m, .3... .hk.‘ i . ta}. .5 um: Rafi. .. 4an ‘ . 9 6%? A)... .5 ., ... .. Paul}... a PM? swuwwfl gm." ‘ f... . 1.51... .w ... Axum...” A»... .. 34%.” yr . . . :4 ..v..I-.l . V ' n! I run I C': .~' -. f‘.‘ a >0 a This is to certify that the dissertation entitled ENHANCING MARKETING INNOVATION THROUGH MARKETING KNOWLEDGE TRANSFER: AN INVESTIGATION OF STRATEGIC ALLIANCES presented by SANGPHET HANVANI CH has been accepted towards fulfillment of the requirements for Ph.D. degree in Business Administration Date /%m51 fimz / f MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE Nov’W-zm 4 6/01 c'JCIRC/DateDuepssop. 1 5 ENHANCING MARKETING INNOVATION THROUGH MARKETING KNOWLEDGE TRANSFER: AN INVESTIGATION OF STRATEGIC ALLIANCES By Sangphet Hanvanich A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY DEPARTMENT OF MARKETING AND SUPPLY CHAIN MANAGEMENT 2002 ABSTRACT ENHANCING MARKETING INNOVATION THROUGH MARKETING KNOWLEDGE TRANSFER: AN INVESTIGATION OF STRATEGIC ALLIANCES By Sangphet Hanvanich This dissertation investigates the outcome of marketing knowledge that firms acquired from their alliance partners, the mechanisms that firms use to transfer the knowledge, and factors that may affect the knowledge transfer process. The dissertation consists of two distinct parts. Part 1 examines these issues from the shareholder perspective and is an event study using secondary data. Part 2 explores the same issues in more detail from the management viewpoint and is based upon the analysis of primary survey data. The results from Part 1 suggest that announcements of marketing knowledge acquisition through alliance formations enhance shareholder value as reflected in positive abnormal returns. This incremental shareholder value is, however, affected by the type of knowledge being acquired, industry relatedness between alliance and parent firm, and national differences between partners. Respectively, the results point to the effects of knowledge tacitness, absorptive capacity and cultural differences. All of these results are reexamined in Part 2. The results from Part 2 suggest that incremental marketing knowledge also enhances marketing innovation of the parent firms. The process of gaining marketing knowledge, however, involves external knowledge transfer from alliance partners and internal knowledge transfer back to the parent. Partner-to—partner knowledge transfer requires coordination and cooperation between alliance partners as key learning mechanisms, whereas alliance-to-parent knowledge transfer requires rotation of marketing personnel. Results from Part 2 also suggest that trust between partners and a firm’s absorptive capacity strengthen the relationships from coordination and cooperation to partner-to-partner knowledge transfer. However, absorptive capacity weakens the relationship between marketing knowledge and marketing innovation. Moderating effects of cultural differences and tacitness are not found in Part 2. These results and plausible explanations are discussed and future research directions are provided. In loving memory of my mother Nongsri Hanvanich iv ACKNOWLEDGEMENTS I would like to express my earnest gratitude to all my committee members who provided me guidance and encouragement throughout the course of the dissertation. I am very gratefiil to Dr. Cornelia Droge, my advisor, for generously giving me her time and advice, patiently teaching me how to write, and graciously supporting me through difficult times in my academic and personal life. I am greatly indebted to Dr. S. Tamer Cavusgil for his belief and confidence in me and for adopting me into the doctoral program. I am very appreciative of Dr. Roger Calantone for navigating me through the difficult data collection phase and for the methodology expertise that enabled me to complete this dissertation. I am thankful to Dr. Stewart Miller for his guidance and insight in Part 1 and for the thorough comments in other parts of the dissertation. I also would like to thank the Center for International Business Education and Research at Michigan State University for financial assistance and to Dr. Robert Nason, the department chair, for his continuing support throughout the years. Sincere thanks to Beverly Wilkins who helped me with much needed personal help when I first arrived at the university; to Marilyn Brookes, Renee Dixon, Laurie Fitch, Patty Geller, Cheryl Lundeen, Kathy Mullins, Tiffany Norwood and Kathy Waldie for their hard work and dedication; to my friends, Rosanna Garcia, Katrina Savitski and Elif Sonmez for making the doctoral program more enjoyable; and to Tunga Kiyak and to Jeff Meese at the Electrical Engineering department, whose computer knowledge and expertise was invaluable. Finally, thanks to my loving husband, Mike, and to my extended family, especially my grandmother “Khun Kim Ang”, for their love, support and belief in me. TABLE OF CONTENTS LIST OF TABLES ......................................................................................... viii LIST OF FIGURES ................................................................................... x CHAPTER 1 INTRODUCTION ....................................................................................... 