focus on the level of product knowledge (Darling, 1987; Pecotich, Pressley, & Roth, 1996; Schaefer, 1997) as well as halo effects (Erickson et al., 1984; Han, 1989; Johansson et al., 1985; Johansson & Thorelli, 1985), consumer ethnocentric tendencies (Klein et al., 1998; Netemeyer, Durvasula, & Lichtenstein, 1991; Pecotich et al., 1996; Shimp & Sharma, 1987), general attitudinal research (Allen, Machleit, & Schultz Kleine, 1992), and results from our qualitative work, we formulate the following hypotheses. H1. The level of knowledge about goods local origin in a product category has a negative impact on the preference of products made in the consumers city of residence. H2. A more positive attitude toward Buy Local leads to a more positive attitude toward products produced in the city of residence. H3. The more positive the attitudes toward the city of residence, the higher the attitude toward local products. H4. The more positive the attitude toward local products the higher the preference for (the willingness to buy) products with local origin.
Methodology To test our hypotheses and the resulting research model, we collected data from 316 participants in three major cities in the Ruhr-Rhine area, using a standardized questionnaire in combination with personal interviews. 104 persons participated in supermarket intercept interviews in Dortmund (a city of 590,000 inhabitants), 106 in neighboring Bochum (380,000 inhabitants), and 106 in Duisburg (510,000 inhabitants), located at a
264
distance of 40 miles from Dortmund. Altogether, the participants reected a variety of different demographical types, with no significant differences in age, education, and income levels found between the three cities. In analyzing these data, we employed both conjoint analysis and structural equation modeling. Conjoint measurement helps us determining the relative importance of a products local origin, while structural equation modeling identies a causal structure between attitudes toward local products, buying local, and preference for local products, among others. The conjoint measurement design consists of four different factors, i.e., city of origin, price, taste, and brand name. Cities of origin included in the design represent the same three major cities in which the data has been collected, so that biases are unlikely. In order to infer about the importance, we subsequently analyze the conjoint data both for the pooled sample as well as separately within each of the three cities. Price also consists of three different facets, i.e., low, medium, and high-priced beer, with actual memory values of 6h, 8h, and 10h, respectively. The price levels have been chosen in alignment with the standard prices on the German beer market. Taste is captured by two major facets that relate to the taste of beer, i.e., mild- and bitter-tasting beer. Finally, brand name represents three different brands, i.e., Harpoon, Lakefront, and Acadian. It is a highly controversial issue whether to choose brand names that are mostly known or mostly unknown to the respondents. Including familiar and well-known brands could lead to overemphasizing the importance of a brand name factor, since many respondents possibly choose the brand that they like or that they are mostly familiar with. Also, since marketing communication related to specic brands largely differs in both quality and quantity, most familiar brands earn their reputation through advertisements and related marketing communications, which diminishes the importance of other characteristics that might be of relevance. However, choosing brand names that are mostly unknown to study participants likely weakens the effect that a particular brand might have. Weighing these arguments, we decided to include brand names that are mostly unknown to participants, since we are also very interested in the remaining three factors of our study, and we fear that wellknown and familiar brand names could unjustiably capture participants attention. The nal brand names chosen represent existing American beer brands, which however should be mostly unknown to German consumers. The second methodological section focuses on a structural equation model in order to test the set of hypotheses. In addition, this part also addresses issues of measurement reliability and validity as well as goodnessof-t tests for the overall structural model. We operationalized our measures
265
based on a combination of existing scales adapted from the literature, and a few additional items created through exploratory qualitative research. In detail, we employed 11 items from, among others, Papadopoulos et al. (1990) to capture Attitude toward City of Residence, 8 items to measure Attitude toward Buy Local (Shimp & Sharma, 1987), and 6 items to operationalize Attitude toward local products (Tse & Gorn, 1993; Tse & Lee, 1993). The level of knowledge of product origin is operationalized through two different facets, namely, objective and subjective knowledge. While the subjective knowledge component was derived from our qualitative research and consists of three items, the objective knowledge is measured by an indicator which captures the number of different beers cities and regions of origin participants were able to identify. Consequently, knowledge of product origin consists of four items, one measuring the objective component, the remaining three focusing on subjective knowledge. Finally, preference for local products is operationalized through three items, with a diverse content. While two items capture the rank order from experiments, where participants were asked to rank different brands of beer as well as different cities as origins of local beer, the third item shows estimated utilities of local brands, as reected by results taken from a conjoint analysis. An overview of the items employed in this study can be found in the appendix.
RESULTS
Before testing our hypotheses, we rst focus on descriptive results obtained from conjoint analyses for the pooled sample. These results show that the city of a products origin evidences a strong importance for a consumers buying decision. In detail, a beers city of origin yields an average relative importance of 32.69%, when making a buying decision. This is even more than is attributed to the price of beer, with a smaller, but still high relative importance of 29.45%. The taste of beer (i.e., either mild or bitter tasting) shows a relatively weak importance for a consumers buying decision, with a value of 21.11%. Not surprisingly, brand names do not show any strong importance (16.75%), since we used brands that are most likely unfamiliar to study participants. Overall, this underlines the attention that should be paid to communicating a products city of origin, specifically when focusing on different brands of beer. This facet is further emphasized when concentrating on the three sub-samples, i.e., on the samples collected in the three different cities. Here we also nd that, as expected, the corresponding city of origin shows the largest relative importance for making a buying decision, with values ranging
266
Table 2.
Sample
Results from Conjoint Analysis.

Dortmund Imp./Util. 18.25% 0.136 0.068 0.068 34.55% 1.243 0.919 0.324 27.43% 0.194 0.236 0.042 19.77% 0.250 0.250 Duisburg Imp./Util. 18.04% 0.016 0.044 0.028 30.96% 0.333 1.072 0.739 29.75% 0.516 0.308 0.208 21.25% 0.446 0.446 Bochum Imp./Util. 13.99% 0.123 0.031 0.154 32.63% 0.513 0.547 1.060 31.11% 0.255 0.321 0.066 22.28% 0.469 0.469
Pooled Imp./Util.
Brand name Harpoon Lakefront Acadian City of origin Dortmund Duisburg Bochum Price 6,- h per box 8,- h per box 10,- h per box Taste Bitter tasting Mild tasting
16.75% 0.002 0.018 0.020 32.69% 0.122 0.124 0.002 29.45% 0.323 0.289 0.034 21.11% 0.390 0.390
from 30.96 to 34.55%. In addition, we nd strong positive utilities for the corresponding cities in its respective sub-sample. For example, in the Dortmund sample we nd a strong positive utility for the facet brewed in Dortmund (1.24), while the other two facets (i.e., brewed in Bochum and brewed in Duisburg) show large negative values of .92 and .32, respectively. Details about results from the conjoint study can be found in Table 2. Following these more descriptive insights, we conducted analyses into the validity and reliability of our measures. First, in an exploratory factor analysis, we nd one factor extracted for each of the ve constructs, based on both eigenvalue criterion and scree plot (Hair, Anderson, Tatham, & Black, 1998), with extracted variances between 58.1 and 69.4%. Furthermore, all items without exception exhibit relatively large loadings (>0.5) on this factor, yielding support for convergent validity (Morrison, 1994). Additionally, Cronbach reliabilities are estimated between 0.63 and 0.94, hence yielding additional support for the measurement properties of our scale (De Vellis, 1991; Nunnally, 1978). Second, in a conrmatory factor analysis, the overall measurement model is clearly supported. In detail, we nd that all item loadings are significantly different from zero (po0.001), with factor reliabilities larger than 0.68, and average variances extracted exceeding highest variance shared, indicating
267
discriminant validity (Fornell & Larcker, 1981). Furthermore, model t indices are also indicative of good model t (w2 531.58 (d.f. 287), NFI 0.881, NNFI 0.933, CFI 0.941, RMSEA 0.049, sRMR 0.069), yielding strong support for our proposed measurement. Owing to the large number of items per construct and the relatively complex structure of our model, we use a partially disaggregated model for further analyses, i.e., we aggregate our original set of items so that nally we employ three aggregated items for each construct. Based on Bagozzi and Heatherton, this is a legitimate procedure once acceptable measurement properties of the completely disaggregated model have been conrmed (Bagozzi & Heatherton, 1994). We therefore aggregate our set of items and re-analyze the aggregated data, again using conrmatory factor analysis. Similarly, we yield highly acceptable estimates for indicator loadings (all po0.001), factor reliabilities (>0.68), and variances extracted (>0.48), with all indices supporting high quality measurement (w2 146.40 (d.f. 80), NFI 0.938, NNFI 0.961, CFI 0.970, RMSEA 0.052, sRMR 0.049). Next, after conrming measurement properties of the scales employed in our study, we investigate the structural part of the model, and thus focus particularly on the hypotheses developed above. The results from a maximum likelihood estimation using LISREL 8.54 can be found in Table 3. Table 3.
Dependent Variable
Results of Structural Equation Model.

R2 b(e) Direct Effects T R2 Indirect Effects b(e) T
Independent Variable
Attitude toward local products
0.07 Attitude toward buy local Attitude toward city of residence 0.52 Knowledge of product origin Attitude toward local products Attitude toward buy local Attitude toward city of residence 0.13(0.06) 0.81(07) 2.25 11.08 0.09(.04) 0.16(.06) 2.18 2.63 0.12(0.05) 0.20(0.07) 2.20 2.66
Preference for local products
Overall Model Fit Indices: w2 156.6(po0.001), NFI 0.933, NNFI 0.958, CFI 0.967, 83 sRMR 0.056, RMSEA 0.053
268
Generally speaking, our hypotheses are strongly conrmed, as they are all supported by the data. In detail, we nd that both attitude toward buying local products and attitude toward the city of residence exert significant and positive effects on a consumers attitude toward local products, with estimated standardized path coefcients of 0.12 (po0.05) and 0.20 (po0.01), respectively. Furthermore, while customers level of knowledge about local product origin, as expected, exerts a negative effect on customers preference for products with local origin (b 0.13, po0.05), attitude toward local products dominantly and positively affects customers preference for locally produced products, with a large path coefcient of 0.81 (po0.001).
DISCUSSION AND CONCLUSION

The present paper for the rst time empirically focuses on city-of-origin mechanisms. Our qualitative as well as our quantitative results clearly show that such an effect exists, and we therefore advocate putting more emphasis on research with regard to origin effects of different geographic and political entities. Following up on the results obtained there is strong evidence that cityof-origin effect mechanisms are very similar to the well-researched countryof-origin effects. Building on research by Knight and Calantone (2000) it is safe to conclude that both halo effects and summary effects shape city-oforigin effects at the same time. We also could show that consumers attitude toward buying local functions in a similar way as attitudes toward foreign products on the national level (e.g., Nijssen & Douglas, 2004). We caution against overgeneralization, but our results provide no evidence that other theoretical concepts might be necessary when trying to capture the mechanisms of origin effects with respect to different entities or loci of origin. The managerial implications of our research are twofold. For the management of local products, the city-of-origin effect provides a basis to stress the local attachment and thus either ght competitors from outside, or to develop a unique selling proposition based on the local competence in a larger market. In more general terms, focusing on differentiated marketing activities (e.g., stressing the local origin in the city and at the same time playing it down in the outside market), the trade-off between higher costs on the one hand and potential benets on the other has to be considered. From an international marketers perspective, knowledge of city-of-origin effects may be used to assess customer-related affective mechanisms in export or international markets which lead to more or less harsh competitive
269
environments. The higher the city-of-origin effects in particular industries the higher the entrance barriers in those markets and hence more localized entry strategies like acquisition of local rms and brands, localized advertising messages, and close co-operation with local distribution channels are effective for increasing market penetration. Considering city-of-origin effects may also help to understand why some products or service providers are more successful in certain parts of a country, but not in others. Our results are exploratory and probably raise more questions than they answer, but they provide a strong proof of the basic effects. The core limitations lie in the specic product category used, the eld work in only one region in one country, as well as in the transfer of theoretical and conceptual ideas albeit done in a very cautious way from quite a different level of abstraction. From a research perspective, important questions are whether in addition to the limited conceptualization we used, a completely different set of antecedents to the preference of local products exists. Consider, for example, intra-cultural differences within a country, more or less homogeneous national institutional backgrounds, regional rivalry, or interregional differences in economic and welfare development as sources for like and dislike of products within or outside a city. Closely related to that question is the generalizability of the city-of-origin effect across industries or product categories. First of all, it seems likely that product categories differ with respect to the importance of the effect. Certain products are by themselves more origin bound than others, like for example food and beverage products or, in contrast, cosmetic products. Another aspect concerns the type of customer involved. Is it justied to assume that we see very homogenous reactions with respect to the city of origin, or is there a considerable difference between segments in the market? Finally, does the effect hold across countries and cultures? Are there societal differences (e.g., the higher mobility of citizens in the US as opposed to Europe) that lead to less city-of-residence identication and pride and thus to less propensity to prefer products from the near surroundings? Since our research is highly exploratory, it obviously provides several avenues for future research. Specifically, since we exclusively focused on cities as geographic entities, it seems worthwhile to expand our ideas to other entities as well, for example, to trade zones on the one end of the spectrum or neighborhoods on the other. Although we do not expect largely different results for these different entities, they need to be investigated before conclusions regarding the generalization of our ndings can be drawn. In general, further research on the effect of products or services
270
origin from different geographic entities could serve as a basis to better understand the interplay between place of residence on the one hand and buying and consumption behaviors on the other, and thus could lead to more differentiated marketing activities with respect to the place of origin in national as well as international marketing contexts. Additionally, since we limited our investigation to one industry and one region, both an analysis of intra-national and international differences with respect to city-of-origin effects across industries should be pursued.
REFERENCES
Agrawal, J., & Kamakura, W. A. (1999). Country of origin: A competitive advantage? International Journal of Research in Marketing, 16, 255267. Allen, C. T., Machleit, K. A., & Schultz Kleine, S. (1992). A comparison of attitudes and emotions as predictors of behavior at diverse levels of behavioral experience. Journal of Consumer Research, 18, 493504. Bagozzi, R. P., & Heatherton, T. F. (1994). A general approach to representing multifaceted personality constructs: Application to state self-esteem. Structural Equation Modeling, 1, 3567. Balabantis, G., Diamantopoulos, A., Mueller, R., & Melewar, T. C. (2000). The impact of nationalism, patriotism and internationalism on consumer ethnocentric tendencies. Journal of International Business Studies, 32, 157175. Bannister, D., & Mair, J. (1968). The evaluation of personal constructs. London: Academic Press. Bellenger, D. N., Bernhardt, K. L., & Goldstucker, J. L. (1976). Qualitative research in marketing. Chicago: American Marketing Association. Bilkey, W. J., & Nes, E. (1982). Country-of-origin effects on product evaluations. Journal of International Business Studies, 13, 8999. Crawford, J. C., & Garland, B. C. (1988). German and American perceptions of product quality. Journal of Consumer Marketing, 1, 6378. Darling, J. R. (1987). A longitudinal analysis of the competitive prole of products and associated marketing practices of selected European and non-European countries. European Journal of Marketing, 21, 1729. De Vellis, R. F. (1991). Scale development: Theory and application (Vol. 26). Newbury Park: Sage. Dow, D. (2000). A note on psychological distance and export market selection. Journal of International Marketing, 8, 5164. Erickson, G. M., Johansson, J. K., & Chao, P. (1984). Image variables in multi-attribute product evaluations: Country-of-origin effects. Journal of Consumer Research, 11, 694699. Ettenson, R., & Gaeth, G. (1991). Consumer perceptions of hybrid (bi-national) products. Journal of Consumer Marketing, 8, 1318. Forgas, J. P., & ODriscoll, M. (1984). Cross-cultural and demographic differences in the perception of nations. Journal of Cross-Cultural Psychology, 15, 199222. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 3950. Gordon, W., & Langmaid, R. (1988). Qualitative market research. A practitioners and buyers guide. Aldershot: Gower.
271
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). New Jersey: Prentice-Hall. Han, C. M. (1989). Country image: Halo or summary construct? Journal of Marketing Research, 26, 222229. Han, C. M., & Terpstra, V. (1988). Country-of-origin effects for uni-national and bi-national products. Journal of International Business Studies, 19, 235255. Heimbach, A. E., Johansson, J. K., & MacLachlan, D. L. (1989). Product familiarity, information processing, and country-of-origin cues. Advances in Consumer Research, 16, 460467. Heslop, L. A., & Papadopoulos, N. (1993). But who knows where or when: Reections on the images of countries and their products. In: N. Papadopoulos & L. A. Heslop (Eds), Product-country images: Impact and role in international marketing (pp. 3975). Binghamton: International Business Press. Hinck, W. (2004). The role of domestic animosity in consumer choice: Empirical evidence from Germany. Journal of Euromarketing, 14, 87104. Hong, S.-T., & Wyer, R. S., Jr. (1989). Effects of country-of-origin and product-attribute information on product evaluation: An information processing perspective. Journal of Consumer Research, 16, 175187. Howard, D. G. (1989). Understanding how American consumers formulate their attitudes about foreign products. Journal of International Consumer Marketing, 2, 722. Jaffe, E. D., & Nebenzahl, I. D. (2001). National image and competitive advantage. Copenhagen: Copenhagen Business School Press. Javalgi, R. G., Khare, V. P., Gross, A. C., & Scherer, R. F. (2005). An application of the consumer ethnocentrism model to French consumers. International Business Review, 14, 325344. Johansson, J. K. (1989). Determinants and effects of the use of made in labels. International Marketing Review, 6, 4758. Johansson, J. K., Douglas, S. P., & Nonaka, I. (1985). Assessing the impact of countryof-origin on product evaluations: A new methodological perspective. Journal of Marketing Research, 22, 388396. Johansson, J. K., & Nebenzahl, I. D. (1986). Multinational production: Effect on brand value. Journal of International Business Studies, 17, 101126. Johansson, J. K., & Thorelli, H. B. (1985). International product positioning. Journal of International Business Studies, 16, 5775. Kelly, G. A. (1955). The psychology of personal constructs. London: Norton. Klein, J. G., Ettenson, R., & Morris, M. D. (1998). The animosity model of foreign product purchase: An empirical test in Peoples Republic of China. Journal of Marketing, 62, 89100. Knight, G. A., & Calantone, R. J. (2000). A exible model of consumer countryof-origin perceptions. A cross-cultural investigation. International Marketing Review, 17, 127145. Krueger, R. A. (1998). Analyzing and reporting focus group results (Vol. 6). Thousand Oaks, London, New Delhi: Sage. LeVine, R. A., & Campbell, D. T. (1972). Ethnocentrism: Theories of conict, ethnic attitudes and group behavior. New York: Wiley. Morgan, D. L. (1998). The focus group guidebook (Vol. 1). Thousand Oaks, London, New Delhi: Sage.
272
Morrison, E. W. (1994). Role definitions and organizational citizenship behavior: The importance of the employees perspective. Academy of Management Journal, 37, 15431567. Muller, S., & Kesselmann, P. (1996). Buy regional: Der Stellenwert des made in Sachsen fur die Kaufentscheidung ostdeutscher Konsumenten. Die Betriebswirtschaft, 363377. Netemeyer, R. G., Durvasula, S., & Lichtenstein, D. R. (1991). A cross-national assessment of the reliability and validity of the CETSCALE. Journal of Marketing Research, 28, 320327. Nijssen, E. J., & Douglas, S. P. (2004). Examining the animosity model in a country with a high level of foreign trade. International Journal of Research in Marketing, 21, 2338. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. Or, C., & Lehmann, D. R. (1986). Measuring images of foreign products. Columbia Journal of World Business, 21, 105108. Papadopoulos, N., & Heslop, L. A. (2003). Country equity and product-country images: State of the art in research and implications. In: S. C. Jain (Ed.), Handbook of research in international marketing (pp. 402433). New York, NY: Edward Elgar Publishing. Papadopoulos, N., & Heslop, L. A. (1993). Product country images: Impact and role in international marketing. New York: International Business Press. Papadopoulos, N., Heslop, L. A., & Bamossy, G. (1990). A comparitive image analysis of domestic versus imported products. International Journal of Research in Marketing, 7, 283294. Parameswaran, R., & Pisharodi, R. M. (1994). Facets of country of origin image: An empirical assessment. Journal of Advertising, 23, 4456. Pecotich, A., Pressley, M., & Roth, D. (1996). The impact of country of origin in the retail service context. Journal of Retailing and Consumer Services, 3, 213224. Pisharodi, R. M., & Parameswaran, R. (1992). Conrmatory factor analysis of a countryof-origin scale: Initial results. In: J. Sherry & B. Sternthal (Eds), Advances in consumer research (19 ed., pp. 706714). Provo: Association for Consumer Research. Schaefer, A. (1997). Consumer knowledge and country of origin effects. European Journal of Marketing, 31, 5672. Shimp, T. A., & Sharma, S. (1987). Consumer ethnocentrism: Construction and validation of the CETSCALE. Journal of Marketing Research, 24, 280289. Stottinger, B., & Schlegelmilch, B. B. (2000). Psychic distance: A concept past its due date? International Marketing Review, 17, 169173. Trognon, L., Lagrange, L., & Janin, S. (1999). Consumer attitudes towards regional food products: A case study for Auvergne. The European Food Consumer, 5, 3039. Tse, D. K., & Gorn, G. J. (1993). An experiment on the salience of country-of-origin in the era of global brands. Journal of International Marketing, 1, 5776. Tse, D. K., & Lee, W. (1993). Removing negative country images: Effect of decomposition, branding, and product experience. Journal of International Marketing, 1, 2548. van Ittersum, K. (2002). The role of region of origin in consumer decision-making and choice. Wageningen, NL: Wageningen University. van Ittersum, K., Candel, M. J. J. M., & Meulenberg, M. T. G. (2003). The inuence of the image of a products region of origin on product evaluation. Journal of Business Research, 56, 215226. Verlegh, P. W. J., Steenkamp, J.-B. E. M., & Meulenberg, M. T. G. (2005). Country-of-origin effects in consumer processing of advertising claims. International Journal of Research in Marketing, 22, 127139.
273
APPENDIX: OPERATIONAL MEASURES OF STUDY CONSTRUCTS

Attitude toward city of residence 6-point scale, Strongly Agree/Strongly Disagree, a 0.74, AVE 0.48, HVS 0.27 oCity> is y y sympathetic y friendly y pleasant y peaceful y attractive y clean y trustworthy y competent y modern y successful y close to my heart Attitude toward buying local 6-point scale, Strongly Agree/Strongly Disagree, a 0.94, AVE 0.83, HVS 0.32 Inhabitants of oCity> should always, when possible, buy products from oCity> instead of buying products from different cities Buy products from oCity>, safeguard our jobs It is not right buying products from different cities whenever it is possible to buy products from oCity>, since people from oCity> could become unemployed A real inhabitant of oCity> should always, whenever possible, buy products produced in oCity> We should only buy those products from different cities which are not produced in oCity> Whenever possible, I always buy products from oCity> in order to support the local economy I would always prefer food from oCity> to others I am proud of our local products Knowledge of product origin 6-point scale, Strongly Agree/Strongly Disagree, a 0.75, AVE 0.51, HVS 0.30
274
I know which beers are produced in oCity> I know in which cities most beers from my region are brewed For most German beer, I know where they are brewed Knowledge index (Number of correctly specied regions and cities of origin for different beers.) Attitude toward local products 6-point scale, Strongly Agree/Strongly Disagree, a 0.89, AVE 0.74, HVS 0.42 Beer from oCity> is sympathetic I enjoy offering beer from oCity> to friends Beer from oCity> is of highest quality Beer from oCity> is guaranteed fresh Beer from oCity> is enjoyable Beer from oCity> tastes very good Preference for local products a 0.68, AVE 0.44, HVS 0.42 Rank order from experiment 1 Rank order from experiment 2 Relative importance of local origin for buying beer taken from conjoint analysis
HOW DO BUSINESS GROUPS FUNCTION AND EVOLVE IN EMERGING MARKETS? THE CASE OF TURKISH BUSINESS GROUPS
Attila Yaprak, Bahattin Karademir and Richard N. Osborn
ABSTRACT
Business groups have become a significant phenomenon in the evolution and functioning of emerging markets. They also provide important partnership opportunities to foreign rms when they enter these markets. Yet, business groups have not received sufcient attention in the international marketing literature. In this paper, we provide an overview of the theories that explain how business groups function and evolve in emerging markets and generate propositions from that theory. We also present evidence on business group evolution from one emerging market, Turkey. Our work should inspire research questions for future study.
275
276
ATTILA YAPRAK ET AL.
INTRODUCTION
In their seminal review of marketing in the New Millennium Issue of the Journal of Marketing, Day and Montgomery (1999) projected that the functioning and evolution of markets would be a significant issue that would require special attention from marketing scholars in the rst decades of the 21st century. They also identied rms relations to their markets as another significant issue, and projected that rms would increasingly relate to their markets through collaborative arrangements and partnerships to improve their performance, especially in those markets that are less familiar to them. In the short span since that review, the evolution of the context in which rms evolve and strive for high performance has become even more significant with the accelerated pace of technological change and globalization. Nowhere is this evolution more pronounced than in the worlds rapidly transforming economies, such as China, India, and Brazil, where rms are evolving with fascinating patterns of development. A key ingredient in this evolution has been the development and success of business groups, often conglomerated family rms, which have been able to develop capabilities that are valuable, rare, and inimitable in these contexts, such as rapid diversication into unrelated product lines and repeated industry entry (Guillen, 2000). This fascinating pattern of evolution has generated considerable interest in the management literature recently; indeed, the Academy of Management Journal dedicated an entire issue to the topic in 2000 and has dedicated considerable publication space to studies examining this issue since then (see, e.g., Chang, 2003). Yet, scholars who study the international dimensions of business have not yet focused sufcient attention on the emergence and functioning of business groups in emerging markets. In fact, this ought to be a significant question for international marketing scholars to examine for a number of reasons. First, since business groups are typically significant players in the economic functioning of emerging markets (Khanna & Yafeh, 2005), their study could provide valuable insights about the nature and functioning of the markets international marketers will penetrate, operate in, and expand from. Second, studies of these groups could unearth important insights about the cultural norms governing exchanges, marketing infrastructures, distribution systems, and regulatory mechanisms in those markets. Third, such studies could provide a better understanding of the partnering behavior by indigenous rms in those markets, whether with prospective foreign partners who are seeking market access or knowledge,
How do Business Groups Function and Evolve in Emerging Markets?
277
or with other networks in their markets who might be pursuing economic gains (Khanna & Palepu, 2000; Guillen, 2000; Kock & Guillen, 2001). Yet, we know very little about the behavior of these groups in emerging economies (Khanna & Yafeh, 2005), such as China and India, even as these economies have been growing as important consumer markets and international expansion platforms recently (Prahalad & Lieberthal, 1998; Prahalad, 2005). We hope to contribute to a better understanding of business groups with our current paper. In response to Day and Montgomerys (1999) call, we hope to examine why business groups develop and prosper in emerging economies, how they might help prospective multinational businesses enter and expand from an emerging economy, and how they might help emerging markets develop and prosper economically. We believe our work is significant on, at least, two grounds. First, we bring insights from the literature on business groups in organizational theory to understand their role more fully in an international business context. Second, we present a discussion of business groups in Turkey as an example of the types of groups that multinational rms might want to embed themselves into in other emerging economies as they operate in these markets and evaluate foreign market expansion opportunities from them. Finally, we offer questions for future research. Our paper is organized as follows. We present the relevant theory of business groups and derive propositions about the evolution and role of business groups from that theory. We then present a brief discussion of the evolution and functioning of business groups in Turkey to underscore the significance of our propositions. We then offer questions that future researchers might address to enhance knowledge about business groups.
RELEVANT THEORY AND PROPOSITIONS

A growing literature on the evolution and role of business groups in emerging economies has surfaced recently (see, e.g., the Academy of Management Journal, 2000, Special Issue). This literature has dened business groups as collections of rms pulled and bound together in formal and informal ways in an intermediate sense (i.e., excluding strategic alliances and rms that are legally consolidated into a single entity) to achieve an economic purpose (Granovetter, 1994), and as coalitions of organizations that go beyond those created by temporary alliances among two or more otherwise independent rms that do not constitute a legally consolidated or integrated company
278
(Guillen, 1997). Business groups are said to be more diversied and generate greater economic impact in their respective markets and invest more heavily and in greater numbers of industries when compared to their independent counterparts in their markets (Kim, Kandemir, & Cavusgil, 2004). Granovetter (1994) suggests that the existence and development of business groups should be viewed through essentially four separate lenses: (1) resource dependence (rms routinely depend on other rms for resources they do not possess); (2) need for strategic partnerships (it is in the nature of markets and consumer demand to change, and the pace of change has accelerated recently due to globalization and the technological revolution); (3) desire to extract rents from the economy or the government through coalitions over and above those which can be gotten in a properly competitive economy; and (4) need to form coalitions of capitalists against social interests (the Marxist explanation). That is, transaction costs theory, institutional theory, and political economy theory should all play a part in explaining the existence of business groups in emerging markets (Guillen, 2000; Khanna & Palepu, 2000). Economists explain the existence of business groups as microeconomic responses to conditions of market failure, especially imperfect markets in capital and intermediate products that plague emerging markets (Granovetter, 1994). Proponents of this position see business groups as responses to transaction costs primarily in capital markets (Williamson, 1991), but in markets for entrepreneurial talent, product markets, labor markets, and cross-border markets for technology as well, which lead to preferential access, for instance, to capital (Khanna & Palepu, 2000; Chang & Choi, 1988), and as responses to agency problems in the market, whereby rms are able to secure intermediate goods with lower costs and less uncertainty by joining or forming groups rather than procuring them through the market or integrating vertically (Granovetter, 1994). Business groups exist in this view only in the absence of a well-functioning market, i.e., they enhance value by serving as functional substitutes for allocation failures in the markets for production inputs (Khanna & Palepu, 2000). There are many market failures in emerging markets such as India, in sharp contrast to advanced economy markets, such as the United States. Khanna and Palepu (2000) argue that these imperfections typically include an inadequate marketing infrastructure, lack of regulatory discipline or frequent changes in the regulatory environment, political and economic instability, and poor dissemination of market information. In markets like the United States, where there is relatively efcient intermediation in markets, entrepreneurs are not as likely to seek association with rms that are
279
diversied across unrelated industries, while in an emerging market such as India, where market intermediation is relatively inefcient, rms will prefer to be associated with other rms that are diversied into many industries to extract potential benets they otherwise could not extract by themselves (Khanna & Palepu, 2000). Further, as Khanna and Palepu (2000) underscore, there is typically inadequate disclosure or opaque governance and control mechanisms; nancial intermediaries such as investment banks are not fully evolved; and the enforcement of securities regulations is typically erratic. Thus, in light of the accumulating transaction costs these imperfections place upon them, independent rms will nd it more profitable to pursue business opportunities when associated with a large, diversied business group that will act as an intermediary between them and imperfect markets (Khanna & Palepu, 2000). Given that these rms can access a variety of other benets in labor, product, and technology market intermediation as well through their association with business groups, they will prefer association over nonafliation. Therefore, business groups present benet opportunities to unafliated rms to overcome market imperfections (Guillen, 2000). In suggesting a curvilinear relationship between enhanced performance effects through afliation with a diversied business group, Khanna and Palepu (2000) argue that there are also costs to this afliation that need to be considered for a completer picture of business group benets. For example, governance costs might rise due to conict of interest among the afliated rms; economically inefcient decisions may continue due to weak disclosure requirements, ineffective governance mechanisms, and underdeveloped corporate control mechanisms; and xed costs may accelerate in enforcing intra-group contracts. While these cost effects will undoubtedly surface, all things considered, business groups should provide a market intermediation opportunity to rms operating in imperfect capital, product, labor, and technology markets, especially in the earlier phases of a markets development. Thus, in parallel with Guillen (2000) and Khanna and Palepu (2000), we propose that: Proposition 1. The deeper the imperfections in capital, product, labor, and technology markets, the greater the intermediation importance of business groups will be in those markets. Sociologists, for their part, explain business group formation by the gains realized through social solidarity and social structure among component rms that are identiable by such things as region, ethnicity, kinship, or religion, i.e., rms are linked by relations of interpersonal trust developed on
280
the bases of personal, ethnic, and communal backgrounds (Granovetter, 1994). For example, kinship is found to be a strong correlate of business group formation in the Korean chaebols; kinship and ethnic bonding in Chinese business groups in Thailand and Indian groups in East Africa; geographic region and ethnicity-based solidarities in the ethnic enclaves of Cuban business groups in Southern Florida, and in the multiple axes of solidarity based on government, local business, and capital elites found in the grupos in Brazil (Granovetter, 1994). Indeed, as Khanna and Palepu (2000) underscore, some sociologists argue that business groups provide a special kind of kinship in which economic and kinship bonds become inextricably intertwined, and where a moral community is created within which the likelihood of opportunistic behavior, a core component of transaction cost economics, is lessened. They further argue that through their encouragement of information dissemination among group members, by reducing the possibilities of contractual disputes, and by providing a mechanism for efcient dispute resolution, business groups provide mechanisms for solidarity and for the lowering of intra-group transaction costs (Khanna & Palepu, 2000). Thus, we propose that: Proposition 2a. The deeper the social solidarity and the social structure needs of component rms in a market, the greater the importance of business groups will be to the functioning of that market. Some sociologists also feel that rms are by-products of the social and institutional contexts surrounding them, and therefore, different social and cultural patterns will spawn different types of organizations (Guillen, 2000; Kock & Guillen, 2001; Kim et al., 2004). Therefore, as Guillen (2000) suggests, the value-creation potential of business groups will change as their ambient institutional context changes. Further, they will likely ourish in markets with vertical relationships, such as in South Korea, but not in societies where reciprocal or horizontal relationships are the norm, such as Taiwan (Guillen, 2000). This sociological rationale is supported by economic reasoning. For example, Chang and Hong (2000) argue that the economic performance of group-afliated rms are likely to be higher when compared to independent rms as group-afliated rms benet from group membership through the sharing of both intangible (social solidarity) and tangible (nancial resources, intra-corporate trade) ties with other member rms, specifically as a result of the cross-subsidization of various intragroup transactions. Chang (2003) underscores the performance benets of this cross-subsidization view through evidence that shows that controlling shareholders in South Korean chaebols typically use insider information to
281
take direct and indirect equity stakes in promising rms and transfer profits to afliates primarily through intra-group trade. Thus, we propose that: Proposition 2b. The more vertical the pattern of relationships are in a society, the greater the importance of business groups will be to the functioning of that market. Management scholars explain the performance effects of business groups on value creation in certain industry sectors and national economies, and suggest that these groups emerge as a result of entrepreneurs and rms enhanced capabilities in emerging markets for repeated industry entry (Khanna & Palepu, 2000) and asymmetric foreign trade and investment environments typically found in emerging economies (Guillen, 1997, 2000). For example, Ghemawat and Khanna (1998) view business groups as socially counterproductive rent seekers that are able to absorb disproportionate amounts of rent in an economy to the detriment of the majority of the population. In the same vein, Guillen (2000) theorizes that autonomous states with the ability to allocate capital and other resources at will, will encourage a few favored entrepreneurs to enter new industries, and will seek the political support of these entrepreneurs by enabling them to exploit rentseeking opportunities. Thus, the greater the autonomy of the state, the greater the importance of business groups will be in that economy. Thus, we propose that: Proposition 3a. The greater the opportunities for favored rent-creation in a market, the more business groups will ourish in that market. An extension of this reasoning, offered by Guillen (2000), stipulates that business groups will develop and ourish in economies where asymmetries in inward and outward trade and investment ows, typical of emerging economies, persist long enough to allow chosen entrepreneurs and rms to develop and maintain an inimitable capability to combine foreign (i.e., capital, technology, know-how, access to foreign markets) and domestic resources (access to home market labor, capital, and product markets, government favoritism) that encourages them to enter multiple industries. Guillen (2000) suggests that the advantages they will enjoy in this context will only be sustainable to the extent that asymmetries in trade and investment persist over time, making their capability inimitable. Otherwise, any entrepreneur, including foreign multinationals, will be able to enter new
282
industries, currently the province of privileged entrepreneurs. Thus, we propose, in line with this reasoning, that: Proposition 3b. The greater the asymmetries in foreign trade and investment in an economy, the more important the business groups will be in that market. Marketing scholars also comment on the significance of business groups in emerging economies. They study these groups through the lenses of exchange theory and market evolution. For example, in the context of exchange theory, Hitt, Dacin, Levitas, Arregle, and Borza (2000) suggest that business groups in emerging markets will be viewed by developed country rms as owners of unique competencies, such as local market knowledge and market access gateways, and thus as opportunities for partnership with which to enter markets, while emerging market rms will seek nancial assets, technical capabilities, and intangible assets such as marketing expertise in partnerships with their developed market counterparts; these needs will provide the bases for efcient utility exchanges between developed country rms and emerging market business groups. Others propose in the context of market evolution that business groups will evolve through life cycle-like trajectories in emerging markets. For example, Kim et al. (2004) suggest that family conglomerates will evolve through three distinct stages. In the introduction stage, these rms will create an enterprise that responds to marketplace opportunities and fullls unmet needs of consumers. Easy access to favored capital such as government contracts, lack of viable competition, and various early mover advantages will facilitate the enterprises growth in this phase. In the growth stage, these rms will diversify into related as well as unrelated businesses, fueled by the transfer of foreign technology and know-how through alliances with foreign rms, creation or acquisition of nancial institutions to meet capital needs, and the transformation of the enterprise through professional management. In the maturity stage, they suggest, the diversication of the enterprise will slow down or cease, giving way to international expansion opportunities, even through wholly owned subsidiaries and joint ventures in globally competitive markets. Others propose that business groups will evolve through a trajectory, much like the wheel of retailing in marketing, where as one type of business group reaches maturity and starts to decline in importance, another type will emerge to take its place, and following a similar trajectory, will eventually give way to yet another type of business group in response to the changes in the market environment. In their study of the evolution of Turkish business
283
groups, for example, Yaprak et al. (2004), hypothesize that traditionally dominant holding groups might have given way to a new type, emergent (regional/ethnic solidarity) groups, which currently appear to be superseded by two new types of groups, which they label encouraged (government favored), and loosely connected, export oriented, regional OEM supplier networks. Each of these prototypes appears to possess a different portfolio of characteristics and serve a particular purpose in economic development in response to the prevailing market imperfections of its time. This is in line with Khanna and Palepus (2000) ndings on Chilean business groups which showed that the evolution of a groups ambient institutional context alters its value-creating potential. For instance, paralleling Guillens (1997, 2000) rationale, the dominant groups, which ourished in the 1960s to the 1980s in Turkey, appear to have been the by-product of the populist/nationalist era in Turkeys economic development which was characterized by protectionism, import substitution manufacturing, government focus on natural assets, and local ownership development (Yaprak et al., 2004). Emergent networks, by-products of Turkeys turn to foreign investment attraction and opening of its markets to global competition in the 1980s, appear to have evolved from the natural institutional frameworks that were congruent with inward economic stability at home coupled with foreign direct investment attraction from abroad. Encouraged networks and regional networks, byproducts of Turkeys rapid transformation into an emerging economy in the 1990s, appear to have been the result of export-led growth in created, not extracted, assets, coupled with a gradually opening external economy. To complete this life cycle, Turkish business groups would have to evolve through one more phase, much like those in Spain in the 1980s and 1990s, that would involve export-led growth based on created assets, but also pragmatic, internal stability that would help attract foreign direct investment (Guillen, 1997, 2000; Yaprak et al., 2004). While the empirical evidence we present here comes primarily from the Turkish context, we feel that our arguments that underscore this evidence will likely apply in other emerging market contexts, as well. Thus, we offer the following proposition: Proposition 4. Business groups will follow life cycle-like trajectories inspired by the ambient institutional context of their economies, such that one type of group will grow, mature, and decline as others will start to emerge, grow, mature, and decline. This literature in marketing has been paralleled by studies in international marketing on the entry modes rms will, and should, use when entering a foreign market (e.g., Johansson, 1997). This literature has focused on
284
recommendations for expansion matched to the strengths and weaknesses of various rm characteristics, national factors, and specic strategic motives (Dunning, 1995). Studies in this stream have also suggested that any entry into a foreign market will tie the entering rm, implicitly or explicitly, to one of a number of business networks in that market; networks that are themselves embedded within the unique socio-political and economic context of that market (e.g., Guillen, 2000). The appropriate selection of such a network during entry will provide the rm not only with specialized opportunities, but also with substantial consequences on the performance of the foreign entry attempt (e.g., Larson, 1992; Lorenzoni & Loipparoni, 1999). Thus, these studies suggest that such network positioning and the type of network entered through a business group can be a significant ingredient in the success of the entry, particularly when entering emerging markets. The partner selection literature suggests that when entering a foreign market, the rm should seek a local partner with enriched capabilities, such as a track record of successful international cooperation (e.g., Geringer & Hebert, 1991). The rm is also warned about the problems it might face in foreign entry, such as likely opportunism on the part of the local partner and the high probability of failure (e.g., Gulati, Nohria, & Zaheer, 2000). Some emphasis has also been placed in the literature on the receptivity, ability, and willingness of the local partner to embed its foreign partner in its network of alliances (Ahuja, 2000; Osborn & Hagedoorn, 1997). Hitt et al. (2000), for example, suggest that prospective partners need to match their motives to one another. The motives typically associated with emerging market rms center on the need for nancial assets, transfer of technical capabilities and competencies, and a willingness to share their expertise in local market access and knowledge. Conversely, rms from developed economies often seek access to market opportunities and knowledge, distribution channels, stimuli for new products and lower cost entry in return for their unique competencies and knowledge. Thus, we propose, consistent with Hitt et al. (2000) that:
Proposition 5. When entering emerging markets, entering rms and their business group partners must see benets, including those that might accrue to the entering rm from the host partners willingness to embed the foreign partner in its local network of independent rms and ready access to its markets, and those that might accrue to the host partner from the foreign partners willingness to embed the host partner in its global network of afliated rms and ready access to its markets.
285
We now turn to an examination of the evolution of business groups in Turkey, an emerging market in Southeastern Europe. This is worthy for several reasons. First, the analysis of the nature and extent of market activity of each of these groups will help emphasize the role that each type might play in market entry of foreign rms into the Turkish market. Second, our examination will help underscore the validity of the propositions we offered earlier for each of the business group types we have identied. Finally, if the evolutionary pattern we are proposing here is universal, this could help shed light on the development and progression of business groups in other markets, such as in Brazil, Chile, South Korea, and Taiwan.
THE EVOLUTION OF BUSINESS GROUPS IN TURKEY

With its unique geographical location, Turkey is a natural gateway for simultaneous entry into Western Europe, Asia, and the Middle East. It has undergone a prolonged and sustained program of economic reform under a series of secular governments. These reforms have promoted a unique pattern of business group formation into allied rms. After abandoning protectionist economic policies, Turkey has been implementing an open-market economic program since 1983. The program has liberalized the Turkish economy and integrated it into the larger world economic system. These reforms have included reducing the states involvement in the economy, minimizing state intervention, implementing a exible exchange rate policy, liberalizing import regulations, increasing imports, encouraging foreign capital investment, establishing free trade zones, and deregulating nancial markets. Implementation of these economic reforms has created significant structural changes in the Turkish economy (Bugra, 1994). The reform program has faced numerous difculties until recently, however. Turkey suffered through two consecutive economic crises in the last two decades. The sustained problems in this context have been high ination, the sizable public deficit, and the continuing, albeit declining, state involvement in the national economy. Although Turkey opened up its economy to the world, initially in the 1980s through the Ozal reforms, and later through such mechanisms as the Customs Union Agreement with the European Union (1996 to present), foreign rms have shied away from investing directly in Turkey during this period for precisely the reasons the theory we discussed earlier would suggest. First, the political and economic
286
environments, marred by successive crises, have not been very attractive. Second, the consumer market has remained shallow because of these economic and political crises. Third, government red tape and the legal procedures governing foreign investment have remained archaic, with unreasonable bureaucracies surrounding taxes and conict resolution through the courts. On top of all of this, the demand for most consumers goods has remained low, due primarily to successive economic crises the country has experienced. For all these market imperfections, direct foreign investment by multinationals has remained very low. Those that have invested chose to do so by using the major Turkish business groups as gateways to this promising emerging market. This gave the opportunity for these groups to prot greatly from diversifying into areas that were outside of their traditional business areas. Business groups, which gained a significant role in the Turkish economy during this period, diversied their investments into the various sectors of the domestic economy and increased their cooperative arrangements and partnerships with foreign rms. Besides the traditional business groups, new groups also emerged. Different from the traditional groups along several dimensions, such as size, scope, technological concentration, management philosophy, strategy, alliance linkages, structure, and internationalization patterns, these emerging groups started playing significant roles in attracting international businesses into Turkey and in expanding Turkeys export effort abroad. Based on research we conducted in Turkey (Karademir, 2000), we suggest that there are four types of business groups in Turkey: (1) holding company business groups, (2) emergent business groups, (3) encouraged networks, and (4) regional networks. We further argue that each type provides a distinct combination of costs and benets and is involved in a different set of relationships with different sets of rms and institutions (Hoskisson et al., 2000). We now discuss the costs and benets involved in partnering with each of these in entry into the Turkish market, underscoring the similarities and differences among them.
1. Holding Company Business Groups (HCBGs) A small number of highly diversied, family-based holding companies dominate the business scene in Turkey (e.g., Sabanci Holding, Koc Holding, Alarko Holding, and Zorlu Holding). These groups share many common features with multi-activity rms found in other late-industrializing
287
countries (Amsden & Hikino, 1994; Kock & Guillen, 2001). Turkish HCBGs tend to be highly diversied with 60100 companies present in their network, and show rapid diversication in a short period of time paralleling the transformation of the Turkish economy from import substitution manufacturing to export expansion liberalization after 1983 (Bugra, 1994). It is apparent that this transformation was fueled by government incentives provided these rms to invest in underdeveloped but high-growth potential sectors and/or regions of the economy (i.e., the textile and tourism sectors initially, banking and nance in the recent past and retailing, e-business, and technology, most recently). While the formal governance mechanisms of these groups seems to vary, all seem to involve a mix of family control and less traditional corporate governance systems focused on sectors where they have major capital investments (Bugra, 1994). While in industry sectors where they dominate they tend to be tightly integrated and coordinated; in other industries, they are more likely structured, and operate as, networks of rms than single hierarchies. In these cases, their structural patterns tend to follow trends that are consistent with their industry and/or technology and size, rather than conforming to a unied corporate form (Nohria & GarciaPont, 1991; Human & Provan, 1997). Partnerships with HCBGs typically involve investment projects where the HCBG provides local knowledge, access to the national government, and substantial local market experience in addition to a low-cost production platform for export and access to sophisticated managerial infrastructures. While the multinational rm bringing technology and access to capital and to foreign markets may well see an alliance with the HCBGs as consistent with a cost minimization strategy, the HCBG is likely to view its alliance with the foreign enterprise as a way of lling structural holes in its enterprise network. This is because HCBGs do not tend to be singular organizational hierarchies with a single mode of organization, one dominant strategy, or even a singular managerial philosophy. As the business group literature suggests, they tend to be clustered networks with strong direct ownership ties that are reinforced by strong social ties based on kinship within a given industry, but comparatively weak, indirect ties across industries (Ahuja, 2000; Walker, Kogut, & Shan, 1997). Thus, from the perspective of the HCBG, a partnership with a multinational enterprise represents a way of internalizing an organizational form in place of the market (Williamson, 1991). While for the multinational enterprise there is always the risk that once this gap is lled when the HCBG internalizes the knowhow, coordination, and control benets from the partnership HCBG managers can make the links with the market directly, this is unlikely to occur
288
as long as the multinational rm is able to keep some of these assets proprietary and continues to innovate and thus preserve its superiority in know-how and other competencies.
2. Emergent Business Groups (EBGs) EBGs are closely diversied businesses in target industries, such as foods and cereals, leather products, automotive products, machinery, tourism, and so on, with networks of subsidiaries composed primarily of small- and medium-sized enterprises. They tend to be entrepreneurial rms that enjoy very high growth rates. These groups appear to have followed growth paths similar to the HCBGs. Like their more dominant counterparts, they owe their extensive growth to government efforts to liberalize the economy. They typically enjoy cost economies, due almost exclusively to their locations, which are targeted by the government for export-led growth, primarily to the Turkic republics in Central Asia and Eastern Europe, but also to advanced economies. They enter businesses faster than their HCBG counterparts and their foreign rivals; hence, some have had a tendency to overdiversify into unrelated industries. While some are suppliers to the HCBGs, they tend to realize that an exclusive focus on serving these larger HCBGs in the form of Original Equipment Manufacturer (OEM) supply might be a disadvantage in the long run. In addition to the lower costs and exibility they offer, many seek to turn their local market knowledge into an important competitive advantage. Thus, EBGs have become attractive enterprises for foreign partnerships. Like their more dominant counterparts in the rst group, they tend to enter international alliances to overcome economic uncertainties and their needs for capital and other resources, such as advanced technology, and managerial know-how. Multinational enterprises that attempt to embed themselves into these networks face risks and challenges, however. EBGs limited international experience and managerial talent as well as their entrepreneurial approach to business pose major difculties, as do their tendency to be undercapitalized and over-diversied. For those enterprises that are looking for partners with initiative, energy and drive, and are willing to cope with these challenges, EBGs offer an excellent opportunity for local market development as well as international expansion. Most EBGs do not possess fully developed closed networks. They tend to be connected via social ties to local nancial institutions, governmental agencies and customers as well as to suppliers and economic agents in labor
289
markets. They are, typically, not completely institutionalized within the national economic fabric, but are regional exemplars with a growing national presence (Baum & Oliver, 1991). Thus, EBGs represent an important future for the Turkish economy, but their readiness for international expansion remains suspect. Those that are likely to be the most effective in internationalization are likely to be the ones that view themselves not as mere production platforms, but as centers of networks linking the East with the West (Lorenzoni & Fuller, 1995). As their potential as productive partners comes not only from their location and internal competencies but also from their advantageous positions in regional networks, they represent excellent opportunities to multinational enterprises for productive partnerships.
3. Encouraged Networks (ENs) With the development of Turkeys Customs Union agreement with the EU in 1996, the government has embarked on a number of initiatives to strengthen the international competitiveness of its small- and medium-sized enterprises. One of these initiatives has been the encouragement of sectoral foreign trade companies. Much like the entities legislated into being through the Export Trading Company Act in the United States in 1983, the purpose of this initiative was to encourage Turkish SMEs into export and international investment involvement. ENs are groups of local and regional rms producing similar products that are encouraged to cooperate in their international involvement efforts, and as such, they are exempt from antitrust legislation. What these business groups provide are economies of scale and scope, and specialization in those areas where they possess a comparative advantage. As the ENs are chartered under government legislation and are exempt from antitrust provisions, they are allowed to expand up, down and/or across their initial value-added chains and industry sectors. Many ENs have evolved into loose confederations of rms where only selected members participate in the international expansion of any one project. The degree to which they have the internal capabilities to effectively promote exports is sometimes questioned, however (see Karademir, 2000). When viewed from a network perspective, these groups present an interesting picture, as they appear to be connected to one another as well as to the central body through common social ties, economic interests, and in some cases, kinship. From a contractual standpoint, they appear to operate as an integrated entity when exporting and dealing with governmental
290
agencies. Internally, however, they appear to be loose confederations of independent suppliers, each with a specic specialty. While we observe legal coordination among the members of these groups, most appear to lack a dened strategy, managerial direction for consistent strategic implementation, and the information systems necessary for effective linkage to their members. Social ties appear to be used to compensate for the lack of formal coordination and control. Most appear to go outside their membership groups for key resources, such as nancial needs and foreign experience. While generally resource poor, ENs present promising opportunities, especially to smaller multinational rms, as potentially productive alliance partners, particularly those that involve the use of Turkey as a production platform for export into the EU and the Turkic republics in Central Asia and Eastern Europe. Beyond the challenges of effective partnering, however, the multinational partner would have to be concerned about the willingness of the EN members to accept centralized strategic and technical guidance from the foreign partner, and its ability to coordinate the often difcult operations of the members (Gulati, 1999; Kogut, 2000).
4. Regional Networks (RNs) Regional networks (RNs) are collaborating enterprises composed of smalland medium-sized rms, which are connected loosely through supplier arrangements, co-production, technical tie-ups and the like in industrial districts with a focus on selected industries (textiles in the Denizli, automotive components in the Kocaeli regions). These networks are formed to generate more advanced technological and marketing knowledge, higherquality products, better knowledge of government incentives and controls, and qualied key personnel (Dyer & Nobeka, 2000). They also provide to participating rms access to nancial resources which individual rms cannot otherwise gain access to as easily. They engage in contractual arrangements around specic projects through which they produce parts or products for original equipment manufacturers, export abroad, and manufacture goods for nal sale in the domestic market. The critical element here is the presence of trust that keeps their loose arrangement together. In many respects, these are similar to the industrial networks of third-tier suppliers in the more advanced economies (see Lorenzoni & Loipparoni, 1999). However, they are not as closely coordinated as industrial districts in more developed countries and often lack the sophisticated technology needed to compete in the international scene.
291
CONCLUSIONS AND FUTURE RESEARCH QUESTIONS

Although business groups have emerged recently as significant forms of industrial organization, especially in emerging markets, they have not received sufcient research attention in the international marketing literature. Further, while their significance in market entry has been known for some time, their role in the functioning and evolution of their ambient environments has not been studied extensively. While extant entry mode literature suggests that foreign market expansion is a function of the match between rm needs, investment site alternatives, environmental factors, international markets, and specic strategic motives (Dunning, 1995), this literature has not considered the role that business groups in emerging markets might play as local networks in the international expansion of developed country rms or their expansion from emerging markets. In this paper, we attempted to contribute to lling this void. We rst presented the relevant literature on why business groups exist and how they evolve in emerging markets through ve sets of propositions we derived from economics, sociology, management, marketing, and international marketing. We then presented a broad-brush picture of the evolution of business groups in one emerging market, Turkey. We showed that entry into an emerging market might involve explicit or implicit ties to a number business networks that are themselves typically embedded within a sociopolitical and economic context. Using Turkeys economy and her business networks as a backdrop, we described four distinct types of business networks and derived a rationale for how each might be used by prospective foreign investors as platforms into other markets. Our work led to a number of questions that could be explored by other researchers. First, much more theory development is needed to better understand how business groups function and evolve in emerging economies. For example, studies should explore how such theories as social exchange in anthropology, resource dependence in sociology, and complexity in organizational behavior might explain the roles of business groups. Second, research could probe the similarities and differences, both contextual and corporate culture based, among the compositions and functioning of different types of groups in different emerging markets, say Chile vs. India, Brazil vs. China, or South Korea vs. Turkey. Third, scholars could question why certain types of groups are more effective while others are not even within the same environmental context, and the antecedents and metrics of successful market entry through partnerships with business groups. Finally,
292
researchers could also examine whether afliation with business groups lead to productive benets for multinational rms as well as for the groups themselves when compared to independent entry into, and expansion from, emerging markets. We hope that our paper will inspire work on these questions as we believe that expansion of the existing and creation of new knowledge in this area should also enhance our understanding of business groups and their role in effective international marketing practice.
ACKNOWLEDGMENTS
The authors are indebted to the two anonymous reviewers and the co-editors of this volume for their constructive comments on an earlier draft of this article. Their careful reading and suggestions improved our work immensely.
REFERENCES
Academy of Management Journal. (2000). Special Issue on Business Groups in Emerging Economies. Ahuja, G. (2000). The duality of collaration: Inducements and opportunities in the formation of interrm linkages. Strategic Management Journal, 21, 317343. Amsden, A. H., & Hikino, T. (1994). Project execution capability, organizational know-how and conglomerate corporate growth in late industrialization. Industrial and Corporate Change, 3, 111148. Baum, J. A. C., & Oliver, C. (1991). Institutional linkages and organizational mortality. Administrative Science Quarterly, 36, 187218. Bugra, A. (1994). State and business in modern Turkey: A comparative study. Albany, NY: State University of New York Press. Chang, S. J. (2003). Ownership structure, expropriation, and performance of group afliated companies in Korea. Academy of Management Journal, 46(2), 238254. Chang, S. J., & Choi, U. (1988). Strategy, structure, and performance of Korean business groups: A transactions cost approach. Journal of Industrial Economics, 37(2), 141159. Chang, S. J., & Hong, U. (2000). Economic performance of group-afliated companies in Korea: Intragroup resource sharing and internal business transaction. Academy of Management Journal, 43, 429448. Day, G. S., & Montgomery, D. B. (1999). Charting new directions for marketing. Journal of Marketing, 63, 313. Special Issue. Dunning, J. H. (1995). Reappraising the eclectic paradigm in an age of alliance capitalism. Journal of International Business Studies, 26(3), 461492. Dyer, J. H., & Nobeka, K. (2000). Creating and managing a high performance knowledge-sharing network: The Toyota case. Strategic Management Journal, 21, 345367. Special Issue. Geringer, J. M., & Hebert, L. (1991). Measuring performance of international joint ventures. Journal of International Business Studies, 22(2), 249264.
293
Ghemawat, P., & Khanna, T. (1998). The nature of diversied business groups: A research design and two case studies. Journal of Industrial Economics, 46, 3562. Granovetter, M. (1994). Business groups. In: N. Smelser & R. Swedberg (Eds), The handbook of economic sociology (pp. 453475). Princeton, NJ: Princeton University Press. Guillen, F. M. (1997). Business groups in economic development. In: Academy of Management Best Paper Proceedings, pp. 170175. Guillen, F. M. (2000). Business groups in emerging economies: A resource based view. Academy of Management Journal, 43(3), 362380. Gulati, R. (1999). Network location and learning: The inuence of network resources and rm capabilities on alliance formation. Strategic Management Journal, 20(5). Gulati, R., Nohria, N., & Zaheer, A. (2000). Strategic networks. Strategic Management Journal, 21, 203215. Hitt, M., Dacin, M. T., Levitas, E., Arregle, J. L., & Borza, A. (2000). Partner selection in emerging and developed market contexts: Resource based and organizational learning perspectives. Academy of Management Journal, 43(3), 449467. Hoskisson, R. E., Eden, L., Ming Lau, C., & Wright, M. (2000). Strategy in emerging economies. Academy of Management Journal, 43(3), 249267. Human, S. E., & Provan, K. G. (1997). An emergent theory of structure and outcomes in small rm strategic manufacturing networks. Academy of Management Journal, 40, 2. Johansson, J. K. (1997). Global marketing: Foreign market entry, local marketing, and global management. Chicago: Richard Irwin. Karademir, B. (2000). KOBIlerin Sebeke Organizasyonlarinda yerAlmalarinin Orgutsel Sonuclari: Sektorel dis Ticaret Sirketleri Uzerine bir Arastirma, Unpublished MBA thesis, Cukurova University, Adana, Turkey. Khanna, T., & Palepu, K. (2000). The future of business groups in emerging markets: Long run evidence from Chile. Academy of Management Journal, 43(3), 265285. Khanna, T., & Yafeh, Y. P. (2005). Business groups in emerging markets: Paragons or parasites? Center for Economic Institutions Working Paper Series, No. 2005-1, 86. Kim, D., Kandemir, D., & Cavusgil, S. T. (2004). The role of family conglomerates in emerging markets: What Western companies should know. Thunderbird International Business Review, 46(1), 1338. Kock, C. J., & Guillen, M. F. (2001). Strategy and structure in developing countries: Business groups as an evolutionary response to opportunities for unrelated diversication. Industrial and Corporate Change, 10(1), 77113. Kogut, B. (2000). The network as knowledge: Generative rules and the emergence of structure. Strategic Management Journal, 21, 405425. Special Issue. Larson, A. (1992). Network dyads in entrepreneurial settings: A study of governance of exchange relationships. Administrative Science Quarterly, 37, 76104. Lorenzoni, G., & Fuller, C. (1995). Creating a strategic center to manage a web of partners. California Management Review, 37, 3. Lorenzoni, G., & Liopparini, A. (1999). The leveraging of interrm relationships as a distinctive organizational capability: A longitudinal study. Strategic Management Journal, 20(4), 317338. Nohria, N., & Garcia-Pont, C. (1991). Global strategic linkages and industry structure. Strategic Management Journal, 12, 105124. Summer Special Issue. Osborn, R. N., & Hagedoorn, J. (1997). The institutionalization and evolutionary dynamics of inter-organizational alliances. Academy of Management Journal, 40, 250268.
294
Prahalad, C. K. (2005). The fortune at the bottom of the pyramid: Eradicating poverty through profits. Philadelphia, PA: Wharton School Publishing. Prahalad, C. K., & Lieberthal, K. (1998). The end of corporate imperialism. Harvard Business Review, 76(4), 6879. Walker, G., Kogut, B., & Shan, W. J. (1997). Social capital, structural holes and the formation of an industry network. Organization Science, 8, 109125. Williamson, O. E. (1991). Comparative economic organization: The analysis of discrete structural alternatives. Administrative Science Quarterly, 36, 269296. Yaprak, A., Karademir, B., Ozgen, H., & Osborn, R. N. (2004). Complex congurations in the Turkish economy. In: N. Boyacigiller & T. Kiyak (Eds), Academy of International Business Annual Conference Proceedings.
SMALL MANUFACTURING FIRMS INVOLVEMENT IN INTERNATIONAL E-BUSINESS ACTIVITIES

Per Servais, Tage Koed Madsen and Erik S. Rasmussen
ABSTRACT
E-business is an important business tool, and the increasing presence on the internet reects this fact. For small- and medium-sized rms (SMEs) interested in internationalizing, their internet offers some advantages, because, with e-business, borders between countries are becoming less relevant, and more direct interaction between business entities is made possible. In this article, we unravel the use of internet usage of different types of rms. First, we present a categorization of different local and international rms, and, second, we focus on the internet usage by born global rms compared to the other types of rms. We conclude that born global rms use the internet to convey their market presence, but only to a limited extent do they sell their products via the internet. Instead, they use the internet to support the already existing relationships by describing their products on web pages, offering services related to their products via
297
298
PER SERVAIS ET AL.
the internet, facilitating product development via the internet, and building and maintaining relations to foreign customers. We also stress the importance of further research on how born global rms adapt to the internet in practice.
INTRODUCTION
Several studies have underpinned that the interaction and communication of small- and medium-sized rms (SMEs) with foreign customers, partners, and other network actors is a central issue in the internationalization of such rms (Coviello & Munro, 1997; Coviello & McAuley, 1999; Ellis, 2000; Saarenketo et al., 2004). Crick and Jones (2000) furthermore argue that some small rms, especially in the high-tech business, often possess the competence, capability, and experience needed to operate on distant international markets from early on in their development. Indeed, the e-business revolution is transforming organizations and organizational processes and creating new opportunities and challenges for international marketers as many markets are becoming borderless and globally integrated (Hitt, Ireland, & Hoskisson, 1999). Quelch and Klein (1996) stress that the usage of e-business tools fosters the emergence of instant international companies having immediate access to potential customers, including both businessto-business customers and end users. Others argue that e-business enables greater integration of businesses across national and organizational borders creating new opportunities for SMEs and their supply chains (Lancioni, Smith, & Oliva, 2000). Such an integrated market environment, however, also levers more potential challenges to the SMEs ability to maintain global efciency (i.e., lowering costs and quickening responses in globally expanded supply chains), while attempting to fulll fast-changing, local requirements within particular foreign markets. The overall purpose of this article is to explore the e-business usage among instant international rms (Hamill, 1997) and compare their e-business behavior with that of other types of rms which are more nationally oriented in their sales and sourcing activities. The main contribution of the article is an empirical study among more than 1,000 Danish manufacturing companies in which it is possible to explore differences between such different types of rms. The reason for this focus is that we expect highly internationally oriented rms to behave differently with regard to ebusiness activities.
SMEs Involvement in International E-Business Activities
299
MANUFACTURING FIRMS INTERNATIONAL EXPANSION

According to Melin (1992), rms internationalization can be perceived as a part of an ongoing strategy process of the rm. The main difference between internationalization and other types of growth strategy processes can be found in two dimensions. First, the rm transfers products, services, and resources across borders. This implies that the company has to select in which country or countries the transaction should be performed. Second, the rm has to choose how to enter foreign markets (Andersen, 1997). Firms international expansion is, however, not conned to export activities alone. In a review of the academic literature (Liang & Parkhe, 1997) reveals a striking imbalance while one side of the coin, the exporter side, has been extensively studied, the other side, the importer side and the associated motives, have largely been neglected. This nding is further stressed by Lye and Hamilton (2000). In a similar vain, Cavusgil (1998) highlights perspectives on some fundamental questions in the eld of international marketing and points out some promising research avenues. Among the mentioned avenues is the analysis of rm expansion in international markets, e.g., inward and outward internationalization. Since the late 1980s, a growing concern for manufacturing rms international sourcing (inward) activities has emerged, stressing the importance of this phenomenon on the rms success (Scully & Fawcett, 1994). The connection between import and export activities within a rm and how it affects the internationalization process of the rm has received only scattered interest in the literature (Korhonen & Welch, 1996). The dovetail has been identied primarily through studies on specic business operations, such as licensing, subcontracting, and countertrade. In a study of a large number of Finnish industrial rms, Korhonen and Welch (1996) found that the vast majority of Finnish rms began international activities on the inward side rather than with export, thus pointing to the potential importance of inward activities as a jumping-off ground for outward activities. Typical inward operations were imports of physical products like raw materials, machinery, and components, but also imported services like installation, testing, servicing, and maintenance, although at a low level compared to physical products. Korhonen and Welch (1996) stress that importexport connections could be found at different stages of the internationalization process. These observations are in accordance with Oviatt and McDougall (1994) whose original idea was that the sourcing of input as well as the selling of
300
PER SERVAIS ET AL.
output should be important dimensions in the categorization of International New Ventures. We suggest dening three categories with regard to sourcing as well as selling. One nationally oriented group of rms with no foreign activities, a group of the most globally oriented rms having more than 25% of activities outside own continent, and a group in between with foreign activities, but less than 25% outside its own continent. The cutoff point (25%) is somewhat arbitrary, but for North American rms it follows the definition suggested by Knight (1997). As regards the time horizon within which a rm should attain the level of sourcing and selling abroad, we advocate three years, which also follows the definition by Knight (1997). Such a time horizon may balance the quest for a short time horizon due to market conditions often being highly globalized and the tradition of entrepreneurship research in which rms are often treated as new up to six years after their inception (Oviatt & McDougall, 1997). In order to limit the number of types of rms, we suggest collapsing them into ve types in the subsequent analyses of our empirical data. As shown in Fig. 1, this results in a 3 3 matrix, but in order to limit the number of types of rms we have collapsed them into ve types in the subsequent analyses of our empirical data. We suggest adopting the name born global rms to the rms whose combined sourcing and selling activities are most internationally oriented. Firms having no foreign activities within three years are domestic new ventures or born local rms, whereas rms that only internationalize their sourcing side or selling side are called born international sourcers or born
Classification of New Ventures
No foreign sourcing after three years Domestic New Ventures /Born Local Firms Born International Seller
Less than 25% of foreign sourcing outside Europe within three years Born International Sourcer
More than 25% of foreign sourcing outside Europe within three years Born International Sourcer
No foreign sales after three years

Less than 25% of foreign sales outside Europe within three years More than 25% of foreign sales outside Europe within three years
Born European Firm
Born Global Firm
Born International Seller
Born Global Firm
Born Global Firm
Fig. 1.
Classication of Firms According to Sales/Sourcing Activities.
301
international sellers, respectively. The born European rms constitute a quite diverse group, including rms that may have limited international involvement, but also rms that are highly international at a European scale. New information technology may serve as a vehicle for the phenomenon of instant internationalization. As shown above, we operate with several types of such rms in our empirical study, and we will explore to what extent the different types, in their business activities, adapt to new information technology in general and e-business tools in particular. Before establishing hypotheses, we will give a short review of the issues discussed in the literature about the use of e-business in internationally oriented rms.
E-BUSINESS AND INTERNATIONALIZATION