1 1.1 Background ................................................................................................ 4 1.2 Research Questions and Objectives ................................................................. 6 1.3 Methodological Basis ................................................................................. 13 1.4 Contributions ............................................................................................ 14 CHAPTER 2 MARKET-BASED EVALUATION ................................................................... 18 2.1 Shareholder Value Creation ......................................................................... 19 2.2 Event Studies: Model and Methodology ......................................................... 20 2.2.1 Cumulative Abnormal Returns ......................................................... 21 2.3 Study I: Characteristics of Knowledge .......................................................... 23 2.3.1 Hypothesis Development ............................................................... 23 2.3.2 Sample Selection .......................................................................... 24 2.3.3 Measurement: Classification of Announcement Content ....................... 25 2.3.4 Analysis and Results ..................................................................... 26 2.4 Study 11: Absorptive Capacity and Cultural Differences .................................... 27 2.4.1 Hypothesis Development ............................................................... 27 2.4.2 Sample Selection ........................................................................... 30 2.4.3 Measurement ............................................................................... 31 2.4.4 Analysis and Results ..................................................................... 32 2.5 Discussion for Event Study I and II ............................................................... 38 2.6 Conclusion for Event Study I and II .............................................................. 41 CHAPTER 3 LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT .......................... 43 3.1 Three Core Processes in Marketing: Domain Specification ................................. 45 3.2 Acquiring and Transferring Marketing Knowledge ........................................... 45 3.2.1 The Role of Strategic Integration ..................................................... 47 3.2.2 Partner-to-Partner Knowledge Transfer ............................................. 48 3.2.3 Alliance-to-Parent Knowledge Transfer ............................................ 49 3.3 Marketing Knowledge and Marketing Innovation ............................................ 51 3.3.1 Marketing Knowledge ................................................................... 52 3.3.2 Marketing Innovation ..................................................................... 57 vi 3.4 Moderating Factors .................................................................................... 61 3.4.1 Characteristics of Knowledge (Tacitness) ........................................... 61 3.4.2 Absorptive Capacity ....................................................................... 63 3.4.3 Trust ........................................................................................... 65 3.4.4 Cultural Difference ........................................................................ 66 3.5 Summary .................................................................................................. 67 CHAPTER4 RESEARCH METHOD ................................................................................... 69 4.1 Unit of Analysis ........................................................................................ 69 4.2 Sampling Frame and Sampling Method .......................................................... 70 4.3 Constructs and Measurements ..................................................................... 73 4.3.1 Alliance Tasks and Objectives ........................................................ 74 4.3.2 Antecedent Constructs ................................................................... 75 4.3.3 Marketing Knowledge and Marketing Innovation Constructs ................. 80 4.3.4 Moderator Constructs .................................................................... 82 4.4 Measurement and Structural Model Testing Approaches .................................... 86 CHAPTER 5 ANALYSIS AND FINDINGS .......................................................................... 87 5.1 Sample Characteristics ............................................................................... 87 5.2 Bootstrapping Procedure ............................................................................ 89 5.2.1 Bootstrapping Procedure for Parameter Estimation ............................... 