E-business is an important business tool, and the increasing presence on the internet reects this fact. For internationally oriented rms, e-business offers some advantages since borders between countries become less relevant because more direct interaction between business entities is made possible. Some researchers even mention the term revolution (Zinkhan & Watson, 1996) as a description for the recent development; others are more moderate and claim that The internet is the single most significant new tool for business, particular for small- and medium-sized enterprises. Prior to the internet, most small companies only had access to the local markets, now on the internet every business is a multinational business in a global marketplace (Ellsworth & Ellsworth, 1995, p. 6). However, regardless of how one prefers to describe the situation, it is a fact that the usage of the internet has increased dramatically during the 1990s (Zinkhan & Watson, 1996). The ability to promote, sell, and deliver goods and services via computer technology has spawned as virtual revolution (Zinkhan & Watson, 1996). This rapid penetration of the internet has led rms to embrace the internet at a remarkable pace (Cockburn & Wilson, 1996), mainly motivated by the fear of missing an opportunity or of damaging their image by not doing so. According to several surveys, rms all over the world have started extending their internet-related activities following a goldrush model, unsupported by a clear strategy (Angehrn, 1997) and without any specic knowledge about the consequences for their network relationships. E-business is particularly appealing for SMEs, removing or reducing some of the traditional barriers they faced in doing business overseas, such as communications costs, long distances, and market entry risks. Indeed, some have argued that the internet has reduced the difference between SMEs and MNCs, giving the
302
PER SERVAIS ET AL.
former a possibility for presence in different international markets as well as easy contact with actual and potential customers, suppliers, and partners abroad (Hamill, 1997). In a recent paper, Petersen et al. (2002) discuss three hypotheses regarding the effect of e-business on the internationalization process of rms. The rst hypothesis is that the effect of the internet on rms foreign market expansion will be very modest. This expectation is based on the notion that an important limiting factor in the internationalization process is the accumulation of experiential knowledge, which remains largely unaffected by the internet. In contrast, their second hypothesis is one of the faster foreign market expansion as a result of the borderless marketplace induced by the internet. Their third hypothesis predicts that a gradual contraction will emerge as geographically diversied rms experience the serious cost disadvantage of foreignness in certain markets. The hypothesis maintains that the internet will allow rms to undertake faster international expansion than has been seen until now. However, attempts to penetrate foreign markets more thoroughly will reveal the borderless marketplace to be illusory, the result being gradual contraction. Petersen et al. (2002) maintain that rms with similar products may be affected in different ways, indicating the complexity of the effect that the internet has on the internationalization process of rms. SMEs can use the e-business to support their export operations in three ways: (1) as a global marketing tool, through which it may send and search information, (2) as a cost-efcient transaction medium, and (3) as a tool of customer care. One advantage of the e-business as a marketing tool is that a web page is available 24 h a day all over the world, which gives the rm a global presence. At the same time, through the adaptation of the displayed information to local visitor needs, it provides an opportunity for the SME to be local in many different markets. Foreign visitors interact with the information provided on the web pages, and this gives many opportunities for the SME to learn more about the interests of visitors, to collect specic information directly, and to involve the visitor in a continuing interaction. Web sites can incorporate substantial amounts of information, that is up to date and easy to access, and they are an efcient and low-cost way of distributing materials (e.g., in pdf format). Such comprehensive information will attract the interest of competitors as well as potential buyers, yet access to some areas of the site can be restricted (e.g., password protected). E-business provides the possibility of carrying out transactions, but is also an option for the SME to facilitate much interaction with customers and partners, including, e.g., ordering procedures (direct or through resellers), price transparency, delivery options, electronic payment methods, and
303
security. E-business effectively promotes direct trading between sellers and buyers, which may cause channel conict if distributors have historically been involved in sales, marketing, and servicing of the products. Intermediaries such as foreign agents and distributors may become superuous, but they may also retain their position in international business, if they are able to provide services that are valued by the parties. In some cases, they have a local servicing capability, whereas in others they assist with access to the customers. So, e-business technology permits SMEs to offer their products and services worldwide, but of course they have to observe the possible drawbacks. For example, competitors may observe their strategies and types of relationships with customers, and a small rm has to be realistic as to what extent it is able to customize information and interaction features in different languages. The e-business readily lends itself to customer-care activities. A variety of tools could be implemented, including developing web site contents that will help buyers install and use the product, providing information on new and complementary products, and listing answers to frequently asked questions. Many companies publish a newsletter that is electronically distributed to interested persons, while others enable return-visitors to quickly access relevant material rather than sending it out through mail. Another vehicle for staying in touch with customers is online questionnaires that invite visitors to give feedback on the company and its offerings. The impact of e-business solutions in general and of the internet is particularly interesting for small businesses as their resources are limited (Poon & Swatman, 1997). Hence they seek solutions in which entry barriers and costs are low (OKeefe et al., 1998). For SMEs, effective use of the internet can provide a low-cost access to international markets (both import and export markets) and may help to overcome many of the barriers to internationalization commonly experienced by such rms. Quelch and Klein (1996) list four barriers to SME internationalization and suggest how the internet can be useful in overcoming these barriers. The rst barrier is psychological: the internet can decrease this barrier by increasing international awareness, condence, and commitment through access to global information sources. The second barrier is operational: the internet simplies operations because export documentation can be handled through electronic data transfers, payments, etc. The third barrier is organizational: the internet provides solutions to the problem of this barrier by offering lowcost market research, improved international knowledge, and reduced dependence on traditional intermediaries. The fourth barrier is related to products/markets: in this respect, the internet facilitates the country/market
304
PER SERVAIS ET AL.
selection, international client feedback, and the adoption of a global niche strategy rather than a country-centered strategy. In the process of seeking foreign opportunities, the internet and other sources of electronic information can be a great source of information about relevant international market opportunities, especially for new rms that lack market-specic knowledge. Moreover, the exploitation of these opportunities is increasingly feasible, also for newly established rms with resource constraints, due to effective and low-cost information technology. Furthermore, regardless of rm age, a systematic investigation of foreign market information can lead the rm to identify opportunities in more markets, because low information costs allow for a more extensive environmental scanning. The realization of exports to more markets can also be the result of unsolicited contact from customers due to internet presence or as a result of matchmaking in virtual communities such as electronic markets (Klein & Quelch, 1996). In any case, integration of e-business in a rms international business model can reduce the complexity of running a geographically disperse organization, making it possible to serve many markets, even with limited resources. Once international markets have been identied, the rm needs to develop an effective market strategy for each specic market. If the rm lacks experiential market knowledge, it can, to a certain extent, compensate for this by obtaining more information from other sources. Hence, e-business-intensive rms, in order to develop market strategies, are likely to rely on information from more sources than rms not using e-business. From the perspective of competitive advantage, advances in e-business have made a broad range of tools (i.e., intranets, extranets, electronic mail services, video conferencing, virtual market places, ERP-systems, etc.) available for small rms at low costs. The consequence is that large rms to a lesser extent have the competitive advantage in terms of the expensive information systems needed to coordinate geographically disperse operations (Knight & Cavusgil, 1996). Other researchers have implied that the adoption or embracing of internet practices is an evolutionary process. For example, Angehrn (1997) presents a model with four elements. The rst one is information space (use of the internet simply as a global billboard), while the second is communication space (interactions occur between businesses and customers as they exchange information, ideas, experiences, opinions, etc.). The third element is distribution space (products are actually distributed via the internet), and the last one is transaction space (formal transactions occur in-between businesses, between businesses and consumers, and in-between consumers). Another model is offered by von Versen (1998) in which it is proposed that the internet
305
is both a market and a medium and, as such, performs three basic functions, namely contact (demanders and suppliers exchange information), contact and distribution (demanders and suppliers perform transactions in a global shopping center), and integration (suppliers form virtual companies, demanders form virtual communities, and they all merge into internet conglomerates). Although a number of other models exist (Timmers, 1998), most of the models reveal a pattern of development from simple, one-to-many information dissemination models to more sophisticated, integrated, and cooperative models. As we have seen, the internet has been dened as both a medium and a market space. A medium implies that it is simply a tool for providing information, whereas a market space implies that it is a driver and facilitator of interaction and integration. Thus, at one end of the continuum we will nd information and at the other end integration. Businesses will nd themselves at various locations along the continuum, depending on their internal strategic considerations as well as on external factors. Additionally, the models have implications both for end-use consumers and/or businessto-business customers. Thus, an integrated model might include a medium/ market continuum within both consumer and business environments. In the ensuing we will explore four dimensions of e-business. The four dimensions attempt to span the main aspects of e-business discussed above. They categorize e-business activities into types of usage, depending on the purpose of the activity: (1) Information. Many companies (and even consumers) rst utilize the internet as a way to provide information about the organization, its products, and/or establish a business image. Firms may simply put product catalogues onto web pages without offering an option of interaction or transactions (similar to the one-to-many mass communication models involving television or radio). These rms may also utilize the internet to access information on competitors, the industry, and perhaps customers. However, they are in many respects passive in their communication form. (2) Interaction. Firms may go further and utilize the internet to actually interact with customers, distributors, suppliers, competitors, etc. This interaction may be in the form of customer support or service, price inquiries, or delivery status. Information is not only provided and sought, but also actually exchanged between the parties, who interact with each other. The interaction is often one to one, but does not yet involve actual transactions. This category can be referred to as interaction space.
306
PER SERVAIS ET AL.
(3) Transaction. A third category involves actual economic transactions. Here, rms and consumers actually exchange products, payments, and services between and among each other. The transactions are often sustained through barter, money, services, etc. Moreover, the transactions not only involve product and price decisions, but also often distribution arrangements as well. This is probably the fastest-growing example of e-business today, as more and more organizations desire to actually sell products and services on the internet. This category can be referred to as transaction space. (4) Integration. Finally, an emerging category involves the use of the internet for integration of rms and customers. This involves the ow of information between multiple parties for multiple purposes and for the establishment of relationships. What emerges is almost a living organism, as rms no longer view themselves as stand-alone rms. Instead, they become part of an international supply chain with other rms and potentially even customers, all of whom are sharing and exchanging value. In this stage, different rms in a supply-chain share processes, not just data, electronically (Simchi-Levi, Kaminsky, & Simchi-Levi, 2000). This category of usage can be referred to as integration space where the impact of e-business on a network orientation for market entry can be maximized. In the following section we formulate hypotheses about the association between the type of new ventures and the usage of e-business.
HYPOTHESES
In the following section, we will analyze the use of the internet in the ve different types of rms categorized earlier; the born local rm, the born international sourcer, the born international seller, the born European rm, and the born global rm with respect to the following four areas: (1) Providing information. As stated earlier, many rms rst utilize the internet as a way to display information about the organization, its products, and to ash news to interested parties. We have operationalized this aspect by the extent to which the rm has an extensive Danish web page, an extensive English web page, whether it describes its products on the web page and has a newsletter on the web. We hypothesize that born global rms in particular are active information providers.
307
H1. Born global rms use the internet for information providing purposes to a higher extent that other types of rms. (2) Active search of information. The internet offers the possibility to search for new customers, new distributors, and information about competitors as well as market opportunities in an efcient and costless way. Since born global rms are dependent of reaching out for new business opportunities, right from the establishment of the rm, it is expected that: H2. Born global rms use the internet for active information search to a higher extent than other types of rms. (3) Interaction. As stated earlier, rms can utilize the internet to interact with customers, distributors, suppliers, competitors, etc. This interaction is measured by the extent to which rms communicate with customers, conduct product developments via the internet, create and maintain relationships to customers and even use the internet as a platform for activities to access customers in distant markets. We expected that: H3. Born global rms use the internet for interaction purposes to a higher extent than other types of rms. (4) Transaction. We have measured transactions dened as selling of products and services via the internet as well as offering services via the internet. Also in this respect, we propose that born global rms are the heavy users of the internet. H4. Born global rms sell or offer products/services via the internet to a higher extent than other types of rms. (5) Integration/results. Integration could have been an interesting topic, but in order to concentrate on the rst three issues we have omitted the integration aspect from the present, quantitative empirical study. Instead, we have explored the perceived achievements resulting from using the internet. We have asked the rms whether they have obtained new customers because of the use of the internet, whether they have reduced the importance of their intermediates, and whether they regard the internet to be an asset. In accordance with the previous hypotheses, we expected that: H5. Born global rms perceive to have obtained better results by the use of the internet to a higher extent than other types of rms.
308
PER SERVAIS ET AL.
METHODOLOGY
The empirical study was carried out in 2004/2005, the unit of analysis being the business rm. The population includes Danish manufacturing rms in the industries with NACE codes 15-37 (manufacturing), 72.21 (development of software), and 73.1 (high-tech rms). Bakeries (NACE 15.81.20) and the graphic industry (NACE 22) were excluded as in Denmark they are primarily very small and locally oriented businesses. The population of the business rms was identied by means of CD-Direct, published by Kbmandstandens Oplysnings-Bureau and listing all Danish private business rms. On February 1, 2004, we selected business rms from the CD-Direct according to the criteria mentioned above. A total of 3,048 rms in Denmark met the criteria. Table 1 shows some basic characteristics of these rms. As it appears, by far, most of the rms are small and medium sized which is very characteristic of the Danish business community in general. In total, 91 duplicates (e.g., same rm registered at two addresses or two rms being mother and daughter registered at the same address) were identied, leading to a revised population of 2,957 rms. In the process, a total of Table 1. Main Characteristics of the Classication.
Born Local Born Born Firm International International Sourcer Seller Number of rms with less than 50 employees Number of rms with more than 50 employees Year of establishment (mean) Sales outside Europe after three years, average (percentage of total sales) Purchases outside Europe after three years, average (percentage of total purchasing) 45 25 24 Born European Firm 38 Born Global Firm
11
58
19
12
23
12
1959 0.0
1971 0.0
1984 7.8
1979 2.0
1988 33.3
0.0
1.8
0.0
3.0
29.1
309
49 wrong registrations were identied (closed rms, further duplicates, rms with wrong NACE code, etc.), which reduced the population to 2,908 rms. Because of budget and time constraints, it was decided to contact each rm only ve times. If it was not possible to reach the CEO after ve phone calls, the rm was dened as unreachable and thus as not belonging to the nal population. This was the case of a total of 381 rms (dened as unreachable rms below), which lead to the nal population size of 2,527 rms. Out of the nal population of 2,527 rms, 1,456 rms refused to participate in the survey. Lack of time was by far the most common reason for not participating. One thousand seventy one rms participated in the survey. This corresponds to a response rate of 42.4%. Most of these rms, however, only answered seven questions attached to an initial letter to the CEO (these questions asked for information about the activities during the rst three years after inception of the rm). A total of 791 CEOs promised to answer the CEO questionnaire during a telephone interview, but only 385 CEOs actually returned a usable questionnaire, which corresponds to a response rate of 48.7%. The response rate for the CEO questionnaire is 15.2%, if calculated on the basis of the 2,527 rms in the nal population. It was tested whether the respondents were representative for the population with respect to founding year, number of employees, pretax profits, return on assets, geographical location in Denmark, and industry membership. The conclusion is that the nal population of 2,527 rms is representative of the total population of 3,048 rms. Out of the 385 rms who completed the CEO questionnaire, between 284 and 286 rms answered the 17 questions concerning the rms use of the internet, hence the sample in this survey. All questions related to a 7-point scale (1 not at all, 7 to a high extent). Table 1 shows the distribution of rms, classied according to the pervious mentioned categories, who answered the questions regarding internet activities. Born local rms seems to have a larger proportion of larger rms (>50 employees) where as born international sources, born international sellers, and born European rms have a higher proportion of smaller rms (o50 employees). In the case of born global rms, the distribution is nearly equal. Born local rms seems to be the oldest rms (on average established in 1959) where born international sourcers and born European rms were established in the 1970s (1971, 1979, respectively), and born international sellers and born global rms in the 1980s (1984, 1988, respectively). When it comes to purchasing and sales outside Europe, three years after the establishment, one can observe large differences between born global rms and
310
PER SERVAIS ET AL.
the other categories, in percentages of total purchasing/sales the born global rms being much more active in both selling and buying from abroad very early after the establishment of the rm.
FINDINGS
As stressed earlier, many rms initially utilize the internet as a way to provide information about the organization, its products, and to ash news to interested parties. We specifically asked whether the rms had a Danish web page or an English web page, to what extent they had a product description on a web page and to what extent they used e-mail to distribute news. Table 2 shows that born local rms, born international sources, born international seller, and born European rms to a much higher degree use a Danish web page compared to born global rms which not surprisingly are given the definition of born global rms. A WallerDuncan test reveals that the difference is significant (0.05 level). It is, however, interesting to notice that both born international seller and born European rms are quite heavy users of web pages in Danish. When it comes to a web page in English, the situation is reversed so that born global rms uses web pages in English to a Table 2. Firms Usage of the Internet as an Information Tool.
Born Local Firm We have an extensive Danish web page We have an extensive English web page Our products are well described on the web page We have an extensive newsletter on our homepage N Mean SD N Mean SD N Mean SD N Mean SD 115 4.31 2.03 115 3.71 2.35 114 4.83 1.85 114 2.11 1.56 Born Born International International Sourcer Seller 44 4.36 2.01 44 2.95 2.322 44 4.80 1.80 44 2.30 1.61 38 4.13 2.24 38 3.29 2.45 38 5.08 1.81 38 2.61 1.97 Born European Firm 65 4.32 2.10 65 3.80 2.41 65 5.23 1.64 65 2.66 1.90 Born Global Firm 24 2.88 2.42 24 5.29 1.88 24 5.21 1.72 24 2.67 1.71
311
very high extent, and significantly more than the other types of rms (at the 0.05 level according to Dunnett T3 test; not equal variances). Describing the rms products on the internet applies to all groups of rms, with a slightly overweight toward that the more international selling rms. None of the rms use their homepages very much to host extensive newsletters. This could be explained by the fact that more detailed news are sent directly to interested parties, and therefore only more general news are posted on the web pages. To conclude, H1 is only partially supported with respect to the usage of web pages in Danish and English. The more active usage of the internet involves the search of new customers, new distributors, and information about competitors as well as market opportunities. As it appears from Table 3, the most widespread use of the internet is related to the search of competitor information, followed by general search of market opportunities. When it comes to search for new customers, all groups use the internet to some extent, but not as much as was the case for previously mentioned searching activities. The involvement in searching for new distributors via the internet is even lower. This may partially be due the fact that nding distributors is a delicate task demanding several personal meetings, screening activities, and participation in trade fairs. Table 3. Firms Active Use of Internet Search.
Born Local Firm We search for new customers via the internet We search for new distributors via the internet We search for information about competitors via the internet We search for new market opportunities via the internet N Mean SD N Mean SD N Mean SD 112 2.95 1.77 115 2.23 1.57 115 4.49 1.78 Born Born International International Seller Sourcer 43 3.67 1.89 43 2.60 1.72 44 4.89 1.93 38 3.08 1.95 38 1.92 1.42 38 5.13 1.58 Born European Firm 65 3.35 2.00 65 2.69 1.64 65 5.11 1.71 Born Global Firm 24 3.71 2.03 24 2.79 1.91 24 5.42 1.77
N Mean SD
115 3.45 1.82
44 3.66 1.90
38 4.03 1.72
65 3.74 1.92
24 4.38 2.00
312
PER SERVAIS ET AL.
In Table 3, significant differences in the use of the internet (at the 0.05 level or better) are only seen in regard to search activities related to competitor information, with born global rms being the most active. A Waller Duncan test explains this significance by the large difference between born local rms and born global rms. Also with respect to search for new distributors, the WallerDuncan test shows significant differences between the two groups of rms (at 0.05 level). So, H2, stating that born globals were expected to be more involved in active search via the internet, is only partially supported in this study. Overall, however, Table 3 reveals some similarities between international sellers, born European rms and born global rms, with the latter being the most active. As previously stressed, one of the advantages of the internet is that it establishes a platform for interaction much more simultaneous and cost efcient than other platforms. Table 4 reveals that, in general, all rms are heavily involved in communication with their customers via the internet. Again, the born global rms constitute the most active group. Born global rms differ significantly from all other groups (WallerDuncan test; 0.05 level) with regard to the extent to which they build and maintain relationships via the internet and accesses more distant markets. Table 4. Interaction with Customers via the Internet.
Born Local Firm We communicate with existing customers via the internet We conduct product development with existing customers via the internet We use the internet to build and maintain relations with customers We us the internet actively to access customers in distant markets N Mean SD N Mean SD N Mean SD N Mean SD 115 4.72 2.02 115 3.01 2.03 115 3.40 1.80 115 2.50 1.74 Born Born International International Sourcer Seller 44 4.27 2.18 44 3.09 2.19 44 3.27 2.15 44 2.48 1.87 38 4.50 2.14 38 2.84 1.99 38 3.42 1.86 38 2.53 2.08 Born European Firm 65 4.88 2.11 65 2.89 2.08 65 3.68 2.03 65 3.02 2.19 Born Global Firm 23 5.57 2.00 24 4.08 2.23 24 5.00 2.00 24 4.50 1.77
313
Table 4 shows no significant differences when it comes to communication with existing customers or product development via the internet. However, a WallerDuncan test reveals that born global rms and born local rms differ significantly (0.05 level). Hence, since born global rms differ from all other rms, we conclude that H3 is specifically supported when it comes to building and maintaining relationships and accessing more distant markets. Closely associated with the interaction dimension is the transaction dimension. At a rst glance, it is interesting to observe the low degree of involvement in sales via the internet among all types of rms. Perhaps this is not surprising, taking into account that the rms are manufacturing rms producing complex products that are very difcult to sell as stand-alone products. Therefore, many buyerseller relationships on industrial markets are typically closely based on interaction and only a few are more temporally in nature. If the products themselves cannot be sold via the internet, some supporting elements like service could be offered via internet. In Table 5, we observe that born global rms to a significant higher degree offer services via the internet. Hence, H4 is supported only in the case of offering services via the internet. Finally, we have attempted to look into the results from using the internet. The rms do report that they have obtained new customers and have reduced the importance of distributors by using the internet, but has the internet really become an asset to the rm? Table 6 shows significant differences between the different groups of rms. A WallerDuncan test further explains that the born global rms are significantly different from all other groups in regard to having reduced the importance of intermediates, whereas they are only significantly different from the born local rms in Table 5. Sales and Services via the Internet.
Born Local Firm We sell our products/ services via the internet We offer service on our products via the internet N Mean SD N Mean SD 113 1.69 1.20 115 1.79 1.51 Born Born International International Sourcer Seller 44 1.91 1.64 44 1.93 1.73 38 1.63 1.40 38 2.08 1.88 Born European Firm 65 1.74 1.02 65 1.85 1.47 Born Global Firm 24 2.08 1.61 24 3.04 1.99
314
PER SERVAIS ET AL.
Table 6. Results Obtained from Using the Internet.

Born Local Firm We have obtained new customers via the internet We have reduced the importance of intermediates by the use of the internet Use of the internet is a large active to our rm N Mean SD N Mean SD N Mean SD 114 4.83 1.85 115 2.33 1.52 115 4.71 1.79 Born Born International International Seller Sourcer 44 4.80 1.80 43 2.35 1.63 44 4.55 1.78 38 5.08 1.81 38 2.26 1.62 38 5.29 1.33 Born European Firm 65 5.23 1.64 65 2.95 2.08 65 5.15 1.64 Born Global Firm 24 5.21 1.72 24 3.58 2.06 24 5.50 1.50
respect to nding new customers and viewing the internet as an asset for the rm. Hence we regard H5 as supported.
CONCLUSION
The survey has revealed that born global rms to a high extent have English web pages, and that they are using the internet actively for communicating with customers, conducting product development, and building relationships also to customers in distant markets. Born global rms also more often than other types of rms offer product services via the internet, and, generally, they consider the internet as an important asset to the rm. In many respects, they are heavy users of the internet, and their level of satisfaction is high. Internet presence, however, cannot give the rm a sustainable competitive advantage alone, but this article seems to suggest, that born global rms are proactive rms that seek new ways of conducting business internationally; therefore, the internet and e-business is not a goal per se to these rms but just a new media offering new challenging opportunities. Only a few foreign markets can be covered and penetrated by exporting rms alone by using the internet (Petersen, Welch, & Liesch, 2002). Foreign market success requires adjustments in terms of product, appropriate after sales service, etc. in order to meet the customer needs, to comply with government regulations, or to match local business practices and available distribution channels. Such adjustments need a local physical presence and
315
local insight. Particularly for rms planning to sell complex products on foreign markets, building and maintenance of relationships are necessary. Especially in the early stages of a relationship, the person-to-person contact is needed. In fact, the success of exporting rms often depends on the acquisition of experiential on-site knowledge. Such knowledge cannot be developed through the internet, because it deals with explicit, codied rather than tacit knowledge. As we have seen, born global rms use the internet to convey and vehicle their market presence, but only to a limited extent do they sell their products via the internet. Instead, they use the internet to support already existing relationships by describing their products on web pages, offering services related to their products via the internet, facilitating product development via internet, and by building and maintaining relations to foreign customers. However, more research is needed to point down how this is done in reality, e.g., how these relations were established at rst hand. Another concern is the information available through the internet. Much information may simply be false. Therefore, when born global rms heavily use the internet in the search of information about markets, customers, distributors, and competitors, it could be extremely interesting to nd out how they validate the information, and, above all, how this is supplemented by other sources of information. A major obstacle that may limit the usefulness of the internet concerns the language spoken, the cultural content employed at the website, the quality of the telecommunication infrastructures in various markets, access to computer systems, and computer skills. Hence, more information is needed on the impact of such issues. As a concluding remark, however, this article has demonstrated and provided empirical evidence that born global rms are highly active users of the internet in many activities related to customers, distributors, and competitors in international markets. It is beyond doubt that other rms will follow their example. For this reason, it is advisable that future research study these rms in more detail. Research that brings about understanding of the opportunities and weaknesses resulting from the heavy use of the internet will be of great value for many managers in internationally oriented rms.
REFERENCES
Andersen, O. (1997). Internationalization and market entry mode: A review of theories and conceptual frameworks. Management International Review, 37(2), 2742.
316
PER SERVAIS ET AL.
Angehrn, A. (1997). Designing mature internet business strategies. European Management Journal, 14, 361370. Cavusgil, S. T. (1998). Perspectives: Knowledge development in international marketing. Journal of International Marketing, 6(2), 103112. Cockburn, C., & Wilson, T. D. (1996). Business use of the world wide web. International Journal of Information Management, 16, 83102. Coviello, N. E., & McAuley, A. (1999). Internationalisation and the smaller rm: A review of contemporary empirical research. Management International Review, 39(3), 223256. Coviello, N. E., & Munro, H. (1997). Network relationships and the internationalization process of small software rms. International Business Review, 6(4), 361386. Crick, D., & Jones, M. V. (2000). Small high-technology rms and international hightechnology markets. Journal of International Marketing, 8(2), 6385. Ellis, P. (2000). Social ties and foreign market entry. Journal of International Business Studies, 31(3), 443469. Ellsworth, J. H., & Ellsworth, M. V. (1995). The internet business book. London: Wiley. Hamill, J. (1997). The internet and international marketing. International Marketing Review, 14(5), 300323. Hitt, M. A., Ireland, D. R., & Hoskisson, R. E. (1999). Strategic management. Cincinnati, OH: Southwestern College Publishing. Knight, G. (1997). Emerging paradigm for international marketing: The born global rm. Unpublished doctoral dissertation, Michigan State University, East Lansing. Knight, G. A., & Cavusgil, S. T. (1996). The born global rm: A challenge to traditional internationalization theory. Advances in International Marketing, 8, 1126. Korhonen, H., & Welch, L. (1996). Internationalization of SMEs: Inwardoutward patterns and government policy. Management International Review, 36(4), 315329. Lancioni, R. A., Smith, M. F., & Oliva, T. A. (2000). The role of the internet in supply chain management. Industrial Marketing Management, 29(1), 4556. Liang, N., & Parkhe, A. (1997). Importer behavior: The neglected counterpart of international exchange. Journal of International Business Studies, 28(3), 495529. Lye, A., & Hamilton, R. T. (2000). Search and performance in international exchange. European Journal of Marketing, 34(1), 176189. Melin, L. (1992). Internationalization as a strategy process. Strategic Management Journal, 13, 99118. O Keefe, R., O Connor, G., & Kung, H. (1998). Early adopters of the web as a retail medium: Small company winners and losers. European Journal of Marketing, 32(7/8), 629643. Oviatt, B. M., & McDougall, P. (1994). Toward a theory of international new ventures. Journal of International Business Studies. (First Quarter), 4564. Oviatt, B. M., & McDougall, P. (1997). Challenges for internationalization process theory: The case of international new ventures. Management International Review, 37(Special issue 2), 8599. Petersen, B., Welch, L., & Liesch, P. (2002). The internet and foreign market expansion by rms. Management International Review, 42(2), 207221. Poon, S., & Swatman, P. (1997). Small business use of the Internet: Findings from Australian case studies. International Marketing Review, 14(5), 385402. Quelch, J. A., & Klein, L. R. (1996). The internet and international marketing. Sloan Management Review, 37(3), 6075.
317
Saarenketo, S., Puumalainen, K., Kuivalainen, O., & Kylaheiko, K. (2004). Dynamic knowledge related learning processes in internationalizing high-tech SMEs. International Journal of Production Economics, 89(3), 363378. Scully, J., & Fawcett, S. (1994). International procurement strategies: Challenges and opportunities for the small rm. Production and Inventory Management Journal, 35(2), 3946. Simchi-Levi, D., Kaminsky, P., & Simchi-Levi, E. (2000). Designing and managing the supply chain management. McGraw-Hill: Boston, MA. Timmers, P. (1998). Business models for electronic markets. Electronic Markets, 8(2), 38. von Versen, K. (1998). The three commercial functions of the internet. Electronic Markets, 8(4), 2529. Zinkhan, G. M., & Watson, R. T. (1996). Advertising trends: Innovation and the process of creative destruction. Journal of Business Research, 37, 163171.
ARE INTERNET FIRMS GLOBAL?

Stephen Chen
ABSTRACT
This study examines to what extent Internet rms have globalized and the key factors that have enabled some rms to globalize more than others. Contrary to arguments that Internet-based rms automatically benet from a global market, this study shows that most Internet rms serve regional markets, consistent with Rugmans (2000) ndings for rms in the FT500. However, there are a few notable exceptions. In these cases a combination of early mover advantages, unique product, technology standards and complementary products and services have created a winner-takes-all market in which a few rms dominate markets worldwide. Implications for globalization theories are discussed.
INTRODUCTION
Undoubtedly one of the most signicant developments to affect marketing worldwide in the 21st century has been the development of the Internet. One of the many claims was that the Internet would facilitate globalization of rms worldwide (e.g. Petersen, Welch, & Liesch, 2002). With the instant worldwide reach of the Internet it is sometimes claimed that Internet-based rms are the epitome of born global rms (Knight & Cavusgil, 1996; Madsen, Rasmussen, & Servais, 2000) that internationalize early in their life.
International Marketing Research: Opportunities and Challenges in the 21st Century Advances in International Marketing, Volume 17, 319345 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1474-7979/doi:10.1016/S1474-7979(06)17012-X
319
320
STEPHEN CHEN
Several arguments have been advanced why the Internet should increase globalization. Wider access to the Internet will, it is argued, bring about a greater commonality in tastes and demand for certain products as exchange of ideas and information increases across borders. For example, the increase in recognition of certain global brands such as Coke, Nike and IBM can be seen as both drivers and outcomes of this phenomenon. Previous studies (Kotha, Rindova, & Rothaermel, 2001; Gandal, 2001) have found that the level of reputation and brands are key factors in the internationalization of Internet rms. The increasing importance of global standards in certain industries such as software is also seen as a key driver. Computer hardware and software are examples a PC bought in one country is similar in most respects to a PC bought in any other. Finally, one of the arguments for increased globalization among Internet-based rms is that digital goods have certain features that favor globalization, namely low cost of reproduction, low degradability and low cost of distribution across a digital network (e.g. Bakos, 2001; Borenstein & Saloner, 2001). However, despite the extensive theoretical arguments, with a few exceptions (e.g. Kotha et al., 2001), there is still little empirical research on the globalization of Internet rms, and the extent of globalization has not been critically examined. This paper examines the extent of globalization in a sample of leading Internetbased rms, some of the key factors that have enabled or hindered globalization and implications for current theories of globalization.
GLOBALIZATION THEORIES AND PREDICTIONS

Although much has been written about the development of global markets, there are differing arguments about the extent to which globalization of rms is in fact happening or will happen and the relative importance of different drivers and constraints of globalization.
Arguments for Globalization One of the earliest arguments predicting globalization of markets was made by Levitt (1983) who predicted globally converging demands as international travel and global communication increased. Levitt argued that the search for economies of scale in production, distribution and marketing would lead to standardized products produced and sold in the same manner worldwide. Another early argument in favor of globalization was the
Are Internet Firms Global?
321
oligopolistic reaction argument made by Knickerbocker (1973), who observed a bandwagon effect in FDI in which US rms in the same industry react to entry by a competitor into a foreign market by likewise entering the market in order to prevent the competitor from gaining a competitive advantage. Similar arguments were made later by Haveman (1993) from an institutional theory perspective in which rms mimic other rms in order to gain legitimacy and examples have been found in other studies of foreign market entry by rms (e.g. Guillen, 2003). Regionalization Arguments Recently, the view of ever-increasing globalization has come under attack from a number of quarters. Researchers such as Rugman (2000), Rugman and Brain (2003) and Rugman and Verbeke (2004) have shown that the extent of globalization has in most cases been overstated and that rms have generally pursued a regional rather than a global strategy. There are several reasons why a regional strategy is likely to be less costly and risky compared to a global strategy. First, regional markets are located at a more proximate distance. This reduces transportation costs. Second, as they are in the same time zone, it is also easier to coordinate activities. Third, markets in the same region are also likely to be part of the same trade blocs and so benet from reduced market entry barriers. Another factor that could encourage regionalization strategies is the need to localize some products. As described by Prahalad and Doz (1987) in their well-known integrationresponsiveness framework, MNCs must balance the twin demands of integration and responsiveness. A regional strategy may allow a balance between the two, allowing MNCs to obtain some of the benets of globalization while remaining responsive to local market needs. Evidence of this can be seen in the automobile industry (Schlie & Yip, 2000) or the white goods industry (Baden-Fuller & Stopford, 1990). However, as Millar, Choi, and Chen (2005) found, there are exceptions and some rms are truly global. These include rms with unique products for which there is a strong global demand and well-developed global distribution networks. Staged Internationalization Arguments Other researchers argue that a regionalization strategy is merely a stage toward a full globalization strategy. Perhaps the most familiar models of staged internationalization are Vernons (1966) product life cycle and
322
STEPHEN CHEN
Johanson and Vahlnes (1977) incremental internationalization model. These models are based on an assumption of incremental learning and development. The typical process of rm internationalization in the latter half of the 20th century was described by Vernon (1966) as a product cycle consisting of domestic product development, followed by exporting, and then by foreign production. Based on their studies of Swedish manufacturing rms, Johanson and Vahlnes (1977) model showed that initial internationalization activities were targeted to psychically close markets and used modes of entry that require less commitment, such as exporting. Johanson and Vahlne (1977) explained that the rm learned and increased its foreign market knowledge over time primarily through experience, and only then did it increase its foreign market commitments and later expand to more psychically distant markets. Born Global Arguments The staged internationalization model, in turn, has recently been challenged by researchers who have noted that some new venture rms skip stages and/ or have been international virtually from inception (Knight & Cavusgil, 1996; Oviatt & McDougall, 1994; Rialp-Criado et al., 2002). These rms have sometimes been referred to as born global rms (Knight & Cavusgil, 1996; Madsen & Servais, 1997; Madsen et al., 2000; Moen & Servais, 2002) and global start-ups (Oviatt & McDougall, 1994). Some factors that have been suggested as contributing to the increase in born global rms include rapidly changing computer, communication and transportation technology, and changes in the political economy, industry conditions, rm effects and management (Oviatt & McDougall, 1994). Among the technological changes, the widespread adoption of the Internet worldwide is seen by many as key and Internet rms are often regarded as the epitome of born global rms. The Internet as a Driver of Globalization The increasing adoption of the Internet has reignited the debate about globalization. As the brief review above shows, there are a number of different theoretical arguments about the extent to which globalization of rms in general is taking place or will take place. Analogous arguments can be made regarding the globalization of Internet-based rms and the impact of the Internet on globalization of rms in general. A number of possible scenarios
323
are possible depending on the effects of the drivers and constraints of globalization (Petersen et al., 2002). Arguments in support of increased globalization among Internet rms are that the Internet directly affects a number of globalization drivers such as global convergence of tastes, global branding, global standards and global purchasing systems that encourage demand for a common product or service worldwide (Yip, 2000). Many have also argued that the Internet lowers global transactions costs and that the global reach of the Internet enables Internet-based rms to benet from a global market (e.g. Evans & Wurster, 2000; Lituchi & Rail, 2000). Some marketers (e.g. Hamill, 1997; Stevenson & Hamill, 2002) argue that this increasing globalization in Internet markets will have signicant implications for international marketing including standardization of prices, facilitating feedback from customers and reducing the importance of intermediaries. This may require a re-examination of traditional marketing paradigms. For instance, Singh and Kundu (2002) proposed that Dunnings (1973) eclectic paradigm of internationalization needs to be extended to incorporate network-based advantages in ecommerce corporations. These include stronger relationships between rms and their customers or between rms and their partners, reconguration of the industry value chain and network externalities.
Constraints to Globalization of Internet Firms On the other hand, some researchers (e.g. Leamer & Storper, 2001) argue that the Internet will not diminish the importance of location-specic factors. This is supported by some studies that conrm the distinctiveness of Internet markets in different countries (Lynch & Beck, 2001). For example, some of the key factors that have been shown to affect both the uptake and the usage of the Internet in different countries include language, educational level, income, population size, availability of credit, availability of venture capital, taxation, government policy and level of telecommunications infrastructure development (Guillen, 2002; Kshetri & Dholakia, 2002; Sprano & Zakak, 2000). Despite the efforts of many governments and international agencies, there also remain stark differences between the level of Internet usage in developed and less-developed countries (Bridges.org Report, 2001). Even in developed countries, Internet markets may remain local in nature owing to differences in language, culture, perceptions and buying behavior (Lynch & Beck, 2001). Liability of foreignness (Hymer, 1976; Zaheer, 1975) still appears to be signicant. Zaheer and Manrakhan (2001) found
324
STEPHEN CHEN
that locational clusters remain important in B2B trading markets while Zaheer and Zaheer (2001) found that despite the supposed threat to local banks posed by the Internet, local banks are still able to compete successfully with global competitors. As Huang, Keser, Leland, and Shachat (2003) found in their study of Internet adoption worldwide, trust appears to be a signicant factor. Customers are still wary of foreign rms with whom they may be unfamiliar and seek the security of rms with a local physical presence even though they may perform many transactions online. In short, although there are clear instances of Internet rms that have successfully globalized, other studies suggest that Internet rms seem to exhibit limited globalization. Given the conicting theoretical arguments and predictions about the globalization of Internet rms, the motivation for this paper was twofold. The rst aim is to determine to what extent Internet rms have in fact globalized and which arguments are supported or disproved. The second aim is to identify the key factors that have enabled or prevented globalization of Internet rms and what the implications are for theories about globalization of rms in general.
METHODOLOGY
In order to answer the rst research question, the extent of globalization among a sample of Internet rms was measured. Internet rms were dened as rms that derive most of their income from one or more Internet-related activities. These can be divided into a number of layers, of which the key activities include search engines, Internet service providers (ISPs), hardware, software, information services and Internet retail (Barua, Pinnell, Shutter, & Whinston, 1999). It was assumed that the largest rms in each layer would show the greatest degree of internationalization and provide the most easily accessible data, so a sample of companies in each layer was identied from the following rankings of leading Internet companies: Ranking of ISPs by size (available at www.jetcafe.org), Internet Retailer magazines ranking of the top 300 retail websites ranked by Internet sales and USA Todays ranking of the Top 50 Internet companies based on market capitalization. Companies that were subsidiaries of larger companies were excluded unless separate nancial reports were available. Data was gathered from
325
annual reports and other sources available via the Internet. The nal sample comprised 63 companies for which geographic sales data was available. Search engines (9) ISPs (9) Information providers (4) Infrastructure (hardware and software) (20) Retail (21)
The nal sample was biased toward infrastructure and retail rms. However, it also reects to some extent the concentration of rms in each market. It was difcult to identify many other search engines, ISPs and information providers that operated on a global scale. Possible reasons for this will be discussed later. Other studies that have examined internationalization of Internet rms (e.g. Kotha et al., 2001) have measured this in terms of number of international websites, not in terms of revenues. However, as Sullivan (1994) found in his study of US manufacturing rms, presence in foreign markets and the extent of international revenues may not necessarily be strongly correlated. This discrepancy is particularly relevant where the cost of foreign entry is relatively low as in the case of Internet websites. Therefore, this study measured globalization in terms of revenues. This measures more directly the extent to which rms are actually capturing a share of the market globally as opposed to the potential market. First, the percentage of foreign sales/total sales was determined for each company. Those rms that had foreign sales of 25% or more (the cut off gure used by Knight & Cavusgil (1996) to identify born global rms) were highlighted. Second, the percentage of each companys sales by region (North America, Europe, Asia Pacic, South America and Africa) was determined. These data were then used to classify companies according to Rugmans (2000) classication of globalization: (a) Home triad region oriented: rms that have at least 50% of their sales in their home region of the triad. (b) Bi-regional: rms with at least 20% of their sales in each of two regions, but less than 50% in any one region. (c) Host triad region oriented: rms that have more than 50% of their sales in a triad market other than the home triad region. (d) Global: rms that have sales of 20% or more in each of the three parts of the triad. (For comparison, the number of country-specic websites was also determined for each company.)
326
STEPHEN CHEN
In the second phase, the rms that were most globalized were then examined in more detail in order to determine how and why they had globalized more than the other rms. Information on the company history and strategy was collected from company websites, company annual reports, press reports and other publications.
RESULTS
Extent of Globalization Table 1 shows the breakdown of sales by region for each of the companies examined. Table 2 shows the percentages in each of Rugmans (2000) categories and the percentage of born global companies according to Knight and Cavusgils (1996) criteria. (All rms in the sample were considerably older than 3 years, the time period used by Knight and Cavusgil, so the extent of internationalization is likely to be higher than that in a sample of rms less than 3 years old.) As Table 2 shows, only 29% of the rms had foreign sales of 25% or more. Using Rugmans (2000) classication, 55 out of the 63 (87%) rms are home triad region oriented, 6 companies (10%) are bi-regional and only 2 companies (3%) are truly global. These percentages are identical to the percentages of 87%, 10% and 3%, respectively, in the FT500 rms examined by Rugman and Verbeke (2004) and conrms that the extent of globalization has been greatly exaggerated, even among leading Internet rms. Furthermore, since the majority of North American revenues can be assumed to be of US origin in US rms, it is possible to go even further and state that not only are US Internet rms primarily home triad based but domestic market based. These ndings based on revenues contrast sharply with globalization as measured by the presence of international sites (Tables 3 and 4) and is consistent with the ndings of Rugman (2000) who found in his study of FT500 rms that the presence of global ofces does not necessarily equate with global revenues. The discrepancy is possibly even greater in the case of websites since the marginal cost of adding a new website is considerably lower than that of establishing a sales ofce or store in a new country. The Global Companies Although the sales data show quite clearly that, despite some predictions, most Internet rms are home region oriented and domestically oriented,
Table 1.
Company Activity Year Founded North America
Breakdown of Sales by Region.

Europe Asia South America Unknown Global Orientation International Foreign Sales>25%
SEARCH Ask Jeeves Findwhat Google Infospace Looksmart Pricegrabber Primedia Shopping.com Yahoo ISP Alltel Charter Comm. Core Comm. Covad Cox Comm. Earthlink Everyones Internet SBC United Online INFORMATION CNET IVillage Jupitermedia Marketwatch RETAIL Digital goods Ameritrade
Search Search Search Search Search Search Search Search Search
1996 1998 1998 1996 1996 1996 1989 1999 1995
65.3% 100.0% 72.0% 92.0% 100.0% 100.0% 100.0% 89.0% 83.0%
30.5%
4.2% 28.0% 8.0%
11.0% 17.0%
H H H H H H H H H
34.7% 28.0% 8.0%
11.0% 17.0%
Y N Y N N N N N N
ISP ISP ISP ISP ISP ISP ISP ISP ISP
1983 1993 1999 1996 1997 1994 1998 1988 1999
100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 90.1% 100.0% 9.9%
H H H H H H H H H 9.9%
N N N N N N N N N
Internet news Womens ezine Market research Market research
1992 1995 1986 1997
79.1% 100% 95.5% 100.0%
20.9% 5%
H H H H
20.9% 5.0%
N N N N
Online stock trading
1975
100.0%
327
Company Activity Year Founded 1963 1981 1995 1995 1996 1998 1989 1998 1989 North America 100.0% 97.7% 65.0% 92.3% 100.0% 99.5% 55.9% 84.8% 100.0% 2.3% 35.0% 4.3% 0.5% 2.9% Europe Asia South America Unknown Global Orientation H H H H H H H H H 2.3% 35.0% 7.7% International Foreign Sales>25% N N N N N N N N N
328
Charles Schwab Checkfree Ebay Etrade Netbank Priceline Sabre Skillsoft University of Phoenix online USA Interactive Physical goods 1800-Flowers Alloy Amazon Drugstore Ecost Freshdirect FTD Netix Overstock Peapod
Online stock trading Electronic commerce only Auction site Online stock trading Bank Retail Travelocity E-learning Online education
0.4% 44.1% 13.0% 2.2%
0.5% 44.1% 18.5%
1976
95.4%
4.6%
4.6%
Flower delivery Clothing Retail Drugstore Discount store Groceries Flower delivery Video rental Unsold stock Groceries
1976 1996 1994 1998 1999 1999 1910 1997 1999 1989
100.0% 100.0% 78.4% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
21.6%
H H H H H H H H H H
21.6%
N N N N N N N N N N
STEPHEN CHEN
CONSUMER INFRASTRUCTURE SERVICES 24/7 Real Software 2003 Media Net2phone Communications 2003 service
60.0% 50.0%
40.0% 50.0%
H H
40.0% 50.0%
Y Y
Real Network
Software
2003
72.9%
15.9%
9.8%
1.4%
27.1%
BUSINESS INFRASTRUCTURE SERVICES BEA Systems Infrastructure 2003 software Broadcom Communication 2003 systems Cisco Network systems 2003 Doubleclick Advertising 2003 JDS Uniphase Fiber optic cable 2003 Juniper Network systems 2003 Networks Level 3 Communication 2003 Macromedia Software 2003 Mercury Systems 2003 NetIQ Systems, security 2003 and web analytics Network Storage solutions 2003 appliance Netratings Advertising rating 2003 Openwave Software 2003 Siebel Software 2003 Tibco Software 2003 Verisign TTP service 2003 Verity 2004 Knowledge management software
54.5% 74.2% 55.9% 79.4% 70.2% 42.2% 81.1% 58.4% 63.6% 76.6%
29.5% 5.9% 27.6% 16.7% 26.6% 17.1% 26.0% 29.5%
16.0% 19.6% 16.6% 20.6% 13.1% 31.2% 1.8% 15.5% 7.0% 23.4% 0.3%
B H B H H G H B B H
45.5% 25.8% 48.9% 22.9% 29.8% 57.8% 19.9% 41.6% 38.6% 23.4%
Y Y Y N Y Y N Y Y N
57.9% 61.9% 47.8% 58.2% 83.2% 89.3% 65.5%
42.1% 38.1% 20.4% 30.9% 31.8% 7.7% 3.3% 16.8% 10.7% 7.2% H G B H H B
42.1% 38.1% 60.2% 41.8% 16.8% 10.7% 34.5%
Y Y Y Y N N Y
27.3%
Note: All data for the year 2003 except where marked otherwise. 2004 data.
329
330
STEPHEN CHEN
Table 2. Breakdown of Companies by Sales.

Home triad oriented Bi-regional Global Total 55 (87%) 6 (10%) 2 (3%) 63 (100%) Domestic market oriented International market oriented 45 (71%) 18 (29%) 63 (100%)
there are some interesting exceptions. Eight of the rms that have globalized more extensively (i.e. are not home region oriented) are examined below and some key elements in their globalization strategies are identied in Table 5. Internet-Based Startups Four of the eight companies have always focused on Internet-related activities and have expanded globally as the Internet market has taken off worldwide. These are Cisco, Juniper, Openwave and Macromedia. Cisco. Cisco is the dominant player in the market for Internet routers and was one of the earliest movers in the market. In 1984, at the time Cisco was founded, the Internet had only 1,000 computers connected to it. Initally the company focused on the domestic US market, which was and still is the largest in the world. However, as the demand for Internet routers increased dramatically from late 1988 onwards, the company increasingly looked to international markets. In 1991, John Chambers was appointed as senior VP for worldwide operations and under his direction Cisco followed a global network supply strategy. First, the company focused on product design and software development, outsourcing most of the other work to partners to whom they were closely linked in IT networks. Second, the company embarked on a string of acquisitions in related markets worldwide (41 between 1993 and 1999) to enable the company to provide end-to-end network solutions as well as routers. This global strategy enabled the company to achieve international sales of 39% by 1993. By 1998, Cisco held a commanding position in the router market worldwide (80% of high-end router market, 70% of the mid-range market and 64% of the low-end market). While it has lost some ground since then to competitors such as Juniper, it still remains the dominant supplier in the router market with a market share of 81% of the enterprise router market and 42% of the service provider router market in 2006. Juniper. Juniper is Ciscos biggest competitor in the market for Internet routers, but unlike Cisco, Juniper was a late entrant in the market. The
331
Table 3. Breakdown of Sites by Region.

Company North America Europe Asia Pacic South and Central America Africa Number of Regions served
SEARCH Ask Jeeves Findwhat Google Infospace Looksmart Pricegrabber Primedia Shopping.com Yahoo ISP Alltel Charter Communications Core communications Covad Cox communications Earthlink Everyones Internet SBC United online INFORMATION CNET Ivillage Jupitermedia Marketwatch RETAIL Digital goods Ameritrade Charles Schwab Checkfree Ebay Etrade Netbank Priceline Sabre Skillsoft University of Phoenix online USA interactive Physical goods 1800-Flowers Alloy
1 (8) 1 3 3 1 (8) 2 (3) 1 1 3 (7)
1 39 1 1 1 9 20 22 17
1 1 5 1 1 1 1 2 3
1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1
1 1 (3) 1 (5) 1
8 1
2 2 1 1
1 1 1 3 2 1 1 2 1 1 1
11 3 1 5 1
2 2
1 1 1 3 3 1 2 2 2 1 1
1 1 (4)
1 1
332
STEPHEN CHEN
Company North America Europe Asia Pacic South and Central America Africa Number of Regions served
Amazon Drugstore Ecost Freshdirect FTD Netix Overstock Peapod
2 1 (3) 1 1 1 1 1 1
1 1 1 1 1 1 1 1
CONSUMER INFRASTRUCTURE SERVICES 24/7 real media 1 Net2phone 1 Real network 1 BUSINESS INFRASTRUCTURE SERVICES BEA Systems 2 16 Broadcom 1 Cisco 2 15 Doubleclick 1 4 JDS Uniphase 1 Juniper Networks 1 Level3 1 12 Macromedia 2 7 Mercury 1 NetIQ 1 5 Network appliance 1 4 Netratings 1 8 Openwave 1 Siebel 3 13 Tibco 1 7 Verisign 2 4 Verity 1 3
1 1 1
2 2 1 1 2 1 3
2 1 2 2 1 1 2 1 1 3 2 3 1 4 2 3 3
Note: Figures in brackets indicate number of languages where the site is multilingual.
Table 4.
Companies Companies Companies Companies Companies Total with with with with with
Breakdown by Presence of International Sites.

in in in in in one region two regions three regions four regions ve regions 42 12 7 1 1 (67%) (19%) (11%) (1.5%) (1.5%)
websites websites websites websites websites
63 (100%)
333
Table 5.
Company Cisco Juniper Openwave Macromedia Siebel Systems
Key Elements in the Strategies of the Highly Global Companies.

Key Elements of Globalization Strategy Pioneer and market leader in Internet routers Global acquisitions in related areas Develop and market leading high performance product worldwide Target large telcos worldwide that require high performance routers Pioneer in establishing global mobile telephony software standard Target large telcos worldwide Pioneer and market leader in professional web design software Product localization using local vendors Pioneer and market leader in CRM Develop and market Integrated applications suite (Siebel everywhere) Develop web-based product to integrate with existing applications Develop and market integrated application suite Establish company as platform provider Develop web-based product using emerging Java standard International partnerships with complementary service providers Target niche in intelligence and publishing industries in markets worldwide Develop product for analysis of information from the Internet Product localization using local vendors Alliances with leading database vendors that offer complementary products Market leader in systems testing Develop product for applications based on emerging Windows standard Product localization using local vendors
BEA Systems
Verity
Mercury Interactive

company was only established in 1996, as the Internet was beginning to be widely adopted by businesses. Unlike Cisco, which adopted a broad market strategy, from the beginning Juniper has focused on specialized markets where it has clear rm-specic advantages over Cisco. The founders correctly identied that trafc on the Internet was being held up by routers that could not transmit packets fast enough for the network demand. Using the latest technology they designed a router, much faster than Ciscos. This consisted of two key elements: a custom-built ASIC (application-specic integrated circuit) and JUNOS, the companys proprietary operating system. Avoiding direct competition with Cisco in the general corporate market, the company clearly targeted the market for high-performance, high-reliability products routers that are needed by the major ISPs and telecommunications companies and using alliances with strategic partners to
334
STEPHEN CHEN
gain international presence. For instance, the company entered into a marketing alliance with French telecommunication equipment manufacturer Alcatel in order to boost sales in Europe and Asia where Alcatel had a strong presence. This strategy enabled the company to increase its international sales from zero in 1998 to 22% in 1999, just 3 years after its founding, and achieve the No. 2 position in the router market after Cisco with 18% of the enterprise router market, 31% of the service provider core router market and 25% of the service provider edge router market in 2006. Openwave. As in the cases of Cisco and Juniper, Openwave has grown by following a growing market. However, in the case of Openwave, it was by focusing on open standards software products and services for the emerging mobile communications industry. The company has beneted both from the rapid growth in the mobile communications market and from the increasing preference for open source software among customers. As one of the early innovators in mobile data and messaging services since 1993, the company has established a strong presence in the market and followed a global strategy early on. In 1997, just 4 years after its founding, international sales accounted for 45% of revenues. As in the case of Juniper one of the key elements of the companys internationalization strategy has been partnering with leading computing and telecommunications companies worldwide. The company has formed partnerships with leading computing rms such as HP, Siemens and IBM, and customers include four of the top ve mobile operators in the United States AT&T Wireless, Verizon Wireless, Sprint PCS and Nextel as well as global companies such as KDDI in Japan and BT Genie in the UK. Currently its products are used by more than 70 operators and more than 47 handset manufacturers worldwide. These include wireless application protocol, or WAP, gateways, which provide the underlying infrastructure to enable data services on mobile phones, and wireless and wireline applications, such as e-mail and multimedia messaging and related services. Macromedia. While in the previous cases the companies have focused on delivering a standardized product or establishing a technology standard, the key feature of Macromedias internationalization strategy is product localization. The company develops software that enables the development of a wide range of Internet solutions including websites, rich media content and Internet applications across multiple platforms and devices. Like Cisco, the company was founded in the early days of the Internet boom in 1992 and quickly established itself as a leader in the market for website design with its
335
key web design product Dreamweaver. In 2000, Dreamweaver had a market share of 80% among professional web developers in the US. While the company initially focused on the domestic US market, with the global diffusion of the Internet, the market for non-English web development has grown rapidly and the company has responded by increasing its international sales. International sales accounted for 26% of revenues in 1994 but this gure has grown to 44% in 2004. One of the reasons why the product has been adopted so widely worldwide has been product localization (Ramdas, Terwiesch, & Lehmbeck, 2001). As well as English, Dreamweaver is available in Spanish, Japanese, French, German, Italian, Swedish, Portuguese, Korean and Chinese. The company uses three approaches to localization. Full localization involves modifying the product so that it has functionality that matches that of the English version. Of the current versions, European English, French, German and Japanese have been fully localized. Partial localization involves localizing some parts of the product, such as manuals. The third form of localization is to outsource the localization to local vendors for a fee ranging from $80,000 to $200,000. The company supplies a localization kit and instructions and the local vendor is expected to translate the code, convert screenshots, redo the help menus and movies, test the product and translate all documentation and packaging. Incumbent Firms The other four global rms in the sample (Siebel Systems, BEA Systems, Verity and Mercury Interactive) all pre-date the Internet boom but have taken advantage of the new market opportunities offered by the Internet to expand globally. In all four cases, the same factors that have facilitated the globalization of Internet-focused rms, such as unique resource advantages and market dominance, also apply. However, another factor has been the ability to utilize complementary assets (Hu, 1995; Teece, Pisano, & Shuen, 1997) in globalization. Studies in other industries such as biotechnology (Rothaermel, 2001) have demonstrated the importance of complementary assets in the success of incumbents in markets undergoing technological change. The incumbent rms here have been able to leverage their existing competencies and relationships in existing markets to rapidly globalize their Internet products and services. Siebel Systems. As in the case of Cisco, Siebel Systems was one of the early movers in its market. In 1993, when the company was founded by Tom Siebel, the market for integrated back ofce software systems was maturing. Large competitors such as SAP and PeopleSoft had successfully
336
STEPHEN CHEN
created and marketed what became to be known as enterprise resource planning (ERP) systems. These software applications automated processes such as nance, accounting, manufacturing and distribution, allowing companies to link and streamline their internal business processes. In contrast, Siebel Systems focused on automating front ofce sales processes and managing customer information. In 1997, Siebel rst released products that incorporated a web browser interface, which allowed the client to access and navigate through Siebels applications from any standard Web browser. This shift to web-focused applications also broadened the companys international reach. International sales accounted for just 10% of revenues in 1996 but this gure jumped to 27% the following year. Since then the company has increasingly focused on web-related products. This was followed in 1998 with Siebel 98, which incorporated a web browser interface into all of its products. In December 1999, the company launched Siebel 99, a fully web-based version of the Siebel product line. Like Cisco, during this time, the company also launched an initiative called Siebel Everywhere, with the goal of making Siebel applications ubiquitously available so that any device connected to the Internet or a clients Intranet would be able to access Siebels applications. By 2002, Siebel Systems offered the industrys most comprehensive suite of customer relationship management (CRM) applications, enabling organizations to deploy sales, marketing and customer service systems across multiple channels, including the Web, call centers, resellers and dealer networks. BEA Systems. When BEA systems was founded in 1993, distributed computing was well on the way to replacing mainframe computing. However, the founder Bill Coleman realized that many companies still used earlier applications running on different operating systems and that this created the opportunity for building an operating system for the network that would allow effective and efcient interoperability of all these information technology components and processes. These insights formed the seed for the idea of what would eventually become an application server. The companys second product was Tuxedo, a transaction processing monitor bought from networking company Novell, Inc. As part of its strategy to rapidly build an international presence, the company also acquired sales and support organizations in France, South Africa, Finland and Australia. By 1998, it had become clear that the Internet was going to transform the computer industry and that the Java standard was going to be important in this emerging environment. Alfred Chuang, who was one of the co-founders
337
and head of product development at the time, identied a company called WebLogic in San Francisco that was developing an application server to support the emerging Java standards and in spite of initial board opposition, persuaded BEA to buy WebLogic in a stock deal valued at the time at $160 million. By the end of 1998, it was estimated that BEA was the market leader in distributed transaction processing with 51% market share against IBMs estimated 20%. A large proportion of this growth in revenues (42% in 1998) came from international sales. Although the direct sales force provided most of the companys sales (69% in 2003), an important factor in the companys rapid growth worldwide has been its use of indirect sales channels, especially in Asia. Sales partners include systems integrators and consultants such as Accenture, HP, EDS, Cap Gemini, KPMG, Deloitte, Price Waterhouse Coopers and Schlumberger, that included BEA products in their custom solutions for clients; system platform companies such as Sun Microsystems that acted as resellers of BEA products, often integrated with their own products; application service providers, such as System SpA, Digex and eBreviate, that provided the hardware, software and support for businesses that did not wish to buy and maintain the system themselves and, nally, distributors that supplement the efforts of the sales force. The third critical element of BEAs strategy was to establish itself as a platform technology provider. As companies were increasingly using the new e-business technologies to streamline its operations internally and integrate with partners systems, there was an increasing demand for systems based on technology standards that allowed compatibility with other vendors products. BEA responded with the launch in June 2002 of the rst unied application infrastructure product that combined application server, integration server and portal product functionality. Verity. Verity is one of the leading suppliers of content management systems including paper-to-electronic and Web-based document capture, electronic forms and process automation. The company was founded by Mike Pliner, co-founder and former CEO of Silicon Valley network vendor Sytek Inc. He was asked by his friend, Dick Wishner, president of Advanced Decision Systems (ADS), a developer of articial intelligence programs, to review technologies in the R&D lab that could be commercialized and he saw the potential of Topic, a concept-based text-retrieval program, for intelligence gathering. Pliner convinced the Strategic Air Command to become a beta test user. Topic was able to cut report-generation time from 2 days to 2 hours. Following this success, a spin-off company Verity was established
338
STEPHEN CHEN
in 1988 with ADS providing the technology and seed capital and Pliner as CEO. Within a year, the company signed joint marketing agreements with leading database vendors such as Oracle, Relational Technology, Sybase and Informix and also distributors in England, Belgium and France. The company rapidly established itself as a leading provider of software to analyze real-time data such as news feeds from Dow Jones and PR Newswire. In 1993, the company shifted its focus to the fast-growing Internet and intranet marketplace, and its products are now used by more than 600 corporations worldwide. International sales accounted for 29% of revenues in 1994. As in the case of Macromedia, one of the factors behind the companys worldwide growth has been localization of its products. The companys products are available in more than 70 languages, many developed with local software companies such as NEC in Japan and 3Soft in Korea. The company also uses a mix of international sales channels. It has an extensive worldwide sales organization including ofces in the United States, United Kingdom, The Netherlands, Germany, France, Sweden, Singapore, Australia, Japan, Mexico, Brazil and South Africa and it also works with a variety of partners, including value-added resellers, OEM customers, independent software vendors and system integrators. Mercury Interactive. Mercury Interactive was founded in 1989 and initially focused on testing and quality assurance of computer systems. The company targeted large clients worldwide early on and international sales have been signicant from an early stage (26% in 1994). However, as in the case of Siebel Systems and BEA Systems, international sales have grown signicantly following the introduction of web-related products. In 1999, the company shifted its focus to website testing. By using the companys Automated Software Quality products, corporate IT departments, system integrators and independent software vendors can identify software bugs more quickly and efciently than with traditional methods. Two technological drivers can be identied for the companys shift. The rst is the widespread adoption of networks, which often mix and match many types of hardware and software, unlike traditional highly standardized mainframe and minicomputer systems typically supplied and maintained by a single vendor. The second is the widespread adoption of Microsofts Windows, which initiated a new style of programming in which the sequence of user responses is unpredictable. Unlike traditional systems that forced the user to pursue a rigid series of keyboard commands to achieve a given task, such as printing the le, this new style allows much more spontaneity on the part of the user.
339
As a result the complexity and need for systems testing has increased signicantly and the company has beneted from the increased demand worldwide. Today Mercury Interactive is the major player in the market for testing of websites with 45% of the market and supplies over 450 major corporations worldwide including Bell Atlantic, Chemical Bank, General Electric, IBM, Lufthansa, the National Association of Securities Dealers, Siemens and Texaco.
DISCUSSION
The ndings from the assessment of the sample of 63 Internet rms and the eight cases of highly global rms shed some light on the various theoretical arguments about globalization posed earlier in the paper. The Globalization versus Regionalization Debate First, the ndings from the sample of 63 Internet rms show that, as in the case of leading traditional bricks and mortar rms, the extent of globalization of Internet-based rms has been greatly exaggerated. Less than 30% of rms have foreign sales of 25%, even those that have been operating in Internet markets for more than 10 years. The majority of rms are still heavily domestically and regionally based. The exceptions are mainly rms that supply global infrastructure services. Despite some predictions, the evidence shows that Internet-based rms are no more global than other rms, at least based on the extent of foreign sales. This is consistent with the ndings of Rugman (2000) who found that regionalization rather than globalization was more common strategy for rms in the FT500. This studys nding that B2C rms were generally less globalized than B2B rms is consistent with other studies that have identied that factors such as Internet access, consumer familiarity with the Internet, trust in conducting commercial transactions over the Internet, language and culture can be barriers in consumer adoption of electronic commerce. It was mentioned earlier that in compiling the sample of rms it was difcult to nd many examples of search engines, ISPs and information providers that had an a priori global presence, and in fact an initial review indicated that in most cases these rms operated locally. In the case of ISPs, the need to gain access to the local network could clearly be a barrier to globalization, while in the case of search engines and information providers there are clearly linguistic and cultural barriers to overcome.
340
STEPHEN CHEN
The barriers appear to be less signicant in the case of B2B rms. For instance, Openwaves largest international market is not Europe but Japan, which is more culturally distant. Two reasons can be advanced. First, the most global B2B rms often have unique products and brands for which there is less direct competition. Second, business buyers are generally better informed about Internet technology so language and culture are less of a problem. In the case of a few highly globalized rms, other reasons can also be identied, which are discussed below.
The Oligopolistic Reaction/Mimetic Market Entry Argument There was no evidence of mimetic market entry or oligopolistic reaction as motives for internationalizing in the case of the eight most global rms. Seven out of the eight most global rms (Cisco, Openwave, Macromedia, Siebel Systems, BEA Systems, Verity and Mercury Interactive) were early movers in their respective product markets. As such they were able to benet from early mover advantages (Lieberman & Montgomery, 1988; Kerin, et al., 1992) such as reputation building, an established customer base and established distribution networks. Only in the case of Juniper could reaction to a competitor (in this case Cisco) be seen as a possible motive for internationalization. However, the fact that the company has generally adopted a strategy of avoiding direct competition with Cisco would suggest that this is unlikely to have been a key motive for internationalization.
The Staged Internationalization versus Born Global Debate The cases clearly support the born global model rather than the staged internationalization model. All the eight highly global rms examined can be classied as born global rms in the sense that they internationalized from an early stage. There was no evidence that gradual learning about international markets was a signicant consideration in their internationalization strategy. Rather, the cases indicate that the main constraint was the state of development of the Internet worldwide and learning by users. The companies have generally entered countries where the Internet market was growing fastest, regardless of psychic distance, as most clearly demonstrated in the case of Openwave, where Japan is the largest foreign market.
341
The Network Effect Argument There was some evidence that supports the arguments that Internet rms may benet from network advantages. A factor that is clearly signicant in all of the cases is a global technology standard. First, a standard lowers the cost of local product adaptation. For instance, Macromedia and Verity were both able to offer local versions of their software at low cost, by using modular designs that allow the required parts of the product to be adapted at low cost and by outsourcing some of this work to local partners. Second, technology standards also allow greater compatibility of products from different suppliers or different products from the same supplier and allow network externality effects to take place. The cases also highlight the important role played by complementary products and services in enabling the rm to globalize rapidly. Both Cisco and Siebel Systems were able to quickly dominate the market by offering an integrated suite of applications and services; BEA Systems and Verity were able to globalize rapidly by forming partnerships with rms that supplied complementary products and services globally; Openwave was similarly able to globalize rapidly by forming partnerships with leading computing companies that already had a global presence and major mobile communication companies in key markets. This conrms the importance of complementary products in the case of markets for systems-based products (Katz & Shapiro, 1994; Farrell, Monroe, & Saloner, 1998) and suggests that complementary products not only may increase local demand but may also facilitate entry into new markets globally.
Winner Takes All The cases also suggest something else. Another common factor that is evident from the cases is that all of the most global rms are the rst or the second leading players in their respective markets. This suggests a winner takes all effect that has been highlighted by other research in technologyintensive markets (Arthur, 1996; Schilling, 2002; Chen, 2003; Noe & Parker, 2005), in which the leading rms capture a disproportionate share of the market. As Arthur (1996) has argued, technology markets often exhibit this effect as they require a high upfront investment to enter the market but once products catch on in the market, marginal costs of production are relatively low. Combined with rst-mover advantages and network externalities, this
342
STEPHEN CHEN
often leads to a situation where rms that gain a slight lead win an increasing share of the market worldwide.
CONCLUSIONS
In conclusion, this study shows that the Internet has had limited impact on the globalization of most Internet rms, particularly in B2C markets. Barriers to globalization such as local brands, language and culture may be signicant in B2C markets. However, there are some exceptions and a few rms operating in B2B markets are highly globalized. Factors that facilitated their globalization include the presence or establishment of a global standard, rst mover advantages and network effects that created a winner takes all effect that extended to markets globally. In these cases, the constraint on internationalization has not been market learning by the rm but rather learning of the technology by customers in the markets in which the rm operates. The results suggest that the main impact of the Internet on rms in global markets has not been as a global sales or distribution channel for rms, as some have claimed, but as a communications network based on a global standard. The rms that have beneted most as regards internationalization have been rms that supply the components necessary to access and utilize that network. This conrms the importance of considering network externality effects resulting from technology standards and complementary products in theories and practice of internationalization strategy in Internet rms or in rms using the Internet as a component of their globalization strategy. However, it should be added that these effects appear to be signicant only in certain cases, mainly B2B rms, and that the effects seem much less signicant in the case of B2C rms, where local market barriers appear to outweigh any network effects. Admittedly this paper has only examined a small number of rms and the results need to be conrmed in a larger sample or other cases. However, as noted earlier, the rms in the sample examined were leaders in their particular market so a priori they would be expected to show the greatest internationalization activity and show the greatest effects of the Internet in facilitating globalization. The fact that extensive globalization was not observed except in a few cases should raise some questions for further research on internationalization of rms, for example, how do drivers and barriers to globalization vary in different geographic markets, for different types of product or service; how relevant are the ndings for non-Internetbased rms?
343
Finally, it should be noted that one of the problems with studying Internet markets is the relative immaturity and fast-changing nature of the markets so some other caveats should be added. It may be that as Internet technology becomes more widely adopted and as consumers gain greater condence in its use, the importance of the Internet as a global sales channel will increase and global sales of Internet rms will also increase. Therefore, more research needs to be done to determine if, and how, the nature and extent of globalization changes as Internet markets mature.
REFERENCES
Arthur, W. B. (1996). Increasing returns and the new world of business. Harvard Business Review, (July-August), 99109. Baden-Fuller, C. W., & Stopford, J. (1990). Globalization frustrated: The case of white goods. Strategic Management Journal, 12, 493507. Bakos, Y. (2001). The emerging landscape for retail e-commerce. Journal of Economic Perspectives, 15, 6980. Barua, A., Pinnell, J., Shutter, J., & Whinston, A. B. (1999). Measuring the Internet economy: An exploratory study. Available at http://crec.mccombs.utexas.edu/works/articles/ internet_economy.pdf Borenstein, S., & Saloner, G. (2001). Economics and electronic commerce. Journal of Economic Perspectives, 15, 312. Bridges.org Report. (2001). Spanning the digital divide: Understanding and tackling the issues. http://www.bridges.org. Chen, S. (2003). The real value of e-business models. Business Horizons, 46(6), 2733. Dunning, J. H. (1973). The determinants of international production. Oxford Economic Papers New Series, 25(3), 289336. Evans, P., & Wurster, T. (2000). Blown to bits: How the economics of information transforms strategy. Boston, MA: Harvard Business School Press. Farrell, J., Monroe, H. K., & Saloner, G. (1998). The vertical organization of industry: Systems competition versus component competition. Journal of Economics & Management Strategy, 7, 143182. Gandal, N. (2001). The dynamics of competition in the Internet search engine market. International Journal of Industrial Organization, 19(7), 11031117. Guillen, M. (2002). What is the best global strategy for the Internet? Business Horizons, 45(3), 3946. Guillen, M. (2003). Experience, imitation and the sequence of foreign entry: Wholly owned and joint-venture manufacturing by South Korean rms and business groups in China, 19871995. Journal of International Business Studies, 34, 185198. Hamill, J. (1997). The Internet and international marketing. International Marketing Review, 14(5), 300323. Haveman, H. A. (1993). Follow the leader: Mimetic isomorphism and entry into new markets. Administrative Science Quarterly, 38(4), 593627.
344
STEPHEN CHEN
Hu, Y.-S. (1995). The international transferability of the rms advantages. California Management Review, 37(4), 7388. Huang, H., Keser, C., Leland, J., & Shachat, J. (2003). Trust, the Internet and the digital divide. IBM Systems Journal, 42(3), 507518. Hymer, S. H. (1976). The international operations of national rms: A study of direct investment. Cambridge, MA: MIT Press. Johanson, J., & Vahlne, J. E. (1977). The internationalization process of the rm: A model of market knowledge and increasing foreign market commitments. Journal of International Business Studies, 8, 2332. Katz, M. L., & Shapiro, C. (1994). Systems competition and network effects. Journal of Economic Perspectives, 8(2), 93115. Kerin, R. A., Varadarajan, R., & Peterson, R. A. (1992). First-mover advantage: A synthesis, conceptual framework, and research propositions. Journal of Marketing, 56, 3352. Knickerbocker, F. T. (1973). Oligopolistic reaction and multinational enterprise. Boston, MA: Harvard University Press. Knight, G. G., & Cavusgil, S. T. (1996). The born global rm: A challenge to traditional internationalization theory. Advances in International Marketing, 8, 1126. Kotha, S., Rindova, V., & Rothaermel, F. T. (2001). Assets and actions: Firm specic factors in the internationalization of US Internet rms. Journal of International Business Studies, 32(4), 769792. Kshetri, N., & Dholakia, N. (2002). Determinants of the global diffusion of B2B e-commerce. Electronic Markets, 12(2), 120129. Leamer, E. E., & Storper, M. (2001). The economic geography of the Internet age. Journal of International Business Studies, 32(4), 641665. Levitt, T. (1983). The globalization of markets. Harvard Business Review, 61(3), 92102. Lieberman, M. B., & Montgomery, D. B. (1988). First-mover advantages. Strategic Management Journal, 9, 4158. Lituchi, T. R., & Rail, A. (2000). Bed and breakfasts, small inns, and the Internet: The impact of technology on the globalization of small businesses. Journal of International Marketing, 8(2), 8697. Lynch, P. D., & Beck, J. C. (2001). Proles of Internet buyers in 20 countries: Evidence for region-specic strategies. Journal of International Business Studies, 32(4), 725748. Madsen, T. K., Rasmussen, E. S., & Servais, P. (2000). Differences and similarities between born globals and other types of exporters. Advances in International Marketing, 10, 247265. Madsen, T. K., & Servais, P. (1997). The internationalization of born globals: An evolutionary process? International Business Review, 6(6), 561583. Millar, C., Choi, C., & Chen, S. (2005). Globalization rediscovered: The case of creative industries. Management International Review, 45(Special Issue), 121128. Moen, ., & Servais, P. (2002). Born global or gradual global? Examining the export behavior of small and medium-sized enterprises. Journal of International Marketing, 10, 4972. Noe, T., & Parker, G. (2005). Winner take all: Competition, strategy, and the structure of returns in the Internet economy. Journal of Economics and Management Strategy, 14(1), 141154. Oviatt, B. M., & McDougall, P. P. (1994). Toward a theory of international new ventures. Journal of International Business Studies, 25(1), 4564.
345
Petersen, B., Welch, L. S., & Liesch, P. (2002). The Internet and foreign market expansion by rms: Theoretical questions and three predictions. Management International Review, 42(2), 207221. Prahalad, C. K., & Doz, Y. (1987). The multinational mission: Balancing local demands and global vision. New York: Free Press. Ramdas, K., Terwiesch, C., & Lehmbeck, C. (2001). Macromedia Inc. Weaving dreams of global markets and new technologies, case UVA-OM-071. Darden Business Publishing, University of Virginia. Rialp-Criado, A., Rialp-Criado, J., & Knight, G. A. (2002). The phenomenon of international new ventures, global start-ups, and born-globals: What do we know after a decade (19932002) of exhaustive scientic inquiry? Proceedings of the 28th EIBA Conference. Athens, Greece, December, 142. Rothaermel, F. T. (2001). Incumbents advantage through exploiting complementary assets via interrm cooperation. Strategic Management Journal, 22(6/7), 687699. Rugman, A. (2000). The end of globalization: Why global strategy is a myth and how to prot from the realities of regional markets. New York: Amacom. Rugman, A. M., & Brain, C. (2003). Multinational enterprises are regional, not global. Multinational Business Review, 11(1), 312. Rugman, A. M., & Verbeke, A. (2004). A perspective on regional and global strategies of multinational enterprises. Journal of International Business Studies, 35(1), 318. Schilling, M. A. (2002). Technology success and failure in winner-take-all markets: Testing a model of technological lock out. Academy of Management Journal, 45, 387398. Schlie, E., & Yip, G. (2000). Regional follows global: Strategy mixes in the world automotive industry. European Management Journal, 18(4), 343354. Singh, N., & Kundu, S. (2002). Explaining the growth of e-commerce corporations: An extension and application of the eclectic paradigm. Journal of International Business Studies, 33(4), 679697. Sprano, E., & Zakak, A. (2000). E-commerce capable: Competitive advantage for countries in the new economy. Competitiveness Review, 10(2), 114123. Stevenson, A., & Hamill, J. (2002). Internet forum. International Marketing Review, 19(2/3), 323326. Sullivan, D. (1994). Measuring the degree of internationalization of a rm. Journal of International Business Studies, 25(2), 325342. Teece, D., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509533. Vernon, R. (1966). International investment and international trade in the product cycle. Quarterly Journal of Economics, 2, 190207. Yip, G. (2000). Global strategy in the Internet era. Business Strategy Review, 11(4), 114. Zaheer, S. (1975). Overcoming the liability of foreignness. Academy of Management Journal, 38, 341363. Zaheer, S., & Manrakhan, S. (2001). Concentration and dispersion in global industries: Remote access and the location of electronic commerce activities. Journal of International Business Studies, 32(4), 667687. Zaheer, S., & Zaheer, A. (2001). Market microstructure in a global B2B network. Strategic Management Journal, 22(9), 859874.
HOW DO THE INTERNET AND INTERNATIONALIZATION AFFECT THE BUYING CENTER? AN EXPLORATORY CASE STUDY
Catherine N. Axinn, Dawn R. Deeter-Schmelz, Brian T. Straley and Ernest J. Zavoral, Jr.
ABSTRACT
Drawing from seminal research on organizational buying behavior (Johnston, 1979; Johnston & Bonoma, 1981), we use a case study format to explore the impact of the Internet and internationalization on todays industrial procurement processes. Interviews with senior managers of an industrial distributor reveal several key insights regarding the impact of the Internet on buyersupplier interactions and the importance of global sourcing. Based on these exploratory ndings, implications for future research are offered.
1. INTRODUCTION
Research published more than 25 years ago transformed the way in which purchasing was viewed (Johnston, 1979; Johnston & Bonoma, 1981).
347
348
CATHERINE N. AXINN ET AL.
It became the basis for sales training in universities and for selling professionals to determine how and with whom to communicate their sales efforts. However, little business literature has evaluated or built upon the original reported framework. Two dynamic drivers demand researching potential changes in the ways in which businesses purchasing entities communicate and operate: the Internet and internationalization. Because neither factor was dominant at the time of the seminal work in this area (e.g., Cyert, Simon, & Trow, 1956; Johnston, 1979; Johnston & Bonoma, 1981; Robinson, Faris, & Wind, 1967; Sheth, 1973; Webster & Wind, 1972, 1996), the relevance of the established structure, function, and interaction processes concern us here. While current business literature proposes that both factors have had an effect on business, none materially measures the effect on the purchasing behaviors of industrial organizations in the context of the buying center. The primary goal of this research is to combine exploration of the impact of internationalization on industrial sourcing with an assessment of the impact of Internet use on the buying center, specically, and industrial purchasing and procurement processes, in general. A better understanding of the effects of these phenomena may create an updated framework for understanding the ways in which industrial companies of the 21st century are operating. Evaluating the effects of internationalization and the Internet on purchasing behaviors will benet both buyers and sellers by allowing more efcient and effective exchanges due to developing a clearer understanding of who the players involved are and how they are interacting with one another. This research will benet business by creating a basis for evaluating the continuing relevance of the buying center in the current operating environment. Mapping the interaction between purchasers and sellers may provide knowledge allowing sales professionals to better utilize the Internet as a marketing tool. In addition, the research may provide a clearer framework by which to understand the utility of global sourcing.
2. BRIEF LITERATURE REVIEW

We begin by taking a look at the literature related to the classical concepts of the buying center, and the more recent move toward formalized buying teams. This is followed with a brief examination of the literature pertaining to two environmental forces increasingly affecting the organizational buying process: (1) the Internet, and (2) internationalization and global sourcing.
How Do the Internet and Internationalization Affect the Buying Center?
349
2.1. The Buying Center Research conceptualizing the buying center as a network of connected individuals spans almost 40 years (e.g., Bristor, 1993; Buckles & Ronchetto, 1996; Cyert et al., 1956; Johnston & Bonoma, 1981; Robinson et al., 1967; Sheth, 1973; Venkatesh, Kohli, & Zaltman, 1995; Webster & Wind, 1972; Weigand, 1968). Beginning with the seminal work of Cyert et al. (1956), Robinson et al. (1967), Webster and Wind (1972, 1996), and Sheth (1973), organizational buying theory has characterized the buying function as a complex process involving multiple individuals rather than the single act of one person. Since this early work was conducted, most literature has focused on the decision-making processes and the methodologies have focused on a specic buying decision (Sheth, 1996). Among the earliest studies to empirically test the buying center concept, Johnston (1979), in his dissertation, and later Johnston and Bonoma (1981) provided evidence of the buying center structure that has received little subsequent attention. As prescribed by organizational buying theory, the authors adopted a communication network approach to study the buying decision process, with the goal of identifying the structure and interaction patterns occurring in the buying center. Johnston and Bonoma (1981) identied ve buying center dimensions: (1) vertical involvement, i.e., the number of levels in the organizations hierarchy exerting inuence in buying center decisions; (2) lateral involvement, i.e., the number of different departments and divisions involved in buying center decisions; (3) extensivity, i.e., the number of people involved in the buying network; (4) connectedness, i.e., the strength of the linkages between members of the buying center; and (5) the centrality of the purchasing manager in the buying center network. In addition, the authors explored ve structural variables: (1) centralization, i.e., the degree to which authority, responsibility, and power are concentrated in the buying center; (2) formalization, i.e., the extent to which buying center activity is formally dened by rules, policies, and procedures; (3) complexity, i.e., the extent to which the company pursues functional specialization; (4) organization size; and (5) participation, i.e., the level of buying center member involvement in the decision-making process. Among the key ndings reported by Johnston (1979) and Johnston and Bonoma (1981) was that the extensivity of the buying center was affected by the degree of formalization, the complexity and importance of the purchase decision, and the purchase class (industrial versus service). Lateral involvement was affected by formalization along with the importance and novelty
350
of the purchase situation. Vertical involvement was most heavily inuenced by purchase class, importance, and complexity, while connectedness was affected by formalization and centralization. In fact, formalization and the importance of the purchase situation had the greatest impact on the buying center dimensions, whereas organization size had no effect. Over the 25 years since the seminal work was published, there has been much research about industrial organization buying behavior, but little has challenged the structure outlined by Johnston (1979) and Johnston and Bonoma (1981). Yet a shift in the general way that marketers are selling to industrial customers has changed with a shift to understanding and inuencing these customers (Sheth, 1996), suggesting that additional research is warranted. As noted by Sheth (1996), in his review of the state of organizational buying research, the business literature proposes that organizational buying behaviors have been changing since the 1970s due to globalization, total quality management (TQM) philosophies, industry consolidation through mergers and acquisitions, and the use of information technologies. Because of global mergers and acquisitions and alliance activities, procurement has shifted from a decentralized administrative function to a centralized strategic function (Sheth, 1996). Organizational restructuring and outsourcing are affecting organizational buying behavior, which in turn affects the decisions made (Lewin & Johnston, 1996). The use of technology has also affected the buying center with electronic data interchange (EDI) and computer networks, which in turn have restructured the procurement philosophy, processes, and platforms (Dadzie, Johnston, Dadzie, & Yoo, 1999; Dzever, Quester, & Chetty, 2001; Stump & Sriram, 1997). Research has also shown that the change in organizational buying behaviors shifts it from a transaction-centered and domestic practice to a relational-centered and global practice (Sheth, 1996; Wilson, 1996). Given these environmental changes affecting the buying function, it seems appropriate to revisit Johnston and Bonomas (1981) early work, updated for todays buying organizations. Thus a primary goal of this case study is to evaluate the potential for replicating their earlier research in todays procurement environment.
2.2. Team Buying The more a company transitions to a relational-centered and global procurement process, the more it will have to create cross-functional teams dedicated to suppliers, which will replace the buying center structure and
351
process (Sheth, 1996). Team buying is, essentially, an extension of the buying center concept. Dened as the process of making purchasing decisions through cross-functional experts cooperating to quickly and effectively make a purchase (Deeter-Schmelz & Ramsey, 1995), team buying reects a higher level of formalization as compared to the traditional buying center model. Current business literature suggests that unlike the buying center with its uid membership, members in formalized buying team structures have specic functions and roles and can be clearly identied through their team membership (Deeter-Schmelz & Ramsey, 1995). Buying team members typically have a functional interdependence on one another based on cross-functional expertise (Blau, 1972). When a need surfaces, the appropriate team is activated for resolution of the problem. Although a core purchasing team exists, other organizational areas may provide members that support the function (Deeter-Schmelz & Ramsey, 1995). Because purchasing teams allow additional personnel to participate in a purchasing decision that otherwise would not have been included, a more strategic decision process is created (Ellram & Pearson, 1993). The goal of working in a team setting for making purchasing decisions is to coordinate efforts to best match the specications of different departments that are affected. In order for the buying team to be successful, it must align its goals with the overall goals of the organization (Johnston & Bonoma, 1981). Team involvement with purchasing most commonly takes place in situations involving problem solving with suppliers, supplier selection, and decisions to make or buy, and facilitates improved communication, awareness, and interaction between different functional departments of an organization (Ellram & Pearson, 1993).
2.3. Internet The impact of the Internet on purchasing activities has changed substantially over the past 25 years. During this time, the Internet has become a signicant inuencer of many business initiatives. Companies are currently taking advantage of the ease in communicating and attaining information at growing speeds. The Internet is radically changing the communication among those involved in purchasing. Leading executives are focusing their attention on business-to-business Internet ventures, reverse auctions, and intranets (Croom, 2000; Deeter-Schmelz & Kennedy, 2002; Essig & Arnold, 2001; Talluri & Ragatz, 2004; Wenninger, 1999); and experts have predicted that by the year 2010, the Internet will be the backbone of electronic
352
industrial purchasing (Carter, Carter, Monczka, Slaight, & Swan, 2000). As many companies are focusing their efforts on creating relationships with foreign suppliers and re-evaluating their budgeting of promotional dollars, understanding the role of the Internet is critical to the effective allocation of time and effort by international marketers. Early on, the Internet was perceived to be an impersonal form of communication. It was used only as a channel for exchanging information both commercially and noncommercially. Commercial use includes interactive websites, while noncommercial use comes in the form of newsletters, bulletin boards, and list serves. One of the rst formal analyses of the use of the Internet as a communication device revealed it to be extremely versatile on the dimensions of symmetry of information ow, media content, diversity of information sources, communication timing, and personal activity (Hoffman & Novak, 1996). More recent use of the Internet has shown it to be helpful as a conduit for personal communication and as a source of information for organizational buyers. A study reported by Deeter-Schmelz and Kennedy (2002) reveals the importance of gathering and sharing information through the use of e-mail. In particular, research suggests that the Internet is used regularly for communication and for gathering information on suppliers and customers (Kennedy & Deeter-Schmelz, 2001). The Internet has been used to compile competitor information, gather supplier information, monitor, and maintain inventory levels through materials requirements planning (MRP) and just-in-time (JIT) acquisition processes, develop customer relationship management processes, and train employees. One can imagine that the Internet might be more useful during the prepurchase information stage. Likewise, buyers might be more likely to engage in online purchasing under the conditions of a straight rebuy (Deeter-Schmelz & Kennedy, 2002), although it is worth noting that current research suggests industrial buyers are using the Internet more frequently as an information-gathering tool than as a transaction (i.e., purchasing) tool (Kennedy & Deeter-Schmelz, 2001). One factor inuencing the use of the Internet is experience. Researchers have argued that those buyers lacking the skills to use the Internet are less likely to do so (Kennedy & Deeter-Schmelz, 2001; Novak, Hoffman, & Yung, 2000). As industrial buyers search for a more efcient channel for carrying out the various functions involved in a purchase, it is inevitable that the Internet will come into play. At this point, those involved must develop a sense of awareness and a knowledge base that will facilitate easier and, therefore, greater use of the Internet.
353
Not only is the Internet disrespectful of national boundaries and national jurisdiction, it renders geography irrelevant; in online transactions, the location of the parties may be unknown (Grant & Bakhru, 2004). As the Internet becomes more prevalent in purchasing situations, those involved will have to re-examine its function to determine any changes that may occur during the purchasing process.
2.4. Internationalization and Global Sourcing The dominant change in the world economy since Johnstons original research is the globalization of markets (Levitt, 1984; Yip, 1992; Bartlett & Ghoshal, 2000). As markets have globalized, rms have internationalized. Some scholars have posited gradual models for internationalizing both the marketing activities of rms (Johanson & Vahlne, 1977, 1990) and the global sourcing activities of rms (Monczka & Trent, 1991), while others have offered a more interactive view of both the inward and outward internationalization processes, which could speed up both processes (Welch & Luostarinen, 1993). In addition, it seems that the pace and extent of globalization have been fairly swift. According to Trent and Monczka (2002), the percentage of U.S. rms making purchases internationally grew dramatically from 21% in 1975, to 56% in 1982, to 71% in 1987. Trent and Monczka (2002, p. 69) further assert that:
The average amount of total purchases from non-U.S. sources by larger U.S. rms has increased from 9 percent of total annual dollar purchases in 1993 to over 25 percent in 2000.
European research reinforces the perception that purchasing has become highly internationalized. Research by Servais and Jensen (2001) revealed that less than 15% of the small Danish manufacturers who were studied had no foreign suppliers, while over 25% spent more than half of their procurement budgets with foreign suppliers. Although denitions of global sourcing vary, most authors agree that there is a signicant strategic component to global sourcing that is absent in international purchasing. Based on their development of a ve-level model of internationalizing procurement, Trent and Monczka (2002, p. 68) posit the following:
Global sourcing, which differs from international purchasing in scope and complexity, involves proactively aggregating volumes and coordinating common items, practices,
354
processes, designs, technologies, and suppliers across world-wide procurement, design, and operating locations.
Murray, Kotabe, and Wildt (1995, p. 181), in conducting their contingency analysis of the performance implications of global sourcing strategy, dene global sourcing as involving,
setting up production operations in different countries to serve various markets, or buying and assembling components, parts or nished products worldwide.
Kohn (1993, p. 18) of A.T. Kearney, Inc. denes global sourcing in terms of both what it is, and what it is not:
Primarily, global sourcing is re-examining purchasing strategies for the majority of a companys purchased material cost base to determine whether the sourcing should change in light of global supply economics and a global supply base. y Global sourcing is not an exercise in nding cheap sources of supply or suppliers of questionable quality y 50 to 60 percent of savings in global sourcing programs are the result of existing suppliers improving their price to world-class levels.
Motives for global sourcing can be as varied as denitions. According to Murray et al. (1995, p. 182),
Improved sales and nancial performance is a major objective of a global sourcing strategy.
Their analyses reveal that a variety of sourcing methods may be effective, depending on the product being sourced, and that sourcing strategies may need to change over time to reect changes in the dynamics of supply markets. Alguire, Frear, and Metcalf (1994, p. 73) examine the determinants of global sourcing strategy and conclude that
global sourcing strategies should be driven at least as much by technology and quality considerations as by cost concerns.
This study examines both motives for global sourcing and barriers to global sourcing. Factor analysis reveals two major motives (competitive advantage and comparative advantage) and two major types of barriers (internal and external). Fagan (1991, p. 21) acknowledges the initial role of costs:
Traditionally, the most widely recognized benet of global sourcing has been lower costs. Less-expensive labor, less restrictive work rules, and lower land and facility costs have enticed companies to foreign suppliers; reduced product cost remains the main attraction for perhaps one third to one half of those companies currently pursuing global sourcing.
355
However, he goes on to detail additional benets from global sourcing, including availability of supply, uniqueness of products, and quality and technical supremacy. Servais and Jensen (2001) compared motives for choosing foreign suppliers to those for choosing domestic suppliers and found lower prices to be the primary factor for both sources of supply, although the secondary factors varied with better quality being a motive for selecting foreign suppliers and better lead time being a motive for selecting domestic suppliers. Trent and Monczkas (2002) research examined factors contributing to more versus less successful applications of global sourcing. The top three factors found to indicate greater success were: (1) following a well-dened process or approach to global sourcing, (2) involving the right individuals as participants or team members, and (3) having well-established communication methods among participants. Partly because communication is so important in successful global sourcing, the Internet has become increasingly vital. Electronic commerce is considered to represent the ultimate manifestation of the globalization of business (Grant & Bakhru, 2004), and experts predict that within the next decade the Internet will become the interface of choice for supply chain management, including the procurement process (Carter et al., 2000). The international face of e-commerce reveals that national markets have only partly succumbed to the onward march of globalization. While some sectors are dominated by multinational players, most of the major businessto-business e-commerce hubs are dominated by global players (Grant & Bakhru, 2004).
3. EXPLORATORY CASE STUDY

Clearly, the industrial marketplace is undergoing tremendous changes, in part due to the effects of the Internet and internationalization. This paper reports the rst of three exploratory case studies. As mentioned previously, a key goal was to evaluate the potential for replicating the research originally conducted by Johnston (1979) in order to ascertain how the Internet and global sourcing might affect the communication relationships inherent in, and structure of, the buying center. Case study method was deemed the most appropriate technique to achieve this goal, since the original research essentially comprised an analysis across 31 case studies. Therefore, we modied the instruments used by Johnston (1979) in his seminal research, and to this framework the questions necessary to facilitate our expanded
356
purpose were added. Some of the questions regarding Internet use were drawn from Deeter-Schmelz and Kennedy (2002). The subject rm in this case study is an industrial distributor. Four industrial distributors had been among the original 31 rms which formed the sample for Johnstons (1979) earlier research. Our subject rm, headquartered in the Midwestern United States, is a major player in the global automotive aftermarket, supplying equipment to mechanics in a wide variety of industries. Interviewees were all the senior managers who were identied as playing a role in the organizations purchasing and procurement processes. These managers included the CEO, a representative of the parent company, the Global Procurement Leader, who is permanently assigned to work with the subject rm, and ve other managers. We were very fortunate to have the complete support of the CEO, who stated in the opening interview:
We will help you in any way we can. We will give you access to as many people as you need access to. y I think that the world has changed radically. Not just globally because of globalization, but because of Internet communication and travel.
4. FINDINGS
4.1. Shaped by the Subject Firm At the close of the same interview, the CEO cautioned:
I ask you to really think about the bigger thing, which is purchasing in multi-divisional environments, global multidivisional environments, what does the organization look like and how can it be more efcient? y I encourage to you keep an open mind.
This comment turned out to be somewhat prophetic, as we soon found ourselves unable to t this organizations purchasing behavior into the frameworks of our carefully designed surveys. This may, in part, be due to the multidivisional structure of the organization, and it may be due to the specic business model of the rm as a distributor. Because of the distributor business model, we spoke with numerous people involved in procuring products critical to its business operations, but had difculty identifying people who had made purchases like those studied by Johnston (1979) in his original research (e.g., capital equipment and industrial services). In the end, we were able to interview one manager who played a principal role in the purchase of a piece of capital equipment, but were unable to build from that point a map of all the communication relevant to the buying center for
357
that purchase. However, we were still able to uncover some aspects of how the Internet and internationalization/global sourcing affect the purchasing practices of this rm, as discussed below.
4.2. Internet Impact on Interactions with Suppliers To begin, we asked each manager to rate the importance of the Internet, as shown in Table 1. Interestingly, their rating scores varied widely, and were often accompanied with various comments. For instance, the CFO noted that there was a tendency to rely more on EDI than online purchasing. He emphasized the importance of direct system-to-system connections, passing orders back and forth. The Director of Merchandising noted that the company was only a few months away from launching a vendor-oriented webpage. He said, y all the new products will come through this webpage. The parent companys Global Procurement Leader reported not using the Internet in his purchasing activities, due in part to his more strategic role in the rm. At the other end of the spectrum, not surprisingly, is the Director of Information Technology, who said that with respect to the purchase of a very large and specialized printer (capital equipment purchase), we use the Internet heavily to compare features and benets. To gain a more detailed view of the respondents Internet-based behaviors, each (except the CEO) was given a short survey drawn from the work of Deeter-Schmelz and Kennedy (2002). The results of the two questions posed in this survey are compared with the results obtained in the previous study in Tables 2 and 3, below. With respect to the Internet use (Table 2), it is not surprising that the respondents in our sample report Table 1.
Respondent Role
Importance of Internet in BuyerSupplier Relationship.

On a Scale of 110, about How Important is the Internet in Your Relationships with Suppliers? 8.5 5 Now it is 2, but soon it will be 8 7 6 8 No use
CEO CFO Director of Merchandising Director of Supply Chain and Inventory Category Manager Director of Information Technology Parent Companys Global Procurement Leader
358
Table 2.
Comparison with Findings from Deeter-Schmelz and Kennedy (2002): Frequency of Internet Use.
Deeter-Schmelz and Kennedy Mean Case Study Mean
Question: Please Rate the Frequency with Which You Use the Internet in Each of the Following Ways:(1 Never, 5 Always) E-mail Searching for new suppliers Electronic data interchange Online customer support Online ordering Online order status checks Online payments Conducting reverse auctions Just-in-time inventory planning Discussion groups with other customers Gathering information regarding current suppliers Gathering product/component information Gathering competitive information for your company Accessing supplier documents (specs, order policies, etc.) Providing information to suppliers (specs, order policies, etc.) Gathering external customer information for your company
3.96 3.24 2.21 2.29 2.47 2.45 1.41 1.18 1.51 1.55 2.94 3.28 2.56 2.37 2.61 2.32
4.20 2.20 3.40 3.00 1.80 3.00 2.60 2.40 2.00 1.40 3.40 3.00 4.00 2.60 2.60 2.60
greater e-mail use, given that availability of the Internet technology has increased over time. Our respondents also indicate greater use of the Internet for EDI, online customer support, and online order status checks in payments. Our respondents also report more use of the Internet for reverse auctions, just-in-time inventory planning, and gathering information on suppliers as well as competitive information for the rm. Although the signicance of these ndings is not clear, given our small sample size, the results do provide preliminary evidence that organizational buyers may be using the Internet with greater frequency, and for a wider range of activities. Interestingly, our respondents reported less use of the Internet for online ordering as compared to the results reported by Deeter-Schmelz and Kennedy (2002), i.e., 1.80 versus 2.47, respectively.
359
Table 3.
Comparison with Findings from Deeter-Schmelz and Kennedy (2002): Importance of Internet Attributes.
Deeter-Schmelz and Kennedy Mean Case Study Mean
Question: How Important Are the Following Practices for Your Job: (1 Not Important, 5 Very Important) Easy online ordering Ability to ask questions online Easy movement around Internet sites Availability of current information Increasing speed of information from suppliers Increasing speed of information to suppliers Ability to customize content of supplier Internet sites Ability to obtain a lower price for products purchased Reducing order processing time Reducing paper ow Ability to connect to a live person when using an Internet site Ability to compare prices from several suppliers easily Ability to compare products from several suppliers easily Ability to offer feedback regarding Internet sites Ability to obtain information that educates me on product uses Ability to exchange information with colleagues
3.92 3.78 4.31 4.44 4.30 4.04 3.22 4.09 4.31 4.34 3.21 3.92 3.87 2.93 3.71 3.40
2.40 2.80 4.20 4.00 4.20 4.20 3.00 3.60 3.80 3.80 2.60 3.80 3.80 2.60 3.60 3.60
Table 3 reveals the ndings relative to the importance of Internet attributes. As shown, respondents in the Deeter-Schmelz and Kennedy (2002) study placed greater importance on easy online ordering and the ability to ask questions online. Likewise, the Deeter-Schmelz and Kennedy (2002) sample placed greater importance on the ability to obtain lower prices for products purchased online, reduced order processing times and paper ows, and the ability to connect to a live person when using an Internet site. The remaining ndings seem relatively comparable. The only attribute on which respondents in our sample placed greater importance than the Deeter-Schmelz and Kennedy (2002) sample was increasing the speed of
360
information to suppliers. Still, results from both studies suggest this attribute is of importance to industrial buyers (4.04 and 4.20 for the two samples). 4.3. Internationalization and the Impact of Global Sourcing Once again, each manager was asked to rate the importance of global sourcing to the rm. There was much more agreement across all relevant managers on this topic than there was on the topic of the Internet, as can be seen in Table 4. Through the managers comments on the rms global sourcing activities, we learned that there were sourcing operations in Shenzhen and Shanghai, in addition to Taiwan, while sales operations exist in Canada, the United Kingdom, and Japan. According to the CEO, We probably do 70% of our business in the U.S. and 30% in the others. We also learned that the parent companys global sourcing activities were broader than this subsidiarys activities. According to the parent companys Global Procurement Leader, Purchasing to us is tactical, procurement and sourcing is strategic. So we are looking at more of a strategic global standpoint y through utilizing preferred suppliers, qualied suppliers, and basically consolidating our volume where we can. y Our global team consists of 16 or 18 people. One theme that was quite recurrent was the idea that global sources provided high-quality, innovative products, rather than just low-cost products. These comments were typical: You are starting to see innovation overseas. y Right now a lot of the ideas are coming out of Asia and from Europe. (Director of Merchandising). Table 4.
Respondent Role
The Importance of Global Sourcing.

On a Scale of 110, about How Important Is Global Sourcing to Your Firm? 9 7 10 8 7 8
CEO CFO Director of Merchandising Director of Supply Chain & Inventory Category Manager Parent Companys Global Procurement Leader
361
China is a big source of products right now. Taiwan is a big source of product for us. y India tends to be a place where you get more y technologically advanced products. y Probably 10 years ago, people viewed Asian source products as lower quality y over the years, the quality has denitely improvedya lot of times now their standard and quality of work coming out of Asia is changing dramatically. (CFO). It used to be, you buy foreign, you [would] sacrice something. y The world has changed y in many cases there is a lot more innovation coming from overseas than in our own back yard. (Category Manager). On the other hand, several people shared their concern that customers expected their products to be Made in the USA. There are many cases y [where] our customers insist on made in the USA y (Category Manager). You have a huge population of people that are very proud of American made products and are willing to pay for American made products. (CFO). However, as noted by the Category Manager, The customer really does not have a problem with offshore products, as long as theyre sharing in the savings. And we werent looking at it quite that way when we started. y they really dont care where it comes from, as long as it stays in one piece and [we] stand behind it with a warranty. And, the Director of Supply Chain Management observed that, y global sourcing is really key to us y it has a competitive advantage. Thus it appears that the sourcing of products sold by this industrial distributor will continue to be a blend of products Made in the USA and products, especially high-quality, innovative products, made throughout the world. Throughout this discussion we have heard echoes of points made earlier in the literature review. Global sourcing has grown in strategic importance. In this rm the physical manifestation of this is the Global Procurement Leader who is a corporate employee, but is permanently stationed at the subsidiary to assist its global procurement efforts. In addition, we also hear repeated emphasis on the quality and technology benets of global sourcing, in addition to the cost benets. 4.4. Learning about the Buying Center Although our original goal was to replicate Johnstons (1979) seminal research on the structure and function of the buying center, taking into
362
account the Internet and internationalization, this proved impossible, as noted in Section 4.1, above. However, we were able to glean some interesting insights into the aspect of formalization of the purchasing process. Table 5 reports our ndings on this topic. The range of answers on this question is quite startling. One aspect on which there was consensus was the existence of a procedure called Delegation of Authority. This process mandates how much money each person can spend on a purchase without higher approval. According to the CFO, We have a delegation of authority that specically lays out, based on type of purchase y where people have different approval levels. y A direct manager can approve something $2500 or less y but anything over $2500 comes to me for $2500 to $50,000 and then goes to [the CEO] for $50,000 to $250,000. And above $250,000, one single expense over $250,000, we have to get corporate approval. The most diverse views were held on the perceived presence or absence of a formal purchasing manual. Three managers reported awareness of a paper manual while two said there was none. Online availability was also in doubt, although this was the point at which all respondents mentioned the delegation of authority. An unexpected result was comments made regarding the positive impact of the Sarbanes-Oxley legislation in formalizing the purchasing processes. According to the CFO, y you are forced to create processes and procedures for everything. Last year we went through pretty extensive exercises y The Director of Merchandising expanded on this point: A lot of people tell you its been miserable, but the reality of it is it forced [you] y to document how you currently do everything y the big eye opener was you would see everything mapped out that weve been doing for years and suddenly go, Boy, that doesnt really make sense when you see it laid out on Table 5.
Respondent Role
Formalization of Purchasing Policies and Procedures.

On a Scale of 110, about How Formalized Would You Say Your Companys Purchasing Policies/Procedures Are in Comparison to Other Companies? 7 9.5 5 4.5 5 7
CEO CFO Director of Merchandising Director of Supply Chain and Inventory Category Manager Parent Companys Global Procurement Leader
363
paper like that. So there have probably been more changes in the last two years y youll look back in ten years and say, That Sarbanes-Oxley changed corporate America! more than anybody even realizes right now. Because its forcing companies to look at how they operate, there should be huge improvement in efciency y
5. LIMITATIONS, INSIGHTS AND FUTURE RESEARCH

5.1. Limitations in Researching the Buying Center As mentioned earlier, a key goal of this study was to evaluate the potential for replicating Johnstons (1979) seminal research in the context of todays industrial procurement environment. We were frustrated in this endeavor in several ways, and subsequent case studies involving a nationally recognized service rm and a Fortune 200 global manufacturer have provided some additional explanatory insight. One key tool used to construct diagrams of buying centers was a question asking respondents to identify other members of the buying center and quantify their communications with them, specifying the percentage of communication that was oral or written. In our attempt to examine how the Internet inuences the buying center, we expanded this question to include communication via the Internet, e-mail, and company intranets. This expansion rendered this question nearly impossible for the respondents to answer, due to both increased complexity and the sheer volume of communications implied. While on one hand, this provides some evidence that the Internet does indeed impact the buying center; on the other hand, it leaves us seeking alternative means of assessing this impact. Additional limitations stem, in part, from our fairly narrow goals for this research. Because of our focus on the buying center, we did not broadly explore the internationalization of the subject rm. We did not examine their outward internationalization nor did we explore possible connections between their inward and outward internationalization. In addition, we did not consider the impact of cultural distance on the respondents, although some culturally related biases may be seen in some comments. Further, we did not consider the network of potential supplier/buyer relations or the possible existence of reciprocity. Finally, we are quite limited in the conclusions we can meaningfully draw since we only examined one industrial distributor. However, there are some insights that came from this work. These are discussed below.
364
5.2. Insights With respect to our goal of exploring the effects of the Internet and internationalization, focused on global sourcing, on the Johnston and Bonoma (1981) model, we did gain some insights. Regarding the effects of the Internet, we learned that our sample rm is using the Internet more for EDI purposes rather than online purchasing, thereby suggesting that the Internet is being used by this rm more frequently in straight rebuy situations. In addition, as suggested by the Director of Information Technology, the sample rm is using the Internet as an information-gathering tool, collecting information on the features and benets of various products. Finally, as noted by the Director of Merchandising, the rm is turning to the Internet as a way to provide product information to its customers. Interestingly, these ndings are similar to those reported by Kennedy and Deeter-Schmelz (2001) and Deeter-Schmelz and Kennedy (2002), who found that industrial buyers were more likely to use the Internet as an information-gathering tool as well as a purchasing tool in straight rebuy situations. One key difference is the emphasis our sample rm placed on EDI via the Internet. We also gained insight regarding the effects of internationalization, and global sourcing in particular. Several themes became apparent. First, our sample rm was using a wide variety of global sources when purchasing products. Second, global sources were associated with innovative, quality products, rather than just cost savings. Third, there is evidence that the global sourcing activities of this rm are increasingly strategic, although they have not become as centrally controlled as might be expected by Trent and Monczka (2002) to be categorized as a Level V organization (having integration and coordination of global sourcing strategies with other functional groups). Indeed, in this organization we nd that procurement activities are quite diffuse among senior managers. This may be attributed, in part, to the customer-driven distributor business model prevailing in the rm. However, it is possible that such diffuse procurement responsibility could be associated with a responsibility-based budgeting process, and/or the shift to atter organizational structures which are becoming common in the 21st century.
5.3. Suggestions for Future Research The insights discussed above have implications for future research. We believe that longitudinal studies are needed to continually monitor changes in
365
the usefulness of the Internet to industrial buyers and sellers, and to examine its effects on the purchasing process. This might be especially fruitful if done from the perspective of buyer/supplier networks, in the context of examining both inward and outward internationalization. One construct that researchers might investigate is formalization. Formalization was identied as a key inuencing variable by Johnston (1979) and Johnston and Bonoma (1981), affecting extensivity, lateral involvement, and connectedness. Likewise, our sample rm reported high levels of formalization. As organizations move toward more formalized buying teams (Sheth, 1996), as well as more formalized selling teams, it seems appropriate to investigate the effects of formalization in greater detail. Will the advent of the Internet and internationalization lead to more formalization? Does that formalization add necessary structure to the procurement process, or does it inhibit organizational processes? Additional research is needed to explore these ideas more fully. Our research also revealed something of a paradox, with some trends pushing rms to place procurement responsibility in the hands of more individuals (in parallel with their operational responsibilities) and other trends pushing procurement responsibility to a more centralized unit to facilitate true global sourcing. Further research will be needed to resolve this apparent paradox. Finally, it is quite likely that categorizing the purchases made by industrial buyers may be helpful in actually mapping the purchasing process. Bunn (1993) identied buying activities (search for information, use of analysis techniques, proactive focusing, and procedural control), buying situational characteristics (purchase importance, task uncertainty, extensiveness of choice set, and perceived buyer power), and buying decision approaches (casual purchase, routine low priority, simple modied rebuy, judgmental new task, complex modied rebuy, and strategic new task). Just as Robinson et al. (1967) proposed that the buying center would vary by purchase situation it seems likely that the effects of the Internet and internationalization might vary depending on the buying activity, situational characteristics, and buying decision approaches. Researchers might use Bunns (1993) taxonomy as a starting point for empirical investigations of effects of the Internet and internationalization on the purchase decision process.
ACKNOWLEDGEMENTS
The authors would like to thank the Sales Centre at Ohio University for supporting this research.
366
REFERENCES
Alguire, M. S., Frear, C. R., & Metcalf, L. E. (1994). An examination of the determinants of global sourcing strategy. Journal of Business & Industrial Marketing, 9(2), 6274. Bartlett, C. A., & Ghoshal, S. (2000). Transnational management. Boston: Irwin McGraw-Hill. Blau, P. M. (1972). Interdependence and hierarchy in organizations. Social Science Research, 1, 124. Bristor, J. M. (1993). Inuence strategies in organizational buying: The importance of connections to the right people in the right places. Journal of Business-to-Business Marketing, 1, 6398. Buckles, T. A., & Ronchetto, J. R., Jr. (1996). Examining an industrial buyers purchasing linkages: A network model and analysis of organizational buying workow. Journal of Business & Industrial Marketing, 11, 7492. Bunn, M. D. (1993). Taxonomy of buying decision approaches. Journal of Marketing, 57, 3856. Carter, P. L., Carter, J. R., Monczka, R. M., Slaight, T. H., & Swan, A. J. (2000). The future of purchasing and supply: A ten-year forecast. Journal of Supply Chain Management, 36, 1426. Croom, S. R. (2000). The impact of web-based procurement on the management of operating resources supply. Journal of Supply Chain Management, 36, 413. Cyert, R. M., Simon, H. A., & Trow, D. B. (1956). Observation of a business decision. Journal of Business, 29, 237248. Dadzie, K. Q., Johnston, W. J., Dadzie, E. W., & Yoo, B. (1999). Inuence in the organizational buying center and logistics automation technology adoption. Journal of Business & Industrial Marketing, 14, 433444. Deeter-Schmelz, D. R., & Kennedy, K. N. (2002). An exploratory study of the Internet as an industrial communication tool examining buyers perceptions. Industrial Marketing Management, 31, 145154. Deeter-Schmelz, D. R., & Ramsey, R. (1995). A conceptualization of the functions and roles of formalized selling and buying teams. Journal of Personal Selling & Sales Management, 15, 4760. Dzever, S., Quester, P. G., & Chetty, S. (2001). Factors affecting industrial procurement decisions in the Asia-Pacic region: A comparative study of Australia, New Zealand, Thailand and China. Journal of the Asia Pacic Economy, 6, 194211. Ellram, L. M., & Pearson, J. N. (1993). The role of the purchasing function: Toward team participation. International Journal of Purchasing and Materials Management, 4 (Summer), 39. Essig, M., & Arnold, U. (2001). Electronic procurement in supply chain management: An information economics-based analysis of electronic markets. Journal of Supply Chain Management, 37, 4349. Fagan, M. L. (1991). A guide to global sourcing. The Journal of Business Strategy, 12(March/ April), 2125. Grant, R. M., & Bakhru, A. (2004). The limitations of internationalisation in e-commerce. European Business Journal, 16(June), 95104. Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated environments: Conceptual foundations. Journal of Marketing, 60, 5068.
367
Johanson, J., & Vahlne, J. E. (1977). The internationalization process of the rm a model of knowledge development and increasing foreign market commitment. Journal of International Business Studies, 8(1), 2332. Johanson, J., & Vahlne, J. E. (1990). The mechanism of internationalization. International Marketing Review, 7(4), 1124. Johnston, W. J. (1979). Communication networks and inuence patterns in industrial buying behavior. Unpublished Ph.D. dissertation, University of Pittsburgh, Pittsburgh. Johnston, W. J., & Bonoma, T. V. (1981). The buying center: Structure and interaction patterns. Journal of Marketing, 45, 144156. Kennedy, K. N., & Deeter-Schmelz, D. R. (2001). Descriptive and predictive analyses of industrial buyers use of online information for purchasing. Journal of Personal Selling & Sales Management, 21, 279290. Kohn, L. F. (1993). Global sourcing: Broadening your supply horizons. Business Forum, 18(Winter/Spring), 1720. Levitt, T. (1984). The globalization of markets. Harvard Business Review, 61(2), 92102. Lewin, J. E., & Johnston, W. J. (1996). The effects of organizational restructuring on industrial buying behavior: 1990 and beyond. Journal of Business & Industrial Marketing, 11, 93111. Monczka, R. M., & Trent, R. J. (1991). Global sourcing: A development approach. International Journal of Purchasing and Materials Management, 27(2), 28. Murray, J. Y., Kotabe, M., & Wildt, A. R. (1995). Strategic and nancial performance implications of global sourcing strategy: A contingency analysis. Journal of International Business Studies, 73, 181202. Novak, T. P., Hoffman, D. L., & Yung, Y. (2000). Measuring the customer experience in online environments: A structural modeling approach. Marketing Science, 19, 2242. Robinson, P. J., Faris, C. W., & Wind, Y. (1967). Industrial buying and creative marketing. Boston: Allyn & Bacon. Servais, P., & Jensen, J. M. (2001). The internationalization of industrial purchasing: The example of small Danish manufacturers. Advances in International Marketing, 11, 227254. Sheth, J. N. (1973). A model of industrial buyer behavior. Journal of Marketing, 37, 5056. Sheth, J. N. (1996). Organizational buying behavior: Past performance and future expectations. Journal of Business & Industrial Marketing, 11, 724. Stump, R. L., & Sriram, V. (1997). Employing information technology in purchasing: Buyer supplier relationships and size of the supplier base. Industrial Marketing Management, 26, 127136. Talluri, S., & Ragatz, G. L. (2004). Multi-attribute reverse auctions in B2B exchanges: A framework for design and implementation. Journal of Supply Chain Management, 40, 5260. Trent, R. J., & Monczka, R. M. (2002). Pursuing competitive advantage through integrated global sourcing. Academy of Management Executive, 16(2), 6680. Venkatesh, R., Kohli, A., & Zaltman, G. (1995). Inuence strategies in buying centers. Journal of Marketing, 59, 7186. Webster, F., Jr., & Wind, Y. (1972). Industrial marketing strategy. New York: Wiley. Webster, F., Jr., & Wind, Y. (1996). A general model for understanding organizational buying behavior. Marketing Management, 4, 5257. Weigand, R. E. (1968). Why studying the purchasing agent is not enough. Journal of Marketing, 32, 4145. Welch, L., & Luostarinen, R. K. (1993). Inwardoutward connections in internationalization. Journal of International Marketing, 1, 4456.
368
Wenninger, J. (1999). Business-to-business electronic commerce. Current Issues in Economics and Finance, 5, 16. Wilson, E. J. (1996). Theory transitions in organizational buying behavior research. Journal of Business & Industrial Marketing, 11, 719. Yip, G. S. (1992). Total global strategy. Englewood Cliffs, NJ: Prentice-Hall.
INTERNET-BASED KNOWLEDGE INTERNALIZATION AND FIRM INTERNATIONALIZATION IN TRANSITION MARKETS

Tho D. Nguyen and Nigel J. Barrett
ABSTRACT
Realizing that the Internet is a source of information and the possibility to transform it into knowledge, this study develops an IBK-Internalization process in which internationalizing rms in transition markets utilize the Internet to search for information about foreign markets, to assess its relevance, and then, to internalize it for their internationalization. It is found that IBK-Internalization underlies international orientation and foreign sales intensity, which in turn, has a reciprocal effect on IBKInternalization. Further, learning orientation facilitates the IBKInternalization process. These ndings suggest that internationalizing rms should promote and value the IBK-Internalization process in order to mitigate their lack of foreign market knowledge.
369
370
THO D. NGUYEN AND NIGEL J. BARRETT
INTRODUCTION
In an economy where the only certainty is uncertainty, the one sure source of competitive advantage is knowledge. When markets shift, technologies proliferate, competitors multiply, and products become obsolete almost over night, successful companies are those that consistently create new knowledge, disseminate it widely through the organization, and quickly embody it in new technology and products (Nonaka, 1991, p. 96).
Knowledge is an invaluable asset that enables rms to remain competitive or even to survive, especially in the knowledge society (e.g., Davenport & Prusak, 2000; Drucker, 1988). In the literature on internationalization, knowledge has also been posited to be a key factor that affects rms internationalization behavior (e.g., Cavusgil, 1980; Johanson & Vahlne, 1977; Liesch & Knight, 1999; Ling-yee, 2004; Rialp & Rialp, 2001). Firms conducting business in foreign markets are confronted with greater risks than they face in their home markets because foreign markets tend to be characterized by heterogeneity, sophistication, and turbulence. Moreover, the differences in cultural, national, economic, political, legal, social, and other environmental inuences have made it more complex and difcult to conduct foreign market research (Craig & Douglas, 2000). As a result, obtaining foreign market knowledge is not an easy task, especially for rms in transition economies, such as Vietnam, because they lack resources for undertaking foreign market research such as foreign market experiments and foreign market visits (Nguyen & Barrett, 2006). The Internet has received a high level of attention by rms in the last decade because it offers several commercial applications for rms around the world (Bennett, 1997; Hamill, 1997; Morgan-Thomas & Bridgewater, 2004). With the inception of the World Wide Web, commercial applications of the Internet have proliferated (Hamill, 1997; Javalgi, Radulovich, Pendleton, & Scherer, 2005; Luo, Zhao, & Du, 2005). Several rms around the world, domestic and multinational, both in developed, developing, and transition economies such as China and Vietnam, are establishing their presence on the Internet. There are also a number of e-trade providers in transition markets such as MeetVietnam.com and MeetChina.com that provide rms in these countries with online tools for their internationalization (Asia Today International, 2000). The Internet may provide a different environment for international marketing in general, and for rm internationalization in particular. The Internets low-cost communication may permit rms with limited capital to become global marketers at an early stage of their development (Luo et al.,
IBK-Internalization and Firm Internationalization
371
2005; Quench & Klein, 1996). The Internet becomes a powerful tool for supporting networks, both internal and external to the rm (Hamill, 1997; Morgan-Thomas & Bridgewater, 2004). Internet connection can substantially improve communications with existing foreign customers, suppliers, agents, and distributors. It can also identify new customers and distributors and generate a wealth of information on market trends, as well as on the latest technology, research, and technical development (Hamill & Gregory, 1997; Kaynak, Tatoglu, & Kula, 2005). It is also a new and efcient medium for conducting market research (e.g., Lescher, 1995; Weible & Wallace, 2001). Using various Internet tools such as e-mails, search engines, and online surveys, internationalizing rms can collect a variety of data and information about foreign markets. This source of information is very promising because it is cost effective and speedy (e.g., Hamill, 1997; Luo et al., 2005; Reedy & Schullo, 2004). It is the innovation of information and communications technologies that has given an opportunity for information acquisition that is substantially more efcient for rms all around the world (Porter & Millar, 1985). Several studies have been conducted to explore the usefulness of the Internet for international marketing (e.g., Bennett, 1997; Hamill & Gregory, 1997; Morgan-Thomas & Bridgewater, 2004; Srensen & Buatsi, 2002; Sheth & Sharma, 2005). However, less attention has been paid to the transformation of data and information obtained from the Internet for rm internationalization. Moreover, the theory of knowledge in internationalization posits that the knowledge for a rms international expansion is mainly experiential or tacit, which can mainly be obtained during its international operations (e.g., Johanson & Vahlne, 1977). Nevertheless, it does not cater for how the rm develops the knowledge. Models of knowledge creation help explain how the rm develops knowledge, however, they have been found to be largely conceptual (e.g., Akbar, 2003; Lam, 2000; Nonaka, 1994; Davenport & Prusak, 2000). More importantly, most research in this area has been conducted primarily in advanced economies. Little has been undertaken in the developing world, especially in transition economies (Chao, Samiee, & Yip, 2003). In an attempt to bridge the above gap, this study examines the process in which a rm searches for data and information on the Internet, assesses their relevance for internationalization, and then, internalizes them with the existing foreign market knowledge in the rm. This process is termed Internet-based knowledge internalization, or simply, IBK-Internalization. An antecedent and outcomes of IBKInternalization are also examined. The antecedent is learning orientation and the outcomes are international orientation and foreign sales intensity.
372
The rest of the article is structured as follows: rst, we propose a conceptual model of IBK-Internalization and rm internationalization and hypotheses. Subsequently, we present the method, data analysis, and results. We, then, discuss the results and implications. We conclude the article by addressing a number of limitations and directions for future research.
IBK-INTERNALIZATION AND FIRM INTERNATIONALIZATION

Fig. 1 depicts a conceptual model of IBK-Internalization. Firms utilize the IBK-Internalization to obtain foreign market knowledge for their internationalization. This will lead to a greater degree of their international orientation and foreign sales. It is also postulated that learning orientation is a key antecedent of the IBK-Internalization process.
IBK-Internalization According to the resource-based view of the rm, although tangible resources are determinants of performance, intangible resources including knowledge are the key factors contributing to a rms competitive advantages (Barney, 1991; Wernerfelt, 1984). It is argued that using computers and access to the Internet are unlikely to be a source of sustainable competitive advantage in internationalization because these are imitable (Barney, 1991; Samiee, 1998). Every rm can access the Internet to collect data and information about foreign markets, nevertheless, sustainable
H1
Foreign Sales Intensity
Learning Orientation
H4
IBKInternalization
H3
International H2 Orientation
373
competitive advantage in the form of knowledge resources can only be gained by rms if they are capable of creating knowledge based on the Internet. The dynamic theory of knowledge creation, especially the internalization process (Nonaka, 1994; Nonaka & Takeuchi, 1995) offers an explanation of how a rm obtains information and then transforms it into knowledge. Internalization is the process of searching for, acquiring, and absorbing both tacit and explicit information and translating it into knowledge which is then applied to some purpose (Liesch & Knight, 1999, p. 385). In this study, IBK-Internalization is dened as the process in which data and information collected from the Internet are acquired, assessed, and compared with other sources, interpreted collectively, and then, used for rm internationalization. Three important features of IBK-Internalization are Internet utilization (Srensen & Buatsi, 2002; Kaynak et al., 2005), assessment of information relevance (Davenport & Prusak, 2000), and information internalization (Knight & Liesch, 2002; Nonaka, 1994). Internet Utilization Internet utilization assists rms in collecting data and information on the Internet. This is made possible due to the wide range of data and information about foreign markets on the Internet (Hamill, 1997; Srensen & Buatsi, 2002). The Internet operates as a virtual library which can be accessed by users everywhere around the world (Ancel, 1999). Webb and Sayer (1998) believe that due to the Internets interactive characteristic of the Internet, it can in fact function as much more than a virtual library. However, the opportunity given by the Internet has not been, in practice, properly utilized (Webb & Sayer, 1998). Frequently utilizing Internet tools such as search engines and e-mails, internationalizing rms are able to collect a considerable amount of data and information about foreign markets (Bennett, 1997; Nguyen & Barrett, 2006; Reedy & Schullo, 2004; Srensen & Buatsi, 2002). Internet Information Relevance Internet information relevance refers to the level of usefulness of information obtained from the Internet for rm internationalization. Internet utilization assists rms in collecting data and information on the Internet. However, to obtain relevant and purposive information for internationalization, rms should use it effectively (i.e., use suitable tools for obtaining information) and efciently (i.e., use those tools skillfully). This implies that rms not only search for relevant and purposive data but also, categorize,
374
correct, and summarize them, i.e., to transform data into information (Davenport & Prusak, 2000). This process can enhance the level of usefulness of information, i.e., relevant information for their internationalization (information about the environment, market characteristics, and the marketing mix; Leonidou & Adam-Florou, 1999). Internet Information Internalization Access to the Internet, where various types of data and information about foreign markets are available, is a promising source of information (e.g., Nguyen & Barrett, 2006; Srensen & Buatsi, 2002). It is knowledge, however, not information per se, that is a signicant factor that affects a rms internationalization behavior (Johanson & Vahlne, 1977). When acquiring relevant information from the Internet, the rm, therefore, is likely to transform it into higher levels of knowledge to satisfy its knowledge need for internationalization. Internet information internalization refers to the interaction between information obtained from the Internet with other sources of information and knowledge within the rm. The interaction relates to the process of information comparison and interpretation, and the use of information for making business decisions (Davenport & Prusak, 2000; Nonaka & Takeuchi, 1995).
IBK-Internalization and Firm Internationalization As discussed previously, foreign market knowledge is a convincing explanation for the incremental manner of internationalization in which rms gradually acquire, integrate, and use knowledge about foreign markets to successively increase their commitment to foreign markets (Johanson & Vahlne, 1977). Liesch and Knight (1999, p. 385) argue that [of] all resources, information and knowledge are perhaps the most critical to the expansion of SMEs into foreign markets. Therefore, systematic acquisition of information and knowledge about foreign markets is critical for rms success during their internationalization. There are a number of ways that rms can acquire information and knowledge about foreign markets. This can be achieved through international marketing research, export assistance, and export intelligence (Leonidou & Adam-Florou, 1999). However, rms in transition markets like Vietnam lack resources for obtaining foreign market information and knowledge through such traditional modes (Nguyen & Barrett, 2006). In addition, compared to rms in advanced economies, rms in transition
375
markets are short of international experience. This is because in the last several years, international business activities of those rms have been arranged by their governments. When these markets have been transformed into a new system, i.e., a market-oriented system, arrangements between governments have almost ceased. International business activities are no longer conducted in the traditional way (Grifn, 1998). Firms must nd foreign markets for their products rather than relying on governments arrangements. In so doing, knowledge about foreign markets is perhaps a critical factor for their success and a reasonable explanation for their internationalization (Ling-yee, 2004; Nguyen & Nguyen, 2001). Consequently, it is important that those rms should nd alternatives to obtain knowledge in order to enhance their internationalization. The IBK-Internalization process may offer those rms an opportunity for obtaining foreign market knowledge, provided that they are fully aware of this possibility and are prepared to exploit it. Firms that are capable of employing the IBK-Internalization process can signicantly enhance its knowledge bank for their internationalization. IBK-Internalization is not a static process; it is a dynamic process in which internalized knowledge about foreign markets is created, which can provide international advantages for rms that utilize it. In short, [i]n the information age, a companys survival depends on its ability to capture intelligence, transform it into usable knowledge, embed it as organizational learning, and diffuse it rapidly throughout the company (Bartlett & Ghoshal, 1995, p. 141). Firms pursue IBK-Internalization to solve the problem of the lack of knowledge about foreign markets. The relationships between knowledge, internationalization attitudes, which reect a rms international orientation, i.e., its belief, preference, and willingness to commit to international expansion (Barrett, 1986; Stump, Athaide, & Axinn, 1998), and internationalization behavior are well documented in the literature on internationalization (Haahti, Madupu, Yavas, & Babakus, 2005; Johanson & Vahlne, 1977). A recent study of exporting rms in China, a transition market, by Ling-yee (2004) also reveals that knowledge about foreign markets plays a key role in export intensity. In addition, several studies have found that internationalization attitudes play a vital part in rms development and performance in foreign markets (e.g., Bilkey, 1978; Francis & Collins-Dodd, 2000; Johnston & Czinkota, 1982; Madsen, 1987; Stump et al., 1998). Accordingly, H1. The greater the degree of IBK-Internalization of a rm, the greater is the degree of its foreign sales.
376
H2. The greater the degree of IBK-Internalization of a rm, the greater is the degree of its international orientation. H3. The greater the degree of international orientation of a rm, the greater is the degree of its foreign sales.
Learning Orientation and IBK-Internalization Learning orientation is an organizational factor that inuences the propensity of a rm to create and use knowledge (Sinkula, Baker, & Noordewier, 1997; Slater & Narver, 1995). A learning-oriented rm creates and encourages a learning environment throughout the rm. The rm continuously promotes the organizational learning process, i.e., information acquisition, information dissemination, and shared interpretation (Sinkula, 1994). The rm endlessly creates and uses new knowledge that has the potential to inuence the rms performance (Teo, Wang, Wei, Sia, & Lee, 2005; Sinkula et al., 1997). Consequently, the rm will be more comfortable in dealing with innovations. This gives rise to the ability to adopt and implement new ideas, processes, or products, i.e., to produce innovative capacity for the rm (Calantone, Cavusgil, & Zhao, 2002). This means that learningoriented rms will never be satised with its existing level of knowledge. This will lead to the need for acquisition, assessment, and transformation of information and knowledge from all accessible sources (Lam, 2000; Nonaka & Takeuchi, 1995; Teo et al., 2005), including the Internet. Therefore, learning-oriented rms are more likely to create a learning culture, which can initiate, support, and maximize the IBK-Internalization process within the rm. Accordingly, H4. The greater the degree of learning orientation of a rm, the greater is the degree of its IBK-Internalization.
Rival Model Rival models play an important role in theory construction. Bagozzi (1984, p. 16) argues that [i]t is important to stress that tests of rival hypotheses should not be reserved for separate studies but should be performed whenever possible within the context of an on-going study. In this way, because subjects, settings, instruments, etc. are held constant, we will have greater condence in the internal validity of the rival hypotheses. Based on this, a
377
rival model is proposed. As discussed previously, when a rm acquires appropriate knowledge about foreign markets the rm tends to enhance its commitment to foreign markets, thus leading to an improvement of its performance in foreign markets. In turn, it could be expected that, when the rm has a greater degree of commitment to foreign markets, knowledge required for its internationalization will amplify. This leads to an increase in IBK-Internalization. Therefore, it is proposed that Hc. The greater the degree of foreign sales of a rm, the greater is the degree of its IBK-Internalization.
METHOD
Measurement IBK-Internalization IBK-Internalization was conceptualized as a high-order construct, comprising three dimensions: Internet utilization; Internet information relevance; and Internet information internalization. Internet utilization was measured by two indicators. The rst indicator was a measure of time spent searching the Internet, i.e., asking respondents how many hours per week the rm uses the Internet to search for information on foreign markets, such as using search engines, visits to websites of foreign distributors, competitors, suppliers, and customers. The second indicator was the frequency of using e-mail for international business purposes. It was measured by asking respondents how many times per week the rm receives and sends e-mail related to international business activities. Even though the electronic survey is an important tool for research on the Internet (e.g., Lescher, 1995), a discussion with managers of a research rm in the market showed that these tools had not been widely used in the market to date. Therefore, this tool was included in the measure of Internet utilization as part of the e-mail tool. Internet information relevance was a second-order construct, comprising three dimensions: market feasibility information; adaptation information; and, background information (Hart, Webb, & Jones, 1994). Fourteen items were used to measure the three dimensions of information relevance. Market feasibility information was measured by three items covering information on potential distributors, buyers, and suppliers in foreign markets. Adaptation information was measured by ve items addressing information on
378
characteristics of markets, such as buyers preferences, competition, market size, and growth. Finally, background information was measured by six items covering general information such as socio-economic information, legal issues, political forces, and infrastructure. These items were adapted from Hart et al.s (1994) scale with modications to suit the setting of this study (i.e., based on the types of usable information on foreign markets that the rm can obtain from the Internet). Internet information internalization was also a second-order construct, comprising two dimensions: information transformation and information use. Information transformation was measured by ve items addressing the process of comparing, relating, and fusing information collected from the Internet with information and knowledge obtained from other sources, including experience of members of the rm (Davenport & Prusak, 2000; Nonaka & Takeuchi, 1995). Information use was measured by ve items based on the scale developed by Diamantopoulos and Souchon (1999). The literature on information utilization conceptualizes three forms of information use, i.e., instrumental use, conceptual use, and symbolic use (e.g., Menon & Varadarajan, 1992). Instrumental use of information relates to the direct application of information obtained to solve a marketing problem. Conceptual use refers to the use of information to develop a managerial knowledge base, i.e., indirect use of information. Symbolic use is the use of information mainly for supporting an opinion or, justifying a decision, rather than using it in a manner consistent with the intended purpose of the rm (Menon & Varadarajan, 1992). Among three types of information use, only instrumental and conceptual uses are of importance for knowledge internalization because these types of information use assist the rm in generating knowledge. Research has found that these two types of information use are unidimensional (Diamantopoulos & Souchon, 1999). Foreign Sales Intensity Foreign sales as a percentage of total sales (fsts) was used to measure foreign sales intensity (Moini, 1995; Ling-yee, 2004). Learning Orientation The conceptualization and measurement of learning orientation in this study were borrowed from Sinkula et al. (1997), and comprised three dimensions: commitment to learning; shared vision; and open-mindedness. Commitment to learning was measured by four items reecting the degree that a rm is willing to commit to learning, i.e., to promote and nourish a learning culture within the rm. Shared vision was also measured by four items embodying
379
the focus of the rm on learning that fosters energy, commitment, and purpose among all members of the rm. Finally, open-mindedness was measured by three items mirroring the unlearning process of the rm. International Orientation International orientation was also a second-order construct, comprising two dimensions: international conviction and international intention (Barrett, 1986). International conviction refers to the belief of managers that international business activities would contribute to the achievement of a rms goals and competitiveness and its preference for international expansion (as opposed to other strategies). It was measured by eight items. International intention reects the managers willingness to commit resources to international business activities, and was measured by four items. Foreign sales as a percentage of total sales (fsts) was measured within a range of 110 (1p10%, 2 1120%, y , 10 91100%). Internet utilization was measured at ratio level, and was then recoded by values varying from 1 to 10 for analysis. All other items were measured by a 5-point rating scale (anchored by 1: strongly disagree and 5: strongly agree). Measurement Renement A focus group was undertaken with six managers who were responsible for international business activities of rms that had used the Internet for their internationalization. Even though most of the scales have been used widely in the past, this step is important because of the difference in the research setting, i.e., in the context of a transition market. A quantitative pilot survey followed to rene the measures. This survey was conducted using faceto-face interviews with 89 rms in Ho Chi Minh City, Vietnam. The scales were assessed via Cronbachs alpha (except for foreign sales intensity and Internet utilization, which were measured by one and two indicators, respectively) and exploratory factor analysis (EFA). Principal axis factoring with promax rotation was used because it accurately reects the underlying structure of the data than that provided by an orthogonal solution such as varimax (Gerbing & Anderson, 1988). The results indicate that these measures achieved a satisfactory level of reliability (a>0.70) and factor loadings (l>0.50). Accordingly, these measures were used in the main survey. The Sample Vietnam was selected to empirically test the model because it represents an under-investigated transition economy (Tsang, 2005). A systematic sample
380
of 306 rms in Ho Chi Minh City, the major business center of Vietnam, was surveyed. The sampling frame, based on the Business Directories in Ho Chi Minh City, consisted of about 5,000 internationalizing rms in all industries, which used the Internet for their international business activities. The single key informant approach, the most commonly used method in organizational research (Kumar, Stern, & Anderson, 1993), was used in this study. Respondents were senior executives of the rms. However, the interviewers were instructed to reach relevant people in the organization in order to request specic information, such as Internet usage, which might not be given by senior managers. The original questionnaire was in English. This English version was translated into Vietnamese because English is not well understood by managers in this market. Back translation was used to ensure the equivalence of meanings. Partial self-administered surveys, in which questionnaires are mailed to the target respondents and are collected by interviewers, were used for this study. Follow-up telephone calls to remind respondents to complete the questionnaires prior to collection were also conducted. Four hundred questionnaires were distributed to rms in the chosen sample and 327 completed questionnaires were collected yielding a response rate of 82%. Among these completed questionnaires, 21 were found to be invalid, due to nonqualied respondents, e.g., respondents not being members of top management responsible for international business activities. Consequently, the remaining 306 valid completed questionnaires comprised the sample for this research. The sample comprised 264 (83.6%) rms involved in indirect exporting, 248 (81%) rms involved in direct exporting, 44 (14.4%) rms involved in contract modes, and four rms (1.3%) involved in foreign direct investment. In terms of rm size, 60 (19.6) rms had more than 300 employees; 170 (55.6%) had from 100 to 300 employees; and 76 (24.8%) rms had fewer than 100 employees. In terms of Internet applications, all rms had used e-mail and search engine tools; 80 (30%) had a website; and only 3 (1%) had used the Internet as a channel of distribution.
DATA ANALYSIS AND RESULTS

Measurement Models The measures were rst assessed via Cronbachs alpha, except for Internet utilization and foreign sales intensity, which were measured by two and one items, respectively. The results indicate that all the scales satised the
381
requirement for reliability (a>0.80). Specically, the Cronbachs alphas of the two dimensions of Internet information internalization, i.e., information transformation and information use, were 0.83 and 0.87, respectively. The Cronbachs alphas of three dimensions of Internet information relevance, i.e., market feasibility information, adaptation information, and background information, were 0.80, 0.93, and 0.87, respectively. The Cronbachs alphas of three dimensions of learning orientation, i.e., commitment to learning, shared vision and open mindedness, were 0.84, 0.82, and 0.81, respectively. Finally, the Cronbachs alphas of the two dimensions of international orientation, i.e., international conviction and international intention, were 0.92 and 0.88, respectively. These scales were then assessed via conrmatory factor analysis (CFA). The screening process shows that the data exhibited slight deviations from normality. Nonetheless, all univariate kurtoses were nonsignicant and all skewnesses were within the range of [1, 1]. Therefore, maximum likelihood estimation was used (Muthen & Kaplan, 1985). IBK-Internalization was a high-order construct comprising two secondorder constructs (Internet information relevance and Internet information internalization) and one rst-order construct (Internet utilization). Learning orientation and international orientation were also second-order constructs. The CFA results of these second-order constructs are shown in Table 1. The ndings indicate that the measurement models of these constructs t the data well. The correlations (with standard errors) between the dimensions of these second-order constructs indicate that they were signicantly different from unity (po0.001). Therefore, the within-construct discriminant validity was achieved (Steenkamp & van Trijp, 1991). The correlations between three dimensions of IBK-Internalization (Internet utilization, Internet information relevance, and Internet information internalization) were also signicantly different from unity (po0.001). Consequently, the within-construct discriminant validity of IBK-Internalization was also achieved (Steenkamp & van Trijp, 1991). The saturated model (nal measurement model) also received a good t to the data: w2 (p 0.000); CFI 0.978; TLI 0.977; [1062] 1233.84 RMSEA 0.023. It is noted that because foreign sales intensity was measured by a single item (fsts), it was assumed to have a reliability of .85. To make the model identied, the error-term variance of this item was xed at 0.623 [(1a) s2(fsts)] (Joreskog & Sorbom, 1982). The factor loadings of all items were high and substantial (the lowest loading was 0.65), and all were signicant (po0.001) (see the appendix for the standardized factor loadings of items). Also, all average variances extracted were equal to or greater than
382
Table 1. Measurement Validation.

Constructs Learning orientation: w2 51.2 [41] (p>0.13) Dimensions rc 0.84 0.82 0.82 0.92 0.88 0.85 rvc 0.57 0.54 0.60 0.65 0.64 0.65 inad"infea 0.58(0.078) Correlation lcom"lsv Lcom"lop lsv"lop inco"intent r(se) 0.64(0.092) 0.53(0.086) 0.58(0.091) 0.63(0.084)
Commitment to learning (lcom) Shared vision (lsv) Open mindedness (lop) International International orientation: conviction (inco) w2 38.68 International [34] (p>0.27) intention (intent) Internet information Market feasibility relevance: information w2 91.7 (infea) [74] (p>0.08) Adaptation information (inad) Background information (inbak) Internet information Information internalization: transformation w2 41.4 (intra) [34] (p>0.17) Information use (inuse) IBI-internalization: Internet utilization w2 334.74 (IUT) [291] (p>0.03); Internet information TLI 0.989; relevance (IIR) CFI 0.990; Internet information RMSEA 0.022 internalization (INT)
0.94 0.90
0.76 0.59
inad"inbak infea"inbak
0.55(0.079) 0.55(0.083)
0.83
0.50
intra"inuse
0.51(0.082)
0.87 0.67
0.58 0.50 IUT"IIR IIR"INT IUT"INT 0.38(0.092) 0.50(0.112) 0.48(0.110)
Note: r(se), correlation (standard error); rc, composite reliability; rvc, average variance extracted.
0.50. These ndings indicate that the scales were unidimensional and convergent validity (within-method) was achieved (e.g., Fornell & Larcker, 1981; Steenkamp & van Trijp, 1991). The correlations between constructs together with their standard errors are shown in Table 2. These ndings indicate that they were signicantly different from unity, thus, supporting the across-construct discriminant validity (Steenkamp & van Trijp, 1991).
383
Table 2.
Correlations
Correlation (r) with Standard Error (se) between Constructs.

r(se) 0.71(0.143) 0.33(0.087) 0.22(0.075) 0.62(0.131) 0.35(0.080) 0.53(0.110)
Learning orientation"IBK-Internalization Learning orientation"International orientation Learning orientation"Foreign sales intensity IBK-Internalization"International orientation International orientation"Foreign sales intensity IBK-internalization"Foreign sales intensity
Table 3.
Structural Paths
Unstandardized Regression Weights in the Proposed and Rival Models.

Proposed Model Est(se) p-value Rival Model Est(se) p-value 1.38(.591) 1.04(0.218) 0.21(0.272) 0.47(0.089) 0.05(0.022) 0.020 0.000 0.450 0.000 0.023
H1: IBK-Internalization -Foreign sales intensity 2.00(0.559) H2: IKB-Internalization -International orientation1.05(0.212) H3: International orientation-Foreign sales intensity 0.29(0.262) H4: Learning orientation -IBK-Internalization 0.49(0.093) Hr: Foreign sales intensity -IBK-Internalization 0.00
Estimate (with standard error). Fixed at 0.
0.000 0.000 0.270 0.000
Structural Models The proposed model received a good t to the data: w2 [1064] 1240.12 (p 0.000); CFI 0.977; TLI 0.976; and RMSEA 0.023. The rival model also t the data well: w2 [1063] 1235.87 (p 0.000); CFI 0.978; TLI 0.977; RMSEA 0.023). A chi-square difference test shows that the rival model did contribute to a better overall model t: Dw2 1] 4.25 [Ddf (po0.05). More importantly, the additional path hypothesized in the rival model was signicant (see Table 3 for the unstandardized estimates and Fig. 2 for the standardized estimates). Consequently, the rival model was selected (Anderson & Gerbing, 1988). Furthermore, the existence of a feedback loop (formed by IBK-Internalization, international orientation, and foreign sales intensity) in the rival model (a nonrecursive model) can cause the system to become unstable. The results show that the stability index was 0.100, falling within the range of (1, +1), indicating that the model was stable (Bentler & Freeman, 1983). It is noted that no improper solution was
384

inem insea infea inad inbak intra inuse 0.73 0.68 0.76 0.75 0.74 IUT lop lsv lcom 0.71 0.82 0.76 Learning Orientation 0.76 0.54 0.76 0.92 0.24S Foreign Sales Intensity 0.24*(Hr) 0.08ns(H3) 0.56**(H2) 0.33S 0.83 inco International Orientation 0.76 intent 0.59S 0.28*(H1) IIR 0.73 0.69 INT fsts
0.63**(H4)
IBKInternalization
2[1063] = 1235.87 (p = 0.000); CFI = 0.978; TLI = 0.977; RMSEA = 0.023

*: significant at p < 0.05; **: significant at p < 0.001; ns: non-significant Hr: hypothesized in the rival model; S: squared multiple correlations
Fig. 2.
Structural Results (Standardized Estimates).
found in any of the CFA or structural models: Heywood cases were absent; all error-term variances were signicant; and all standardized residuals were less than|2.58|. Consistent with H1 and H2, IBK-Internalization had positive impacts on both international orientation and foreign sales intensity. A positive relationship between learning orientation and IBK-Internalization was also found to be high and signicant, supporting H4. However, inconsistent with H3, the relationship between international orientation and foreign sales intensity was not signicant, although it was in the same hypothesized direction. In addition, the impact of foreign sales intensity and IBKInternalization hypothesized in the rival model (Hr) was signicant.
DISCUSSION AND IMPLICATIONS

The objective of this study is to explore the possibility of the Internet as a source of foreign market knowledge for rm internationalization. Realizing that the Internet provides a promising source of data and information about foreign markets (Hamill, 1997; Kaynak et al., 2005; Srensen & Buatsi, 2002), and the possibility to transform data and information into knowledge (Nonaka, 1994; Akbar, 2003), we focus on IBK-Internalization. It is a
385
process in which internationalizing rms utilize the Internet to search for data and information about foreign markets, to assess the relevance of data and information, and then, to internalize them for their internationalization. We also examine a key antecedent (learning orientation) and outcomes (international orientation and foreign sales intensity) of IBK-Internalization. The results indicate that the IBK-Internalization process is a useful source of foreign market knowledge for rm internationalization. When rms utilize this source of knowledge, an improvement of internationalization attitudes, reected in international orientation, and performance, reected in foreign sales intensity, will be made. In turn, with an increase in foreign sales, rms would require more knowledge about foreign markets to facilitate their internationalization. This stimulates rms to utilize the IBKInternalization process to a greater extent, which has been conrmed in this study. This dynamic process enhances and extends rms knowledge about foreign markets. These ndings are consistent with the literature on internationalization, i.e., experiential or tacit knowledge is more meaningful in explaining rms international performance (Johanson & Vahlne, 1977) as well as the theory of knowledge creation and use (Davenport & Prusak, 2000; Menon & Varadarajan, 1992; Nonaka, 1994). Furthermore, learning orientation has been found to be a key organizational culture that inuences rms attitudes and behaviors (Slater & Narver, 1995). This role of learning orientation has also been conrmed. We have found that learning orientation motivates rms to utilize the IBKInternalization process. This is because learning orientation facilitates innovation as well as stimulates the process of information acquisition, processing, and use (Sinkula et al., 1997), i.e., knowledge creation. It is also noted that the relationship between international orientation and foreign sales intensity was not signicant. In general, attitudes and behavior are positively related (Frazier & Sheth, 1985). Nevertheless, some rms lack resources to translate their committed attitudes to committed behavior (Stump et al., 1998). This may be the case of Vietnamese internationalizing rms. The ndings show that the impact of IBK-Internalization on international orientation is much higher than that of on foreign sales intensity. This implies that foreign market knowledge is a key determinant of rm international orientation. However, it must be noted that although knowledge is a key factor, it is not the only one that assists rms in translating internationalization attitudes into internationalization performance. The ndings of this study suggest a number of implications for managers in internationalizing rms. The resource-based view of the rm recognizes that knowledge is a key resource contributing to rms competitive
386
advantage (Penrose, 1959; Wernerfelt, 1984), and the Internet is one source of knowledge. Firms should therefore invest in physical resources such as computers and Internet access. However, rms should be aware that the use of computers or access to the Internet is not sufcient to achieve sustainable competitive advantage because these are imitable (Barney, 1991; Samiee, 1998). Managers should keep in mind that access to the Internet to obtain data and information about foreign markets is only one part of the IBKInternalization process. IBK-Internalization involves a process in which data and information from the Internet are acquired, assessed, and compared with other sources, interpreted collectively, and then, used for rm internationalization. In addition, full utilization of IBK-Internalization requires rms to create a learning environment that promotes and values the knowledge creation process. It is expected that the results of this research will encourage managers to utilize this possibility to increase their level of knowledge about foreign markets. In addition, although the study investigates internationalizing rms in a transition market Vietnam, it is expected that the ndings of this study would be of value for rms in other transition markets and other internationalizing rms around the world.
LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH

This study has a number of limitations, which has been addressed in the following. First, although the proposed model received support in a transition market (Vietnam), further replication, extension, and critical evaluation within similar and/or different markets are required, particularly in markets that have different economic, political, and cultural backgrounds. Second, the Internet is a relatively recent innovation and its potential as an effective and efcient source of data and information about foreign markets has not yet been fully exploited, e.g., the use of online surveys. Consequently, the measures of IBK-Internalization dimensions should be modied and extended in future research. Further, foreign sales intensity was the only measure of internationalization performance in this study. It is recommended that future research should use several other measures such as foreign sales growth and prots in foreign markets (Cavusgil & Zou, 1994). Third, a comparison between the utilization of the IBK-Internalization process by different types of internationalizing rms, such as indirect, direct exporting, franchising, and foreign direct investment rms will be of interest
387
in future research. Fourth, the IBK-Internalization process involves dynamic interactions between different organizational members as well as between different levels of knowledge. Therefore, using a more appropriate method of investigation, such as action research, to explore the process of information and knowledge acquisition and conversion within internationalizing rms is worthy of investigation in future research. Finally, although the key informant approach is commonly used in organizational research (Kumar et al., 1993), collecting data from multiple informants is an alternative method recommended for future research, particularly for research that involves several members of the rm.
ACKNOWLEDGMENTS
The authors would like to thank the special issue editors, Alex Rialp and Josep Rialp, the two anonymous reviewers, and the participants of the CIMaR 2005 at the Autonomous University of Barcelona, Spain, for their constructive and insightful comments on the previous version of the article.
REFERENCES
Akbar, H. (2003). Knowledge levels and their transformation: Towards the integration of knowledge creation and individual learning. Journal of Management Studies, 40(8), 19972021. Ancel, B. (1999). Using the internet: Exploring international markets. International Trade Forum, 1, 1416. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411423. Asia Today International. (2000). MeetVietnam.com on-line. Asia Today International, 18(7), 21. Bagozzi, R. P. (1984). A prospectus for theory construction in marketing. Journal of Marketing, 48(1), 1129. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99120. Barrett, N. J. (1986). A study of the internationalization of Australian rms. Unpublished doctoral dissertation. University of New South Wales, New South Wales. Bartlett, C. A., & Ghoshal, S. (1995). Changing the role of top management: From systems to people. Harvard Business Review, 73(3), 132142. Bennett, R. (1997). Export marketing and the internet: Experiences of web site use and perceptions of export barriers among UK businesses. International Marketing Review, 14(5), 324344. Bentler, P. M., & Freeman, E. H. (1983). Tests for stability in linear structural equation systems. Psychometrika, 48(1), 143145.
388
Bilkey, W. J. (1978). An attempted integration of the literature on the export behavior of rms. Journal of International Business Studies, 9(1), 3346. Calantone, R. J., Cavusgil, S. T., & Zhao, Y. (2002). Learning orientation, rm innovation capacity, and rm performance. Industrial Marketing Management, 31(6), 515524. Cavusgil, S. T. (1980). On the internationalization process of the rms. European Research, 8(6), 273281. Cavusgil, S. T., & Zou, S. (1994). Marketing strategyperformance relationship: An investigation of the empirical link in export market ventures. Journal of Marketing, 58(1), 121. Chao, P., Samiee, S., & Yip, L. S.-C. (2003). International marketing and the Asia-Pacic region developments, opportunities, and research issues. International Marketing Review, 20(5), 480492. Craig, S. C., & Douglas, S. P. (2000). International marketing research (2nd ed.). New York: Wiley. Davenport, T. H., & Prusak, L. (2000). Working knowledge: How organizations manage what they know. Boston: Harvard Business School Press. Diamantopoulos, A., & Souchon, A. L. (1999). Measuring export information use: Scale development and validation. Journal of Business Research, 46(1), 114. Drucker, P. (1988). The coming of the new organization. Harvard Business Review, 66(1), 4554. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobserved variables and measurement error. Journal of Marketing Research, 28(1), 3950. Francis, J., & Collins-Dodd, C. (2000). The impact of rms export orientation on the export performance of high-tech small and medium-sized enterprises. Journal of International Marketing, 8(3), 84103. Frazier, G. L., & Sheth, N. J. (1985). An attitudebehavior framework for distribution channel management. Journal of Marketing, 49(2), 3848. Gerbing, W. D., & Anderson, J. C. (1988). An update paradigm for scale development incorporating unidimensionality and its assessments. Journal of Marketing Research, 25(2), 186192. Grifn, K. (1998). Restructuring and economic reforms. In: K. Grifn (Ed.), Economic reform in Vietnam (pp. 119). London: Macmillan. Haahti, A., Madupu, V., Yavas, U., & Babakus, E. (2005). Cooperative strategy, knowledge intensity and export performance of small and medium sized enterprises. Journal of World Business, 40(2), 124138. Hamill, J. (1997). The internet and international marketing. International Marketing Review, 14(5), 300323. Hamill, J., & Gregory, K. (1997). Internet marketing in the internationalization of UK SMEs. Journal of Marketing Management, 13(13), 928. Hart, S. J., Webb, J. R., & Jones, M. V. (1994). Export marketing research and the effect of export experience in industrial SMEs. International Marketing Review, 11(6), 422. Javalgi, R. G., Radulovich, L. P., Pendleton, G., & Scherer, R. F. (2005). Sustainable competitive advantage of internet rms: A strategic framework and implications for global marketers. International Business Review, 22(5), 658672. Johanson, J., & Vahlne, J.-E. (1977). The internationalisation process of the rm a model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 8(1), 2332. Johnston, W. J., & Czinkota, M. R. (1982). Managerial motivations as determinants of industrial export behavior. In: M. R. Czinkota & D. Tesar (Eds), Export management: An international context (pp. 317). New York: Praeger.
389
Joreskog, K. G., & Sorbom, D. (1982). Recent developments in structural equation modeling. Journal of Marketing Research, 19(4), 404416. Kaynak, E., Tatoglu, E., & Kula, V. (2005). An analysis of the factors affecting the adoption of electronic commerce by SMEs: Evidence from an emerging market. International Marketing Review, 22(6), 623640. Knight, G. A., & Liesch, P. W. (2002). Information internalization in internationalizing the rm. Journal of Business Research, 55(12), 981995. Kumar, N., Stern, L. W., & Anderson, J. C. (1993). Conducting organizational research using key informants. Academy of Management Journal, 36(6), 16331651. Lam, A. (2000). Tacit knowledge, organizational learning and societal institution: An integrated framework. Organization Studies, 21(3), 487513. Leonidou, L. C., & Adams-Florou, A. S. (1999). Types and sources of export information: Insights from the small business. International Small Business Journal, 17(3), 3048. Lescher, J. F. (1995). Online market research: Cost-effective searching of the internet and online databases. New York: Addison-Wesley. Liesch, P. W., & Knight, G. A. (1999). Information internalization and hurdle rates in small and medium enterprise internationalization. Journal of International Business Studies, 30(1), 383394. Ling-yee, L. (2004). An examination of the foreign market knowledge of exporting rms based in the Peoples Republic of China: Its determinants and effect on export intensity. Industrial Marketing Management, 33(7), 561572. Luo, Y., Zhao, J. H., & Du, J. (2005). The internationalization speed of e-commerce companies: An empirical analysis. International Marketing Review, 22(6), 693709. Madsen, T. K. (1987). Empirical export performance studies: A review of contributions and ndings. Advances in International Marketing, 2, 177198. Menon, A., & Varadarajan, P. R. (1992). A model of marketing knowledge use within rms. Journal of Marketing, 56(4), 5371. Moini, A. H. (1995). An inquiry into successful exporting: An empirical investigation using a three-stage model. Journal of Small Business Management, 33(3), 925. Morgan-Thomas, A., & Bridgewater, S. (2004). Internet and exporting: Determinants of success in virtual export channels. International Marketing Review, 21(4/5), 393408. Muthen, B., & Kaplan, D. (1985). A comparison of some methodologies for the factor analysis of non-normal likert variables. British Journal of Mathematical and Statistical Psychology, 38(May), 171180. Nguyen, T. D., & Barrett, N. J. (2006). The adoption of the internet by export rms in transitional markets. Asia Pacic Journal of Marketing and Logistics, 18(1), 2942. Nguyen, T. D., & Nguyen, T. M. T. (2001). Knowledge internalization in the rm and its implications for business education. Economic Development, 11(December), 3133 (in Vietnamese). Nonaka, I. (1991). The knowledge-creating company. Harvard Business Review, 69(6), 96104. Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 1437. Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company: How Japanese companies create the dynamic of innovation. New York: Oxford University Press. Penrose, T. E. (1959). The theory of the growth of the rm. Oxford: Basil Blackwell. Porter, M. E., & Millar, V. E. (1985). How information gives you competitive advantage. Harvard Business Review, 63(4), 149160.
390
Quelch, J. A., & Klein, L. R. (1996). The Internet and international marketing. Sloan Management Review, 37(3), 6075. Reedy, J., & Schullo, S. (2004). Electronic marketing: Integrating electronic resources into the marketing process (2nd ed.). Ohio: Thomson. Rialp, A., & Rialp, J. (2001). Conceptual frameworks on SMEs internationalization: Past, present and future trends of research. Advances in International Marketing, 11, 4978. Samiee, S. (1998). Exporting and the internet: A conceptual perspective. International Marketing Review, 15(5), 413426. Sheth, J. N., & Sharma, A. (2005). International e-marketing: Opportunities and issues. International Marketing Review, 22(6), 611622. Sinkula, J. M. (1994). Market information processing and organizational learning. Journal of Marketing, 58(1), 3545. Sinkula, J. M., Baker, W. E., & Noordewier, T. G. (1997). A framework for market-based organizational learning: Linking values, knowledge and behavior. Journal of the Academy of Marketing Science, 25(4), 305318. Slater, S. F., & Narver, J. C. (1995). Market orientation and the learning organization. Journal of Marketing, 59(3), 6374. Srensen, O. J., & Buatsi, S. (2002). Internet and exporting: The case of Ghana. Journal of Business and Industrial Marketing, 17(6), 481500. Steenkamp, J.-B. E. M., & van Trijp, H. C. M. (1991). The use of LISREL in validating marketing constructs. International Journal of Research in Marketing, 8(4), 283299. Stump, L. R., Athaide, G. A., & Axinn, C. N. (1998). The contingent effect of the dimensions of export commitment on exporting nancial performance: An empirical investigation. Journal Global Marketing, 12(1), 725. Teo, H.-H., Wang, X., Wei, K.-K., Sia, C.-L., & Lee, M. K. O. (2005). Organizational learning capacity and attitudes toward complex technological innovations: An empirical study. Journal of the American Society for Information Science & Technology, 57(2), 264279. Tsang, E. W. K. (2005). Inuences on foreign ownership level and entry mode choice in Vietnam. International Business Review, 14(4), 441463. Webb, B., & Sayer, R. (1998). Benchmarking small companies on the internet. Long Range Planning, 31(6), 815827. Weible, R., & Wallace, J. (2001). The impact of the internet on data collection. In: P. Richardson (Ed.), Internet marketing (pp. 274281). New York: McGraw-Hill. Wernerfelt, B. (1984). A resource-based view of the rm. Strategic Management Journal, 5(2), 171180.
391
APPENDIX: CFA FACTOR LOADINGS OF ITEMS (STANDARDIZED)

Items Loading t-value
Learning orientation (second-order, 5-point Likert scale) Commitment to learning Managers basically agree that our rms 0.75 ability to learn is the key to our competitive advantage The basic values of our rm include 0.77 learning as a key to improvement In our rm, employee learning is an 0.74 investment, not an expense Learning in our rm is seen as a key 0.76 commodity necessary to guarantee organizational survival Shared vision There is a commonality of purpose in our rm There is total agreement on our organizational vision across all levels, functions, and divisions All employees are committed to the goals of our rm Employees view themselves as partners in charting the direction of our rm Open mindedness We are not afraid to reect critically on the shared assumptions we have made about our markets Personnel in our rm realize that the very way they perceive the marketplace must be continually questioned We often collectively question our own biases about the way we interpret market information 0.70 0.81
12.63 12.17 12.58
12.05
0.78 0.65
11.76 10.06
0.71
0.86
12.21
0.74
11.38
392
Internet information internalization (second-order, 5-point Likert scale) Information transformation Compare foreign market information 0.66 10.97 obtained from the Internet with information collected from other sources Interpret the foreign market information 0.71 11.76 obtained from the Internet to discover its implication for decision-making Spend time sharing foreign market 0.67 11.12 information obtained from the Internet Spend time discussing about foreign 0.70 11.56 market information obtained form the Internet Top management spend time discussing 0.77 with staff involved in international business activities about foreign market information obtained from the Internet Information use Decisions based on information obtained from the Internet are more accurate than wholly intuitive ones Condence in making international business decisions increases as a result of information obtained from the Internet Uncertainty associated with international business activities is greatly reduced by information obtained from the Internet Information obtained from the Internet is used to keep updated with international business knowledge Information obtained from the Internet plays an important role in making international business decisions 0.74 12.15
0.79
12.82
0.74
12.12
0.79
12.91
0.72
393
Internet information relevance (second-order, 5-point Likert scale) Market feasibility information: The Internet can provide information about Potential distributors of products 0.81 Potential buyers of products 0.83 14.63 Potential suppliers of raw materials 0.78 13.97 Adaptation information: The Internet can provide information about Competitors 0.86 Buyers preference 0.88 20.36 Market size 0.88 20.37 Market growth 0.88 20.41 Price trends 0.86 19.87 Background information: The Internet can provide information about Exchange rate uctuations Legal requirements for entry Potential barriers to entry Social/political background Economic background Transport infrastructure Internet utilization Hours per week the rm uses the Internet to search for foreign market information Times per week the rm receives and sends e-mail related to international business activities 0.70 0.71 0.80 0.85 0.79 0.75 0.69 11.56 12.91 13.52 12.70 12.16 7.79
0.73
Foreign sales intensity (10-point scale: 1p10%, 2 1120%, y , 10 91 100%) 0.92 Percentage of the rms foreign sales compared to total sales for the last nancial year International orientation (second-order, 5-point Likert scale) International conviction International business activities are an 0.77 essential part contributing to our rms competitive advantage
14.60
394
Items International business activities enhance our rms competitiveness by acquiring market knowledge International business activities increase the prestige of our rm in the Vietnamese market International business activities contribute to our rms long-term expansion International business activities represent an opportunity for our rm to exploit an expanded market International business activities reduce our rms risks by selling to diverse markets International intention Formal planning is a necessity for international business activities International business activities are a necessity whether or not Vietnamese market sales come up to expectation International business activities should be considered whether or not opportunities in Vietnam are completely exhausted International business activities are so important to the national interest that every Vietnamese rm should commit its resources to the international business drive Loading 0.81 t-value 15.57
0.80
15.30
0.86
16.73
0.80
15.39
0.79
0.80 0.83
13.67 14.16
0.81
13.83
0.74
HOW ADVANCED ARE WEBSITES OF SME EXPORTERS? AN INVESTIGATION INTO DRIVERS AND INHIBITORS
Heidi Winklhofer, Kathryn Houghton and Thomas Chesney
ABSTRACT
Despite the much publicised advantages of a website for SME exporters, the level of website sophistication, as well as the factors which inhibit or stimulate exporting SMEs to develop their website beyond a basic level of sophistication, are still unknown. The literature is prone to discuss website establishment and development simultaneously, splitting rms into adopters and non-adopters, yet websites may be established and then neglected, or be continually developed. This paper introduces an instrument for measuring website sophistication within an export marketing context, and proposes and empirically tests a model that depicts factors impacting on perceived advantages of a website and website sophistication levels. The results identify export diversity and environmental pressure as key determinants of perceived advantage of a website which in turn is a good predictor of website sophistication. The rm internal resources, i.e. Information and Communication Technology (ICT) knowledge and time, in
395
396
HEIDI WINKLHOFER ET AL.
conjunction with entrepreneurship orientation also determine an SME exporters website sophistication level.
1. INTRODUCTION
The benets of the Internet have been much publicised, and are deemed to be particularly relevant for exporting SMEs (Samiee, 1998; Ritchie & Brindley, 2000; Prasad, Ramamurthy, & Naidu, 2001). For instance, it has been claimed that the Internet will accelerate the internationalisation of small- and medium-sized enterprises (Quelch & Klein, 1996). Kotler (2000, p. 670) has suggested that it will create a level playing eld for SMEs in relation to their larger competitors, as it reduces the traditional importance of scale economies which makes global advertising more affordable and extends smaller rms market reach globally. Although not specically focusing on exporting, a recent survey of 25,000 UK, German and US SMEs (with websites) conrms the benets of websites for SMEs, with 94% believing that their site had contributed to a growth in their business in the 12 months following its launch (1&1 Internet Limited, 2004). The report concluded that websites have contributed to increases in small business sales revenue, growth, effective marketing and better communication (1&1 Internet Limited, 2004). With a few exceptions, previous studies have taken Internet adoption as a dichotomous variable, splitting rms into adopters and non-adopters (e.g. Premkumar & Roberts, 1999; DTI, 2005), or assessed rms intention to adopt. This ignores the Internets extent of use in the organisation, as website adoption is a process in which sophistication can be increased (Daniel, Wilson, & Myers, 2002). Alternatively, some writers have investigated adoption as a series of stages of maturity (e.g. Thelwell, 2000; Willcocks, Sauer, & Associates, 2000) through which a business must progress to achieve whatever level of sophistication is required. Levy and Powell (2003), on the other hand, found little evidence that such a stages of growth model is appropriate. In general, the literature is prone to discuss website establishment and development simultaneously, yet websites may be established, and then neglected or continually developed. In line with previous work focusing on website sophistication, we dene website sophistication as the variety of relevant website features included on a website, whereby more features equate to higher sophistication levels (see Karakaya & Khalil, 2004). A growing body of literature highlights the range
How Advanced are Websites of SME Exporters?
397
of features available to website developers and especially rms operating in international markets (e.g. the use of foreign languages (Mousley & Simintiras, 2001)). While it has been stressed that the level of adoption is important (Palvia, Means, & Jackson, 1994; Karahanna, Straub, & Chervany, 1999), the few empirical studies undertaken in this area (e.g. Dou, Nielsen, & Tan, 2002) are mainly descriptive and there is a notable lack of knowledge on website sophistication, post-adoption. Although descriptive reports on website sophistication are useful, they fail to provide conceptual understanding of the factors inhibiting or driving further developments. Commercial research in this area concluded that perceived cost is a key obstacle (1&1 Internet Limited, 2005). In this context, research among UK, German and US SMEs showed that 65% spend less than 300 a year on creating and maintaining their website (1&1 Internet Limited, 2004). While the current study includes cost as a potential barrier, the literature presented later on in this paper would suggest that further development of a website is inuenced by a range of factors. Also, SMEs have so far been treated as a homogenous entity when in practice they are not (Westhead, Howorth, & Cowling, 2002), and website relevance may vary with different types of SMEs, and as a function of export activity. Despite the much publicised advantages of a website for SME exporters, the level of website sophistication, as well as the factors which inhibit or stimulate exporting SMEs to actively use their website as an export marketing tool, and develop it beyond a basic level of sophistication are still unknown. The current paper makes two contributions to work in this area. Firstly it proposes, based on existing literature, an instrument for measuring website sophistication within an export marketing context. This allows for a more comprehensive measure of sophistication, made up of a combination of different website elements, than merely classifying organisations as adopters or non-adopters, or as a group (e.g. website maturity stage 1). Secondly, it proposes, based on a literature review and interviews with SME exporters, a series of factors that may determine their websites sophistication. These factors are empirically tested and the results discussed. The paper continues with an overview of the literature on website sophistication, followed by a proposed conceptual model which incorporates the factors suggested to impact on website sophistication. Next, we introduce measurements and the data collection process. The analysis starts with a description of the popularity of the various website features among SME exporters and then tests the conceptual model using LISREL. The paper concludes with a discussion of the results and some avenues for future research.
398
2. WEBSITE SOPHISTICATION
This study examines the factors that impact on website sophistication in an exporting context. While several instruments measuring website sophistication were examined, that differentiate between various uses of the Internet (e.g. Levy & Powell, 2003; Karakaya & Khalil, 2004), to the authors best knowledge, no instrument is currently available that focuses specically on website sophistication within the context of export marketing. This context deserves its own instrument as the needs of export customers and overseas intermediaries will require different website features than websites of a rm operating only in the domestic market. The most obvious difference is that the website will probably need to be available in other languages (Dou et al., 2002), and may also have to serve the needs of local agents and distributors (Ghose & Dou, 1998). The following briey reviews the generic instruments discussed in the literature and then introduces the website sophistication index for SME exporters. Levy and Powells (2003) work, based on 12 case studies, differentiates between various uses of the Internet within the rm. They distinguish between e-mail, website with brochureware, website used to interact with customers and website used for e-commerce with customers. Similarly, in their study of factors related to Internet adoption, Karakaya and Khalils (2004) survey examined e-mail use, website sophistication, Internet use for marketing support and Internet use for marketing intelligence. To measure website sophistication they used 10 elements, the absence or presence of each made up the website sophistication score. Their list was: company information, product/service information, company nancial data, links to other companies, online transactions, stock/inventory status, online catalogue, customer service area, advertisements of other company products and advertisement of own company products/services. Although many of these features can be of use to SMEs as a general marketing tool, they lack the specic focus of a website as an export marketing tool. The instrument for measuring website sophistication of exporters presented here is specic to marketing and to exporters. Although it is recognised that website sophistication can be attributed to technical adeptness such as ease of navigation and quality of links (Loiacono, Watson, & Goodhue, 2002; Bauer & Scharl, 2000), as this study is concerned with the marketing potential of a website, technical elements are only included when they will affect the marketing capability of the site. Likewise, e-commerce activities were excluded. While a website that allows customers to buy products will be considered more sophisticated than one that does
399
not, e-commerce activities fall outside of the scope of this study as the factors driving e-commerce are likely to be different from those impacting on website sophistication. The 17 elements that make up the website sophistication score for this study are shown in Table 1. To ensure the features selected were relevant to SME exporters, the instrument to measure sophistication came from a critical review of relevant literature and an analysis of features that exporting SMEs currently have on their websites. A great deal of normative advice is available in this area directed specically at SMEs (e.g. Lynn, Lipp, Akgun, & Cortez, 2002; Leong, Ewing, & Pitt, 2002; Thelwall, 2000; Chaffey, Mayer, Johnston, & Ellis-Chadwick, 2000), but, as mentioned earlier, empirical studies are scarce. The elements included in our sophistication index are based on Karakaya and Khalils (2004) sophistication measure which was described earlier. They were adapted and expanded to specically examine marketing practices of exporting SMEs. Table 1 details the origin of the elements. Although Karakaya and Khalil (2004) included company nancial data as an element in the sophistication index, we felt that this is inappropriate for the context of SMEs, as they are usually not prepared to disclose such information. We also excluded advertisement of own and other company products as the product/service range was already covered by three other items. Finally, online transactions was excluded, as justied earlier. In turn, we added elements which were discussed in the literature as components of website marketing: Thelwall (2000) argues that websites offer SMEs the opportunity to adopt a pull strategy which requires high visibility through search engines. Consequently, we included website details are regularly submitted to search engines as a further component of the sophistication index. Visibility or awareness efciency can be measured through website hits, which if used correctly, provide a wealth of information for marketing and prospecting (Sterne, 1999). Thus we included visitor information is used for marketing purposes. Finally, the instrument captures whether the website has marketing content specically designed for it and whether the website content is regularly updated, as both have been suggested to improve the marketing communication role of a website (Bickerton et al., 1998). To capture the export-specic nature of this study, we included website is written in languages other than English. We also added whether the website domain has been registered other than as co.uk, as this will impact on users perceptions of the site. Based on Ghose and Dous (1998) work, we expanded the element links to other companies (Karakaya & Khalil, 2004) and included instead, websites has links to export distributors/agents/ retailers/wholesalers websites and websites displays contact details of
400
Table 1.
Element
Elements Comprising the Website Sophistication Index.

This Study Karakaya and Khalil (2004) Company information Examples of Other Supporting References Bickerton, Bickerton, and Simpson-Holley (1998) Strauss and Frost (1999) Webb (2002) Quelch and Klein (1996)
2 3 4
10
11
Website includes recent press releases about our company Website includes an online product catalogue Website displays product prices Visitors can download demonstrations of the product or product manuals Website allows visitors to check the availability of our products Website has links to export distributors/agents/ retailers/wholesalers website(s) Website displays contact details of export distributors/agents/ retailers/ wholesalers Website provides ways for export customers to contact the rm directly (for example by email or telephone) Website has segregated password protected areas for export customers Website has segregated password protected areas for export distributors/ agents/retailers/ wholesalers Website has a feedback form(s) for customers to complete Not included Not included Not included
Online catalogue Product/service information
Stock/inventory status
Quelch and Klein (1996)
Links to other companies
Ghose and Dou (1998)
Customer service area
Ghose and Dou (1998)
Dutta and Biren (2001)
Not included
Company nancial data Advertisements of other company products Advertisement of own company products/ services Online transactions
401
Element This Study Karakaya and Khalil (2004) Not included Examples of Other Supporting References Bickerton et al. (1998)
12
13 14
15
16
17
Website has marketing content specically designed for it The website content is regularly updated Website details are regularly submitted to search engines Visitor information (for example hit rates) is used for marketing purposes Website is written in languages other than English Website domain(s) other than .com or .co.uk have been registered (for example, .je or .de)
Not included Not included
Bickerton et al. (1998) Thelwall (2000)
Not included
Sterne (1999)
Not included
Dou et al. (2002)
Not included
Dou et al. (2002)
export distributors/agents/retailers/wholesalers. Furthermore, since password-protected areas can aid communication ows between an exporter and an intermediary (Dutta & Biren, 2001), we included whether such passwordprotected areas are installed. The sophistication index is intended to provide a more comprehensive measure of adoption than employed in previous studies on website and Internet use of SMEs (e.g. Bennett, 1997; Premkumar & Roberts, 1999; DTI, 2005). The 17 elements should be viewed as causing website sophistication rather than being caused by it; thus they act as causal indicators (MacCallum & Browne, 1993, p. 533). Website sophistication, the latent variable, is therefore dened as a linear function of the indicators, plus a disturbance term1 (MacCallum & Browne, 1993). Accordingly, and in line with previous work on sophistication indices (Karakaya & Khalil, 2004), to obtain an overall sophistication index score, all 17 elements are equally weighted and summed. Although one could argue that several themes (e.g. communication) are captured by the index, for the purpose of this study, an overall index is used to provide some initial insights into what type of rms have invested in a larger number of website features and consequently score higher on the sophistication index. All but one of the elements in Table 1 (element 15) are objective and the sophistication score could be calculated largely without
402
the rms cooperation, although their input will probably be needed to answer whether their website has marketing content specically designed for it (element 12) and how often the content is updated (element 13).
3. CONCEPTUALISATION
3.1. Adoption Models As outlined earlier, sophistication levels of post-adoption and their antecedents are somewhat neglected. However, from a company perspective, each individual feature included on a website could be regarded as an adoption decision. Against this background we will briey review the key conceptual models within the IT adoption literature in order to assess their suitability to predict website sophistication among exporting SMEs. Conceptually, research on IT adoption can draw on the Technology Acceptance Model (TAM) (Davis, 1989) and Diffusion of Innovation Theory (Rogers, 1995). TAM, originally developed by Davis (1989) to predict user acceptance of computer technology in the workplace, has the advantage of being well grounded in established social psychology theory. It is based on the Theory of Reasoned Action (TRA) (Ajzen & Fishbein, 1980), and is described as an analytical simplication of how functionality and interface characteristics relate to adoption decisions (Deng, Doll, Hendrickson, & Scazzero, 2005, p. 746). The main determinants of technology acceptance behaviours are the belief constructs: perceived usefulness (i.e. the prospective users subjective probability that using a specic application system will increase his or her job performance within an organizational context) and perceived ease of use (i.e. the degree to which the prospective user expects the target system to be free of effort) (Davis, 1989, p. 985). Key linkages are specied between these two key belief constructs and users attitudes, intentions and adoption behaviour. The effects of external variables such as individual differences or situational constraints are also expected to impact on user acceptance only as far as they are mediated by the two key belief constructs of perceived usefulness and perceived ease of use. Rogers (1962) conceptual model links the belief constructs: relative advantage, compatibility, diversity, trialability and observability to adoption decision. Thus, an individual, or organisation, forms a judgment about an innovation based on these perceived characteristics. Several authors have used the Rogers model and amended, or complimented it with additional constructs (e.g. Moore & Benbasat, 1994; Plouffe, Vandenbosch, & Hulland, 2001; Premkumar & Roberts, 1999; Thong, 1999).
403
These models are based on well-researched attitude and/or believe structures and have been shown to explain adoption as a dichotomous variable (yes/no), or intention to adopt (e.g. Plouffe et al., 2001). Regarding postadoption, there is evidence that the importance of widely used beliefs and attitudes towards usage of IT, change from pre- to post-adoption (Karahanna et al., 1999). Although beliefs and attitudes about the innovation are good predictors of adoption or intention to adopt, generic rm characteristics relating to a variety of resources (e.g. size; IT knowledge, top management support) have also been found to explain adoption of IT within organisations. A brief summary of the factors studied is provided in the following section, and distinguishes between internal and external factors. Among the internal factors inuencing adoption, past research included company characteristics, such as rm size (Dandridge & Levenbury, 2000; Premkumar & Roberts, 1999), company age (Dandridge & Levenbury, 2000), staff workload (Corbitt, 2000; Cragg & King, 1993), nancial support restrictions (Premkumar & Roberts, 1999), IT knowledge and exposure (Premkumar & Roberts, 1999; Thong, 1999; Hamill & Gregory, 1997; Corbitt, 2000), organisational readiness (Lee & Cheung, 2004; Mehrtens, Cragg, & Mills, 2001) and top management support (Premkumar & Roberts, 1999; Sultan & Chan, 2000; Cragg & King, 1993). In terms of external factors the key predictor variables studied were external Information and Communication Technology (ICT) support (Premkumar & Roberts, 1999; Igbaria & Zinatelli, 1997), competition-related factors in ICT supply (Gatignon & Robertson, 1989; Frambach, Barkema, Nooteboom, & Wedel, 1998), government inuence (Corbitt, 2000), external pressure (Premkumar & Roberts, 1999; Gatignon & Robertson, 1989; Lee & Cheung, 2004; Mehrtens et al., 2001) and technology readiness of other parties (Holmes & Srivastava, 1999). Despite the rich empirical literature on Internet and website adoption, there are several weaknesses: rstly, it is not specically applicable to exporters, and secondly it deals overwhelmingly with adoption. Although adoption of a website is a prerequisite for our study, the focus is on further development of it (i.e. website sophistication level). In summary, the literature review resulted in an unparsimonious set of factors. 3.2. Interview Phase To gain further insight into the factors stimulating or hindering website sophistication, we conducted qualitative longitudinal interviews with multiple respondents from 25 SME exporters. A longitudinal approach
404
was regarded necessary as website development is a dynamic process and a cross-sectional approach would be strongly inuenced by informants memory of events rather than capturing events as they occur. Therefore, for the purpose of theory development we conducted between two and three semistructured interviews over a period of 12 months, resulting in a total of 51 interviews. The interviewees were owner/managers and senior managers, or whoever the companies regarded as most knowledgeable concerning the rms website development. It should be mentioned that this study was part of a larger project that also examined e-commerce adoption within SME exporters. As part of the semi-structured nature of the interviews, an interview guide was developed which covered a series of questions pertaining to a companys website development. The key focus was on establishing how it has changed over time and for what reason; identifying who was involved in development of the website and what role it plays in export marketing activities. The interviews lasted from 35 minutes to 2 hours and were taperecorded and transcribed. Within-case followed by cross-case analysis was performed. Within-case analysis began through coding of the transcripts using a combination of codes developed prior to the analysis, using the literature and codes drawn from the interviews (Miles & Huberman, 1994). To facilitate the analysis we used data displays, which compress information and aid the analyst in drawing warranted conclusions (Miles & Huberman, 1994). This was followed by cross-case analysis which enables deeper understanding and explanation than that derived from studying isolated cases (Miles & Huberman, 1994). The objective was to identify drivers and obstacles of website development. For this purpose, we used the standardised displays derived from the within-case analysis in conjunction with the original transcripts. The following conceptualisation draws together the previously reviewed literature and ndings from the interviews. We propose a two-stage model, with a rms export prole and its external environment impacting on their perception of a websites relative advantage. This in turn, in conjunction with resource constraints, and general entrepreneurship orientation, is linked to website sophistication. The model is shown in Fig. 1 and is detailed and justied in the following sections. 3.3. Factors Impacting on a Websites Perceived Relative Advantage The interviews highlighted the importance of export characteristics. Export enthusiasm, i.e. the managements view that exporting is a worthwhile endeavour and the desire to enter new export markets (Diamantopoulos,
405
(+) Export characteristics: Export enthusiasm (H1) Export dependence (H2) Export diversity (H3) (+) External: Environmental pressure (H4)
Perceived relative advantage of website
(H11)(+)
Resources-drivers: ICT knowledge (H6) Personal contact network (H7) Perceived awareness and use of supplier services (H8) Resources obstacles: Staff workload (H9) Perceived cost of website development (H10)
(+)
(-)
Website sophistication
General firm characteristics:

Entrepreneurship orientation (H5)
(+)
Time since website established
(+)
Schlegelmilch, & Allpress, 1990), emerged as a potential driving factor. Interview data indicated that new export expansion programmes served as an impetus for further website development to better suit export markets. For example, a rm that has seen its export sales decline from 40% to 5% over the last two years has launched an export expansion programme which also included revamping the website.
I think the entire site is more internationally avoured now, rather than just a couple of pages on the site y a telecommunications case study for Abu [Dhabi] and one for Dubai. Brewing can be offensive in the Middle East obviously, so we now have a water case study instead.
Not surprisingly, a common theme emerging from the interviews was how the website facilitated and simplied communication for rms operating with export customers and intermediaries dispersed around the globe (see also Samiee, 1998; Ritchie & Brindley, 2000; Prasad et al., 2001). In line with this argument, we found some indication that exporters more dependent on export sales, appear to be more enthusiastic about the benets offered by their website. Illustrative of this attitude is the following quote from a
406
Managing Director whose company is highly enthusiastic and dependent on export sales:
At this stage all of our resources, all of our prot has gone into the development of the product itself the Internet site we have is very basic. Weve already made the decision that we need a professional Website developer. Thats going to be one of the next things on the marketing strategy y
This led to the rst three hypotheses. H1. Export enthusiasm is positively related to a websites perceived relative advantage. H2. Export dependence is positively related to a websites perceived relative advantage. H3. Export diversity is positively related to a websites perceived relative advantage. With regards to company external factors, environmental pressure, resulting from competitors actions and customers demands, as well as a sense of fatalism, especially among consumer-goods rms, acted as a strong incentive to further develop the website, which is in line with the work by Mehrtens et al. (2001). Several participating rms emphasised the increasing expectations of their customers. For example, one interviewee after discussing several features on their website, mentioned:
some companies already have that. So we are playing catch up with some of our competitors y . People we work with expect us to have a certain level of presence.
A marketing manager highlighted:

I think [a website] is actually fundamental. I think weve taken a very sort of active approach to the Internet. We started off quite early on y . Which was really just doing a bit of advertising on the Internet about the company. And I would say that was primarily driven from the export markets. Not so much from the UK. Because what had happened over the years, [company], X is a medium sized business but when you actually go out selling international projects, companies that you are selling to, their perception of you is quite different, particularly if theyve never actually been to your home base. And our competitors in the marketplace are big companies like [names of competitors] and we were seen in the Middle East and certain parts of the Far East as being exactly the same size as they were. And we were getting comments from our export guys well these companies have got websites so you know we need a website because people need to have the same sort of feel. So to a certain extent although we were keen on driving it there was quite a push from the export market place. Probably more so than from the UK and I think that was quite important. The website is a very good communication tool, its very efcient. Its very effective. It actually means that you can get immediate responses.
407
This illustrates that customers expect to communicate with their suppliers via their website, and consequently such rms are likely to perceive their website as more advantageous. We therefore propose the following: H4. Environmental pressure is positively related to a websites perceived relative advantage. 3.4. Factors Impacting on Website Sophistication 3.4.1. Entrepreneurship Orientation Regarding general company characteristics, previous work (Poon & Swatman, 1997) suggested entrepreneurship orientation as the factor distinguishing between more reactive and proactive adoption of the full potential of Internet technology. Indications of this were also found in the interviews. Firms with a high entrepreneurship orientation are innovative, willing to take risks, are proactive and are aggressive in competition (Lumpkin & Dess, 1996). Entrepreneurship orientation encompasses innovativeness, which in this context refers to the degree to which an individual or group is early in adopting new ideas relative to the other members of a system (Rogers, 1995). It also captures the need to take risks in resource allocation when developing a website, where return on investment is uncertain. As expected, entrepreneurship was exhibited within most participants website activity. In line with this, we suggest a positive link between level of entrepreneurship orientation and website sophistication. H5. The level of website sophistication of SME exporters is positively related to the rms level of entrepreneurship orientation. 3.4.2. Company Resources Company resources appear to be a key determinant of website sophistication levels and their presence or absence can act as stimulant of obstacle respectively. Stimulating company resources seem to be (1) ICT knowledge, (2) personal contact network and (3) perceived awareness and use of supplier service. In-house ICT knowledge was seen as pivotal in further developing the website by the rms sampled, which conrms ndings in related adoption studies (e.g. Hamill & Gregory, 1997; Premkumar & Roberts, 1999; Thong, 1999; Hamill & Gregory, 1997; Corbitt, 2000). This was partly due to lack of resources to pay external parties and partly because owner managers have been shown to resist external assistance because of a reluctance to compromise their independence (e.g. Curran & Blackburn, 2001; Bennett & Robson, 1999). Firms sampled also suggested that a lack of resources, such
408
as ICT knowledge, can be compensated for or supplemented by personal contact networks, which include people who help the SME remain committed to the use and development of the innovation (McGowan & Durkin, 2002; Carson, Cromie, McGowan, & Hill, 1995). Illustrative of this is the following quote of a marketing manager of a highly experienced exporting rm with 100 employees:
My assistant will be taking over the website development y and a lot of his friends have designed their own websites. All their ideas are kind of pooled. So I think he has consulted but not on a professional level.
We therefore propose that the extent of the personal contact network is likely to positively impact on website sophistication. We also found support for indications that website development could be linked to the level of awareness of the services offered by technology vendors and website developers (i.e. level of perceived supplier service, see also Igbaria and Zinatelli, 1997). Several of the interviewees appeared to have only a vague understanding of what is available in terms of technical support and incentives for SMEs in terms of website development which appeared to be negatively reected in their website. Although the SMEs interviewed had plans to add further features to their site or change its current format, work commitments meant that they had to prioritise, and other activities were seen as more important than website development. Thus, staff workload was seen as a major impediment in developing the website (see also Corbitt, 2000; Cragg & King, 1993). Finally, if in-house expertise is not available, the perceived cost of additional features appeared to act as an impediment to further developments (see also 1&1 Internet Limited, 2005; Premkumar & Roberts, 1999), given that SMEs, by their nature, are resource constrained (Westhead et al., 2002). We therefore propose: H6. The level of website sophistication of SME exporters is positively related to the rms level of ICT knowledge. H7. The level of website sophistication of SME exporters is positively related to the rms personal contact network. H8. The level of website sophistication of SME exporters is positively related to the level of perceived supplier service. H9. The level of website sophistication of SME exporters is negatively related to the rms staff workload.
409
H10. The level of website sophistication of SME exporters is negatively related to the perceived cost of website development. In line with the ndings on IT and Internet adoption (Premkumar & Roberts, 1999; Thong, 1999; Rogers, 1995), SME exporters appeared to be highly selective in their website construction. There was a need to see a tangible relative advantage (compared to traditional communication channels) in spending more effort on their website. Thus, perceived relative advantage is likely to impact on website sophistication. In other words, SMEs are more likely to use their website for a certain function if it gives them an advantage, relative to some other means of achieving the same function. For example, some participants considered an up-to-date website with an online product catalogue to be superior to distributing paper catalogues. Another participant developed a password-protected area from which authorised parties can download high-quality product images, the relative advantage over printing and posting the images being savings in time and money. We therefore propose: H11. The level of website sophistication of SME exporters is positively related to its perceived relative advantage. Given the longitudinal nature of the interviews, we observed that many of the case study rms had developed their websites further over time. We therefore included time since website established as a control variable. Fig. 1 depicts the linkages proposed. To summarise, the extensive interview process partly conrmed the potential relevance of factors identied in more general IT or Internet-related adoption research. On the other hand, it also demonstrated that TAM and Innovation Diffusion Theory are not well suited to capture website sophistication. Apart from perceived relative advantage (Rogers, 1962, 1995), the other belief constructs or attitudes appear to be less critical in the decision to further develop the website, i.e. add extra features to a website (see also Premkumar & Roberts, 1999). It seems that once the decision of adopting a website has been made, the key focus is on the perceived advantages of the website (which is related to perceived usefulness in TAM). Several general rm characteristics previously shown to impact on ICT adoption, such as rm size, had little impact when focusing on SME exporters. Moreover, government initiatives did not appear to be of any relevance. The interviews further highlighted additional factors such as: export enthusiasm and personal contact networks. The qualitative work also suggested replacing top management support (e.g. Premkumar & Roberts, 1999) with
410
entrepreneurship orientation. The following section will outline the methodology employed.
4. METHODOLOGY
4.1. Measures Employed The measurements of the constructs discussed above were mainly based on adaptations of previously developed scales (see Appendix A). ICT knowledge comprises a ve-item scale by Grewal, Comer, and Mehta (2001), while Entrepreneurship orientation is based on a nine-item rened measure reported in Covin (1991), Premkumar and Roberts (1999) measurements were adapted for the constructs environmental pressure (three items), and the innovation characteristic perceived relative advantage (four items). To capture a rms perception of supplier service (three items), as well as to measure perceived cost of website development (two items), adaptations of the scales reported in Premkumar and Roberts (1999) were used. Export enthusiasm is an adaptation of the scale by Diamantopoulos et al. (1990) (six items). In line with previous work, we used the number of export regions as an indicator of export diversity (e.g. Schlegelmilch, Diamantopoulos, & Tse, 1993), and percentage of sales derived through exports as a measure of export dependence (e.g. Souchon et al., 2003). Measurements for the personal contact network and staff workload constructs were developed based on the qualitative interviews preceding the survey. The degree to which people rely on friends, family and for website development is represented by three items reecting the new construct personal contact network. The construct staff workload was captured by two items relating to the difculty of nding time to work on website development. The dependent variable, website sophistication was captured using the measure discussed in Section 2 (see Table 1). The respondents were presented with the items listed in Table 1 and asked to tick all those that applied to them. As was stated earlier, all features were equally weighted and summed to give their website sophistication score. 4.2. Data Collection The measures were incorporated into a mail questionnaire, which was extensively pre-tested. For the main survey, a random sample of 1,000 UK SME exporters in the manufacturing eld was targeted, derived from a Dun
411
and Bradstreet database. We conducted telephone interviews with 100 nonrespondents, which revealed that the main reasons for non-response were ineligibility (e.g. the rm did no longer export) or time pressure. After adjusting for ineligibles, the 130 usable responses represented an effective response rate of 16.5%. We conducted further non-response analysis by comparing early and late responses (see Armstrong & Overton, 1977) on the key variables. T-tests for independent samples revealed no signicant differences (5% signicance level) between the two groups. Consequently, nonresponse error was unlikely to be a problem in this study. The size of the rms ranged from 2 to 250 employees, who had been exporting for an average of 29.5 years. The majority exported to the European Union, followed by North America and Asia. The rms derived an average of 39.7% of their sales from exporting and represented a range of exporting diversity.
5. ANALYSIS AND FINDINGS

5.1. Scale Development and Validation In order to purify the scales measuring the constructs, we used a combination of (1) item inter-correlations, (2) item-total correlations, (3) exploratory factor analysis and (4) conrmatory factor analysis (CFA). Based on this, poorly performing items were identied and eliminated (see Appendix A for a listing of the puried scales). To comply with minimum sample size to parameter ratios (e.g. Bentler & Chou, 1987), we were unable to test the full measurement model with CFA simultaneously. We have therefore broken it down into three sets of CFAs. Set 1 includes export and general rm-specic constructs, i.e. export enthusiasm and entrepreneurship orientation. Set 2 contains resource-specic constructs, i.e. ICT knowledge, personal contact network, perceived awareness and use of supplier service, while Set 3 comprises of perceived relative advantage and environmental pressure. Table 2 details the t indices obtained. The items display adequate convergent validity as each indicators estimated pattern coefcient is signicantly related to its underlying construct (Anderson & Gerbing, 1988). Discriminant validity of the constructs was assessed by constraining the correlation between each pair of constructs (across all three sets) equal to 1 and comparing the w2 values of the original with the constraint model (see Anderson & Gerbing, 1988). The w2 differences obtained exceeded the value of 3.84 (5% critical value for the w2
412
Table 2.
Model Set 1 Set 2 Set 3 w2 (d.f.)
Fit Indices for Measurement Models.

RMSEA 0.082 0.045 0.038 GFI 0.910 0.941 0.967 NNFI 0.941 0.980 0.994 CFI 0.958 0.985 0.996
63.99 (34) p 0.001 40.26 (32) p 0.149 15.47 (13) p 0.279
Notes: Set 1 contains export enthusiasm, entrepreneurship orientation. Set 2 contains ICT knowledge, personal contact network, perceived awareness and use of supplier service. Set 3 contains perceived relative advantage, environmental pressure.
distribution with 1 d.f.) for each pair, consequently the construct measurements demonstrate discriminant validity. Composite reliabilities are above their thresholds of 0.6, (Bagozzi & Yi, 1988) for all but one of the multi-item constructs (see Appendix A). The perceived cost of website development (two items), and staff workload (two items) constructs performed poorly in the CFA and we had to reduce them to single-item measurements. 5.2. Pattern of Website Sophistication Initially, we set out to identify a pattern of website sophistication among the SME exporters. Inspection of the frequency of each feature (Table 3) showed that almost all (98.5%) included e-mail addresses/telephone numbers, so that export customers could contact rms directly. The vast majority of websites also included marketing content that was specically designed for the website. Regular up-dating and on-line product catalogues were available in two-thirds of the respondents websites. Not surprisingly, more resource intensive components such as websites written in languages other than English, and the possibility of checking product availability, were less common. The least prevalent features were segregated passwordprotected areas and the display of prices. On a more aggregated level, the results demonstrate that the inclusion of specic export marketing features is underdeveloped, while website marketing unique features are considerably more common. On average, the respondents reported that their websites include 7 out of 17 features, albeit in varying combinations. 5.3. Model Estimation Owing to a combination of limited sample size and complexity of the model, we calculated single indicants for each multi-item scale by averaging across the items (Bagozzi & Heatherton, 1994; Cadogan, Sundqvist,
413
Table 3. Elements Comprising the Website Sophistication Index Descriptive Analysis.

Element 1 2 3 4 5 6 7 8 Website Sophistication Index Website includes recent press releases about our company Website includes an online product catalogue Website displays product prices Visitors can download demonstrations of the product or product manuals Website allows visitors to check the availability of our products Website has links to export distributors/agents/ retailers/wholesalers website(s) Website displays contact details of export distributors/agents/retailers/ wholesalers Website provides ways for export customers to contact the rm directly (for example by email or telephone) Website has segregated password protected areas for export customers Website has segregated password protected areas for export distributors/ agents/retailers/ wholesalers Website has a feedback form(s) for customers to complete Website has marketing content specically designed for it The website content is regularly updated Website details are regularly submitted to search engines Visitor information (for example hit rates) is used for marketing purposes Website is written in languages other than English Website domain(s) other than .com or .co.uk have been registered (for example, .je or .de) Frequency 61 84 22 54 22 40 53 128 % 46.9 64.6 16.9 41.5 16.9 30.8 40.8 98.5
9 10
6 16
4.6 12.3
11 12 13 14 15 16 17
47 105 84 71 55 22 31
36.2 80.8 64.6 54.6 42.3 16.9 23.8
Notes: Respondents were presented with a list of these elements and asked to tick all those that applied to their website. All elements were equally weighted and summed to give a websites overall sophistication score.
Salminen, & Puumalainen, 2005). Appendix B shows the nal measurement results and the correlation matrix. We employed LISREL 8.54 to simultaneously test all hypotheses. The modelling is based on a covariance matrix + + and maximum likelihood estimation procedure (Joreskog & Sorbom, 1993).
414
The model t indices obtained indicate a good t (w2 13.46, d.f. 11, p 0.263; RMSEA 0.044; GFI 0.984; NNFI 0.951; CFI 0.993). The estimates (Table 4) conrm that perceived environmental pressure, exercised by customers and competitors, is positively linked to perceived relative advantage of a website (H4, g 0.567, po0.05). In terms of export characteristics, only diversity of export markets leads rms to more strongly appreciate the advantages of a website (H3, g 0.230, po0.05), while more export dependent, as well as export enthusiastic rms do not appear to perceive their websites as more advantageous than their less export enthusiastic or dependent counterparts, thus refuting propositions H2 and H1, respectively. The ndings reveal that the key factor positively impacting on website sophistication is perceived relative advantage (H11, b 0.266, po0.05). This is followed by the positive impact of ICT knowledge (H6, g 0.211, po0.05). Somewhat smaller is the negative effect of high levels of staff workload (H9) on website sophistication (g 0.168, po0.05). Website sophistication is also positively related to a rms entrepreneurship orientation (H5, g 0.167, po0.05) as well as the age of the website (g 0.154, po0.01). In summary, with regards to website sophistication, our ndings support propositions H6 (ICT knowledge), H9 (staff workload), H5 (entrepreneurship
Table 4. Standardised Parameter Estimates.

Proposition: Path H1: Export enthusiasm - perceived relative advantage H2: Export dependence - perceived relative advantage H3: Export diversity - perceived relative advantage H4: Environmental pressure - perceived relative advantage H5: Entrepreneurship orientation - website sophistication H6: ICT knowledge - website sophistication H7: Personal contact network - website sophistication H8: Perceived awareness and use of supplier service website sophistication H9: Staff workload - website sophistication H10: Perceived cost of website development - website sophistication H11: Perceived relative advantage - website sophistication Time since website established (control variable) Estimate 0.054 0.132 0.230 0.567 0.167 0.211 0.067 0.000 0.168 0.042 0.266 0.154 t-value 0.645 1.363 2.438 7.343 1.922 2.340 0.861 0.013 2.150 0.541 3.486 1.888 Signicance ns ns 0.05 0.05 0.05 0.05 ns ns 0.05 ns 0.05 0.10
415
orientation), and H11 (perceived relative advantage), and, our control variable, age of a website, also shows a signicant impact. As can be seen from Table 4, of the 11 propositions tested six were supported, explaining 38.6% of the variation in perceived relative advantage of a website and 31.6% in website sophistication. Our results reveal environmental pressure (H4) as the strongest factor impacting on perceived relative advantage of a website. Environmental pressure, an external factor, comprises of competitive pressure and increased customer expectations. Thus, it appears that rms are predominantly reactive in their appreciation of the benets of a website. Our ndings also show that export diversity (H3) has a positive impact on perceived advantages of a website. The strongest resource driver for website sophistication is ICT knowledge (H6). ICT knowledge has already been shown to impact on initial adoption (Hamill & Gregory, 1997; Thong, 1999) but appears to be equally crucial in further development. As anticipated, staff workload was negatively related to website sophistication (H9), conrming ndings from earlier qualitative research (e.g. Corbitt, 2000) which focused on adoption. As expected, entrepreneurship orientation demonstrates a positive effect on website sophistication (H5). Hence, the more innovative, risk taking, autonomous and proactive the company is in their business activities, the higher the level of website sophistication. Against expectations, personal contact networks (H7), perceived supplier service (H8) and perceived cost associated with website development (H10) were not signicant inuences on website sophistication.
6. DISCUSSION AND CONCLUSION

This study has examined website sophistication among SME exporters with the aim of nding key inuencing factors. We have introduced an index of website sophistication and shown that the level of website sophistication varies greatly among SME exporting rms. Hence, there is a need for future studies to employ measures of website adoption which capture the level of sophistication rather than just treat it as a dichotomous variable. This is of particular relevance when considering the effect to which the Internet replaces or supplements existing marketing channels. Moreover, it also appears that website sophistication is only partly an evolutionary process which develops over time. SMEs develop different levels of sophistication depending on how benecial they perceive their website. Characteristics of individuals, particularly owner managers, have an inuence on the SME
416
decision-making process. Where there is a greater degree of entrepreneurialism, SMEs were more likely to risk allocating assets to website development. The measure of entrepreneurialism employed may also capture an increased tendency towards innovation per se. Implicit in sophistication levels is the extent to which websites supplement traditional marketing activities. It is encouraging that website marketing features (e.g. marketing content specically designed for the website) are widely used. Disappointingly, websites are still fairly underdeveloped in terms of export marketing features. This includes websites being written in languages other than English, registration with international website domains, links to websites of export intermediaries and segregated passwordprotected areas for export intermediaries. Only 30.8% of the sample linked their website to those of their export intermediaries, suggesting that many SME exporters continue to operate in traditional ways with intermediaries. Particularly, rms highly dependent on exports or those enthusiastic about exporting, are not necessarily more positive about the perceived advantages of a website. Instead, export diversity appears to be the only export-related factor that impacts on a rms perception of a websites advantages over traditional communication channels. Thus, one could conclude that the potential of a website as an export marketing tool is so far only appreciated by rms with a more widely spread customer base. As found in this study, perceived relative advantage of a website is the key driver for website sophistication, thus it is therefore not surprising that export-specic features were among the least prevalent features. In other words, the potential offered by websites to SME exporters has not yet been fully exploited and it does not appear that websites are considering replacing traditional export marketing channels. The ndings also illustrate the importance of considering both internal and external factors in website sophistication. Different factors provide opposing inhibiting and stimulating factors, which the SME must balance and respond to accordingly. Indeed, meeting external demands against resource constraints is an eternal struggle for many SMEs. However, our ndings show that one needs to differentiate between the types of resources required. Although in-house resources (i.e. staff workload and ICT knowledge) impact on website sophistication, nancial resources (i.e. the perceived cost of further website development) have not been found to hamper further development. This is in contrast to the ndings by commercial research which identied perceived cost a key obstacle (1&1 Internet Limited, 2005). The fact that perceived cost of further developing the website is not seen as an obstacle is intriguing, as lack of ICT knowledge and time can be
417
overcome by outsourcing further website development. In addition, their awareness of potential rms who could undertake such a task (i.e. perceived supplier services) is not an obstacle. However, comments received from the qualitative study highlighted that time to develop websites was very limited. Even respondents who worked with external website developers found it difcult to give time to their information requests, or to consider its development. For many, the priority was to deal with current orders and fulll customers needs. An alternative explanation is that SMEs tend to prefer developing their websites in-house. This is in line with claims that owner managers are reluctant to compromise their independence and therefore resist external assistance (Curran & Blackburn, 2001; Bennett & Robson, 1999). Unlike in the interview phase, personal contacts do not appear to be inuential in further developing a website. Work undertaken by personal contacts would overcome resource constraints and unwanted involvement of ICT suppliers. It is therefore doubtful that personal contacts are capable of developing a website beyond a basic level of sophistication. The implications of this research are that policy makers should be aware that export active SMEs vary greatly in their appreciation of a website as a marketing tool which is reected in their websites sophistication levels. This is partly due to the variety of markets the SMEs are operating in as well as the pressure exerted by their competitors and the expectations of their customers. In terms of implications for SMEs, rms should look at ways of easing the time constraints to developing this media. Indeed, a number of SME owners interviewed developed their rms websites in their spare time. Time constrained SMEs should be aware of taking a short-sighted view because they are busy satisfying the demands of a full order book. Research in this area is still in the early stages, however the present study highlighted that traditional adoption models (i.e. TAM and Innovation Diffusion Theory) are only partly suited for predicting post-adoption sophistication levels. Thus, there are many fruitful avenues for future work, including the impact or inuence of marketing channels or relationships with intermediaries. A particularly fruitful avenue for further research would be a study that links exporters website sophistication to export performance. Clearly website sophistication is a moving target and even those who score highly against the criteria used here now, may nd that as technology develops, they slip behind their competitors. Future research may wish to examine how the sophistication level changes across time and as technology advances.
418
NOTES
1. It should be noted that the instrument for measuring website sophistication does constitute a formative measure (Bollen & Lennox, 1991) and as such conventional tests for reliability and validity are not suitable (Diamantopoulos & Winklhofer, 2001).
REFERENCES
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Prentice Hall: Engelwood Cliffs, NJ. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411423. Armstrong, J. S., & Overton, T. S. (1977). Estimating non-response bias in mail surveys. Journal of Marketing Research, 14, 396402. Bagozzi, R. P., & Heatherton, T. F. (1994). A general approach to representing multifaceter personality constructs; application to state self-esteem. Structural Equation Modelling, 1(1), 3567. Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16(1), 7494. Bauer, C., & Scharl, A. (2000). Quantitative evaluation of website content and structure. Internet Research: Electronic Networking Applications and Policy, 10(1), 3143. Bennett, R. J. (1997). Export marketing and the Internet experiences of web site use and perceptions of export barriers among UK businesses. International Marketing Review, 14(5), 324344. Bennett, R. J., & Robson, P. J. A. (1999). Intensity of interaction in supply of business advice and client impact: A comparison of consultancy, business associations and government support initiatives for SMEs. British Journal of Management, 10, 251369. Bentler, P. M., & Chou, C. P. (1987). Practical issues in structural equation modelling. Sociological Methods and Research, 16(1), 78117. Bickerton, P., Bickerton, M., & Simpson-Holley, K. (1998). Cyberstrategy: Business strategy for extranets, intranets and the Internet. Oxford: Butterworth Heinemann. Bollen, K. A., & Lennox, R. (1991). Conventional wisdom on measurement: A structural equation perspective. Psychological Bulletin, 110(2), 305314. Cadogan, J. W., Sundqvist, S., Salminen, R. T., & Puumalainen, K. (2005). Export marketing interfunctional interactions, and performance consequences. Journal of the Academy of Marketing Science, 33(4), 520535. Carson, D., Cromie, S., McGowan, P., & Hill, J. (1995). Marketing and entrepreneurship in SMEs. London, UK: Prentice-Hall. Chaffey, D., Mayer, R., Johnston, K., & Ellis-Chadwick, F. (2000). Internet marketing. New Jersey: Pearson Education Limited. Corbitt, B. J. (2000). Developing intraorganizational electronic commerce strategy: An ethnographic study. Journal of Information Technology, 15(2), 119130. Covin, J. G. (1991). Entrepreneurial versus conservative rms: A comparison of strategies and performance. Journal of Management Studies, 2(5), 439462.
419
Cragg, P. B., & King, M. (1993). Small-rm computing: Motivators and inhibitors. MIS Quarterly, 17(1), 4760. Curran, J., & Blackburn, R. A. (2001). Researching the small enterprise. London, UK: Sage. Dandridge, T., & Levenbury, N. M. (2000). High-tech potential? An exploratory study of very small rms usage of the Internet. International Small Business Journal, 18(2), 8191. Daniel, E., Wilson, H., & Myers, A. (2002). Adoption of e-commerce by SMEs in the UK. International Small Business Journal, 20(3), 253270. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319340. Deng, X., Doll, W. J., Hendrickson, A. R., & Scazzero, J. A. (2005). A multi-group analysis of structural invariance: An illustration using the technology acceptance model. Information and Management, 42, 745759. Diamantopoulos, A., Schlegelmilch, B. B., & Allpress, C. (1990). Export marketing research in practice: A comparison of users and non-users. Journal of Marketing Management, 6(3), 257273. Diamantopoulos, D., & Winklhofer, H. (2001). Index construction with formative indicators: An alternative to scale development. Journal of Marketing Research, 38(May), 269277. Dou, W., Nielsen, U., & Tan, C. M. (2002). Using corporate websites for export marketing. Journal of Advertising Research, 42(5), 105115. Dutta, S., & Biren, B. (2001). Business transformation on the Internet: Results from the 2000 study. European Management Journal, 19(5), 449462. DTI. (2005). http://www.dti.gov.uk/industries/telecoms/pdf/UK_Broadband_Fund_Evaluation. pdf. Frambach, R. T., Barkema, H. G., Nooteboom, B., & Wedel, M. (1998). Adoption of a service innovation in the business market: An empirical test of supply-side variables. Journal of Business Research, 41(2), 161174. Gatignon, H., & Robertson, T. S. (1989). Technology diffusion: An empirical test of competitive effects. Journal of Marketing, 53(January), 3549. Ghose, S., & Dou, W. (1998). Interactive functions and their impacts on the appeal of Internet presence sites. In: P. Richardson (Ed.), Internet marketing, readings and online resources (pp. 234251). Boston: McGraw-Hill. Grewal, R., Comer, J., & Mehta, R. (2001). An investigation into the antecedents of organizational participation in business-to-business electronic markets. Journal of Marketing, 65, 1733. Hamill, J., & Gregory, K. (1997). Internet marketing in the internationalisation of UK SMEs. Journal of Marketing Management, 13, 928. Holmes, T. L., & Srivastava, R. (1999). Effects of relationalism and readiness on EDI collaboration and outcomes. The Journal of Business and Industrial Marketing, 14(5/6), 390402. Igbaria, M., & Zinatelli, N. (1997). Personal computing acceptance factors in small rms: A structural equation model (1). MIS Quarterly, 21(3), 279305. + + Joreskog, K. G., & Sorbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Hillsdale, NJ: Lawrence Erlbaum Associates. Karahanna, E., Straub, D. W., & Chervany, N. L. (1999). Information technology adoption across time: A cross-sectional comparison of pre-adoption and post-adoption beliefs. Management Information Systems, 23(2), 183213.
420
Karakaya, F., & Khalil, O. (2004). Determinants of Internet adoption in small and medium-sized enterprises. International Journal of Internet and Enterprise Management, 2(4), 341365. Kotler, P. (2000). Marketing management (10th ed., International Edition). London: PrenticeHall. Lee, M. K. O., & Cheung, C. M. K. (2004). Internet retailing adoption by small-to-medium sized enterprises (SMEs): A multiple-case study. Information Systems Frontiers, 6(4), 385397. Leong, E. K. F., Ewing, M. T., & Pitt, L. F. (2002). E-comprehension evaluating B2B websites using readability formulae. Industrial Marketing Management, 31, 125131. Levy, M., & Powell, P. (2003). Exploring SME Internet adoption: Towards a contingent model. Electronic Markets, 13(2), 173181. Loiacono, E., Watson, R. T., & Goodhue, D. L. (2002). WebQual: A measure of website quality. American Marketing Association, 2002 Winter Educators Conference Proceedings, 13, 432438. Lumpkin, G. T., & Dess, G. G. (1996). Clarifying the entrepreneurial orientation construct and linking it to performance. Academy of Management Review, 21(1), 135172. Lynn, G. S., Lipp, S. M., Akgun, A. E., & Cortez, A. (2002). Factors impacting the adoption and effectiveness of the World Wide Web in marketing. Industrial Marketing Management, 31, 3549. MacCallum, R. C., & Browne, M. W. (1993). The use of causal indicators in covariance structure models: Some practical issues. Psychological Bulletin, 114(3), 533541. McGowan, P., & Durkin, M. G. (2002). Toward an understanding of Internet adoption at the marketing/entrepreneurship interface. Journal of Marketing Management, 18, 361377. Mehrtens, J., Cragg, P., & Mills, A. (2001). A model of Internet adoption by SMEs. Information and Management, 39, 165176. Miles, M. B., & Huberman, A. M. (1994). An expanded sourcebook: Qualitative data analysis. Beverly Hills, CA: Sage. Moore, G. C., & Benbasat, I. (1994). Integrating diffusion of innovations and theory of reasoned action models to predict utilization of information technology by end-users. In: K. Kautz & J. Pries-Heje (Eds), Proceedings: International Federation for Information Processing (IFIP) Working Conference on Diffusion and Adoption of Information Technology. Oslo, Norway: Chapman & Hall. Mousley, B., & Simintiras, A. (2001). E-marketing: Can one web site serve the global marketplace? The 30th EMAC Conference Proceedings, Bergen, Norway (CD-ROM). Palvia, P., Means, D. B., & Jackson, W. M. (1994). Determinants of computing in very small businesses. Information and Management, 27, 161174. Plouffe, C. R., Vandenbosch, M., & Hulland, J. (2001). Intermediating technologies and multigroup adoption: A comparison of consumer and merchant adoption intentions toward a new electronic payment system. The Journal of Product Innovation Management, 18, 6581. Poon, S., & Swatman, P. M. C. (1997). Small business use of the Internet: Findings from Australian case studies. International Marketing Review, 14(5), 385402. Prasad, V. K., Ramamurthy, K., & Naidu, G. M. (2001). The inuence of Internet-marketing integration on marketing competencies and export performance. Journal of International Marketing, 9(4), 82110. Premkumar, G., & Roberts, M. (1999). Adoption of new information technologies in rural small businesses. Omega, 27(4), 467484.
421
Quelch, J. A., & Klein, L. R. (1996). The Internet and international marketing. Sloan Management Review, 37(3), 6075. Ritchie, B., & Brindley, C. (2000). Disintermediation, disintegration and risk in the SME global supply chain. Management Decision, 38(8), 575583. Rogers, E. (1962). Diffusion of innovations. New York: Free Press. Rogers, E. (1995). Diffusion of innovations (4th ed.). New York: Free Press. Samiee, S. (1998). Exporting and the Internet: A conceptual perspective. International Marketing Review, 15(5), 413426. Schlegelmilch, B. B., Diamantopoulos, A., & Tse, K. (1993). Determinants of export marketing research usage: Testing some hypotheses on UK exporters. In: M. Baker (Ed.), Perspectives on marketing management III (pp. 119139). London: Wiley. Souchon, A. L., Diamantopoulos, A., Holzmuller, H. H., Axinn, C. N., Sinkula, J. M., Simmet, H., & Durden, G. R. (2003). Export information use: A ve-country investigation of key determinants. Journal of International Marketing, 11(3), 106127. Sterne, J. (1999). World Wide Web marketing: Integrating the web into your marketing strategy (2nd ed.). New York: Wiley. Strauss, J., & Frost, R. (1999). Marketing and the Internet: Principles of online marketing. New Jersey: Prentice-Hall. Sultan, F., & Chan, L. (2000). The adoption of new technology: The case of object-orientated computing in software companies. IEEE Transactions on Engineering Management, 47(1), 106126. Thelwall, M. (2000). Effective websites for small and medium-sized enterprises. Journal of Small Business and Enterprise Development, 7(2), 149159. Thong, J. Y. L. (1999). An integrated model of information systems adoption in small businesses. Journal of Management Information Systems, 15(4), 187214. Webb, K. L. (2002). Managing channels of distribution in the age of electronic commerce. Industrial Marketing Management, 31, 95102. Westhead, P., Howorth, C., & Cowling, M. (2002). Ownership and management issues in rst and multi-generation family rms. Entrepreneurship and Regional Development, 14(3), 247269. Willcocks, L., Sauer, C., & Associates (2000). Moving to E-business. Random House. 1&1 Internet Limited. (2004). Research shows UK small businesses ride the waver of Internet usage. http://press.1and1.co.uk. 1&1 Internet Limited. (2005). 1&1 Internet launches 35,000 SMEs online with free web hosting trial. http://press.1and1.co.uk.
422
APPENDIX A. MEASUREMENTS
1. Export Characteristics Export enthusiasm (H1): based on Diamantopoulos et al. (1990) 1. Our company needs to enter new export markets. 2. Our company needs to grow existing export markets. 3. Exporting is a vital part of our business. 4. We commit a lot of resources to growing existing export markets. 5. We commit a lot of resources to nding new export markets. 6. Exporting is likely to expand in our business. Composite reliability 0.817 Export diversity (H3): Number of key export regions (e.g. Schlegelmilch et al., 1993) Export dependence (H2): Percentage of sales derived through exports (e.g. Souchon et al., 2003) 2. General Firm Characteristics Entrepreneurship orientation (H5): based on Covin (1991) In general, the top managers in my rm favour y 1. A strong emphasis on the marketing of tried and tested products or services/a strong emphasis on R&D, technological leadership and innovations. How many new lines of products or services has your rm marketed in the past 5 years (or since its establishment)? 2. No new lines of products or services/very many new lines of products or services. 3. Changes in product or service lines have been mostly of a minor nature/changes in product or service lines have usually been quite dramatic. In dealing with its competitors, my rm y 4. Typically responds to actions which competitors initiate/typically initiates actions which competitors then respond to. 5. Is very seldom the rst business to introduce new products/ services, administrative techniques, operating technologies, etc./is very often the rst business to introduce new products/ services, administrative techniques, operating technologies, etc.
423
APPENDIX A (Continued)
Typically seeks to avoid competitive clashes, preferring a live-and-let-live strategy/typically adopts a very competitive undo-the-competitors strategy. In general, the top managers of my rm have y 7. A strong tendency for low-risk projects (with normal and certain rates of return)/a strong tendency for high-risk projects (with chances of very high returns). In general, the top managers of my rm believe that y 8. Owing to the nature of the environment, it is best to explore it gradually via timid, incremental behaviour/owing to the nature of the environment, bold, wide-ranging acts are necessary to achieve the rms objectives. When confronted with decision-making situations involving uncertainty, my rm 9. Typically adopts a cautious, wait-and-see strategy in order to minimise the probability of making costly decisions/typically adopts a bold, aggressive strategy in order to minimise the probability of exploiting potential opportunities. Composite reliability 0.868 General rm characteristics: Resources ICT knowledge (H6): based on Grewal et al. (2001) Our company: 1. has strong information technology planning capabilities. 2. has very capable technical support staff. 3. has an adequate in-house knowledge about information technology. 4. gives high importance to the strategic use of technology. 5. is experienced with information technology. Composite reliability 0.914 Personal contact network (H7): Newly developed scale 1. Advice from friends and family is extremely useful for developing our website(s). 2. We often sought advice and help from family on website development. 6.
424
We often sought advice and help from friends on website development. Composite reliability 0.844 Perceived supplier service (H8): based on Premkumar and Roberts (1999) 1. We have always developed the website(s) in-house, without any external assistance. (Recoded) 2. There are businesses, that (could have) provided effective technical support of websites(s). 3. External website developers are essential to providing us with a professional website. Composite reliability 0.540 Staff workload (H9): Newly developed scale 1. Although we have ideas to improve the website(s) we do not have sufcient time to undertake these. 2. There are many other important business tasks that distract us from developing the website(s). Perceived cost of website development (H10): based on Premkumar and Roberts (1999) 1. The costs of employing external consultants to develop our website(s) are far greater than the benets obtained. 2. Further development of our website(s) would take signicant resources. Relative advantage (H11): based on Premkumar and Roberts (1999) Our company website(s): 1. allows us to better communicate with our international customers. 2. allows us to cut costs in our international marketing communications. 3. is more effective than our traditional marketing communication methods. 4. allows us to better communicate with potential international customers. Composite reliability 0.868 3.
425
Environmental pressure (H4): based on Premkumar and Roberts (1999) 1. Export customers expect us to continue to develop our website(s). 2. We feel it is a strategic necessity to continue to develop our website(s) to compete in the export market place. 3. We believe we will lose our export customers to our competitors if we do not continue to develop our website(s). Composite reliability 0.798
Item removed during measure purication.
426
APPENDIX B. CORRELATION MATRIX

Measure 1. Export enthusiasma 2. Export dependenceb 3. Export diversityc 4. Entrepreneurship orientationa 5. ICT knowledgea 6. Personal contact networka 7. Perceived awareness and use of supplier servicea 8. Staff workloada 9. Perceived cost of website developmenta 10. Perceived relative advantage a 11. Environmental pressurea 12. Time since website established d 13. Website sophisticatione
a
Mean SD 5.78 39.67 5.72 4.40 4.70 2.55 4.94 4.54 4.34 4.54 4.79 4.20 6.93
10
11
12
13
1.38 1.00 28.54 0.437 1.000 2.72 0.425 0.623 1.000 1.06 0.097 0.096 0.147 1.000 1.47 0.023 0.027 0.115 0.416 1.000 1.64 0.057 0.042 0.009 0.100 0.085 1.000 1.39 0.082 0.052 0.113 0.051 0.345 0.006 1.78 0.049 0.098 0.012 0.176 0.272 0.119 1.60 0.006 0.049 0.011 0.108 0.116 0.240 1.45 0.251 0.217 0.293 0.169 0.131 0.006 1.48 0.278 0.321 0.215 0.185 0.181 0.030 1.63 0.170 0.340 0.300 0.360 0.273 0.010 3.18 0.144 0.125 0.239 0.379 0.409 0.094
1.000 0.059 0.019 0.081 0.079 0.112 0.129
1.000 0.150 1.000 0.149 0.040 1.000 0.081 0.157 0.588 1.000 0.174 0.011 0.267 0.340 1.000 0.332 0.115 0.382 0.408 0.368 1.000
7-point Likert scale (1 strongly disagree; 7 strongly agree). Percentage. c Number of export regions (19). d 7-point scale (1 less than 1 year; 7 more than 10 years). e Index (1 1 feature; 17 17 features).
b

List of Contributors: - Ukurova University, Adana, Turkey

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

List of Contributors: - Ukurova University, Adana, Turkey

Hochgeladen von

Copyright:

Verfügbare Formate

LIST OF CONTRIBUTORS

Nigel J. Barrett Stephen Chen

Dawn R. DeeterSchmelz Pervez Ghauri

Ca Gura lin u Hartmut H. Holzmu ller Kathryn Houghton

Jorma Larimo Kannika Leelapanyalert

Patrick Lentz Tage Koed Madsen

Michael R. Mullen Tho D. Nguyen

Trang T. M. Nguyen Aristides Olivares-Mesa

Giada Palamara Jose Pla Barber

Alex Rialp Criado

Josep Rialp Criado

nchez Peinado Esther Sa Eric Schirrmann

Shirley Ye Sheng Brian T. Straley Sonia Suarez-Ortega

Ernest J. Zavoral, Jr. Antonella Zucchella

INTERNATIONAL MARKETING RESEARCH: OPPORTUNITIES AND CHALLENGES IN THE 21ST CENTURY

ALEX RIALP AND JOSEP RIALP

International Marketing Research

ALEX RIALP AND JOSEP RIALP

THE SPECIAL ISSUE: ORIGIN, STRUCTURE AND CONTENTS

International Marketing Research

ALEX RIALP AND JOSEP RIALP

PART I: EXPORT BEHAVIOR, DEVELOPMENT PROCESS, AND PERFORMANCE

International Marketing Research

ALEX RIALP AND JOSEP RIALP

PART II: STRATEGIC INTERNATIONALIZATION PROCESS IN DIFFERENT SECTORAL SETTINGS

International Marketing Research

ALEX RIALP AND JOSEP RIALP

PART IV: BUSINESS INTERNATIONALIZATION AND INFORMATION TECHNOLOGIES

International Marketing Research

ALEX RIALP AND JOSEP RIALP

International Marketing Research

DIFFERENT TYPES OF EXPORTING SMEs: SIMILARITIES AND DIFFERENCES IN EXPORT PERFORMANCE

Different Types of Exporting SMEs

2. TYPES OF EXPORTING SMEs: TRADITIONAL EXPORTERS VS. BORN INTERNATIONALS

Different Types of Exporting SMEs

Different Types of Exporting SMEs

3. LITERATURE REVIEW OF EXPORT PERFORMANCE STUDIES AND DEVELOPMENT OF HYPOTHESES

Different Types of Exporting SMEs

Different Types of Exporting SMEs

POS NEG. NS IMP. IMP

POS. NEG. NS IMP. IMP.

POS. NEG. NS IMP. IMP.

POS. NEG. NS IMP. IMP.

POS. NEG. NS IMP. IMP.

Different Types of Exporting SMEs

1990s America Europe Other Industrial

0.000 0.000 0.000

0.049 0.002 0.000

0.049 0.013 0.122

0.049 0.038 0.100 0.325

0.001 0.000 0.000

0.100 0.019 0.000

0.099 0.007 0.000

0.001 0.000 0.001

0.099 0.011 0.000

0.001 0.000 0.084

Source: Leonidou et al. (2002).

Different Types of Exporting SMEs

Different Types of Exporting SMEs

Different Types of Exporting SMEs

Different Types of Exporting SMEs

4. SAMPLE, METHODOLOGY, AND OPERATIONALIZATIONS OF THE MEASURES

Different Types of Exporting SMEs