90 5.2.2 Bootstrapping Procedure for Assessing Model Fit ................................ 93 5.3 Main Model ............................................................................................. 97 5.3.1 Measurement Validation for Main Models .......................................... 97 5.3.2 Hypothesized Structural Models (Main Models) ................................. 109 5.4 Moderation Models .................................................................................. 114 5.4.1 Measurement Validation for Moderation Models ................................ 114 5.4.2 Hypothesized Structural Models (Moderation Models) ...................... 123 5.5 Summary ................................................................................................ 127 CHAPTER 6 DISCUSSION OF RESULTS, LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH .................................................................................. 130 6.1 Discussion of the Results ........................................................................... 131 6.1.1 Marketing Knowledge .................................................................. 131 6.1.2 Marketing Innovation ................................................................... 133 6.1.3 Mechanisms of Knowledge Transfer ................................................ 134 6.1.4 Moderation of Knowledge Transfer ................................................. 140 6.2 Limitations and Additional Future Research Directions .................................... 146 BIBLIOGRAPHY .......................................................................................... 149 vii Table 1.1: Table 2.1: Table 2.2: Table 2.3: Table 2.4: Table 2.5: Table 2.6: Table 2.7: Table 3.1: Table 5.1: Table 5.2: Table 5.3: Table 5.4: Table 5.5: Table 5.6: Table 5.7: Table 5.8: Table 5.9: LIST OF TABLES Overview of Contents in Part 1 and Part 2 ............................................... 2 Analysis of CARS by Different Types of Knowledge Acquisition ............... 27 Ownership Structures and Cultural Difference Between Partners ............... 33 Analysis by Joint Venture Ownership Structure ...................................... 34 Analysis by Type of JV -Parent Relatedness ........................................... 35 Analysis by Type of Partner-Partner Relatedness .................................... 37 Analysis by Cultural Difference .......................................................... 38 Summary of Preliminary Study I and Preliminary Study 11 Results ............. 42 Summary of the Hypotheses ............................................................... 68 Sample Characteristics ....................................................................... 87 Non-Response Bias ........................................................................... 88 Evaluation of Fit Indices: CF A for Antecedents of Marketing Knowledge ...99 Evaluation of Fit Indices: CF A for Marketing Knowledge Construct .......... 101 Evaluation of Fit Indices: CFA for Marketing Knowledge'Construct and its Antecedents .......................................................................... 102 Evaluation of Fit Indices: CFA for Marketing Innovation Construct .......... 104 Evaluation of Fit Indices: CFA for Marketing Knowledge and Marketing Innovation Constructs ........................................................ 105 Final Items and Reliabilities of the Constructs in Main Models ................. 106 Evaluation of Fit Indices: Structural Model for the Relationships Between Marketing Knowledge and its Antecedents ........................................... 110 Table 5.10: Evaluation of Fit Indices: Structural Model for Relationship Between Marketing Knowledge and Marketing Innovation ................................. 113 viii Table 5.11: Final Items and Reliabilities for the Moderators ................................... 117 Table 5.12: Evaluation of Fit Indices: CFAs for Moderation Models ........................ 122 Table 5.13: Moderation Models ....................................................................... 125 Table 5.14: Moderator Parameter Estimates ........................................................ 126 Table 5.15: Summary of Hypothesis Testing Results ............................................ 128 Table 6.1: Summary of Hypothesized Moderating Effects ...................................... 141 ix LIST OF FIGURES Figure 1.1: Conceptual Framework ....................................................................... 3 Figure 3.1: Hypothesized Main Effect Relationships .............................................. 44 Figure 3.2: The Role of Strategic Integration ........................................................ 47 Figure 3.3: Partner-to-Partner Knowledge Transfer ................................................ 49 Figure 3.4: Alliance-to-Parent Knowledge Transfer ............................................... 50 Figure 3.5: Marketing Knowledge and Marketing Innovation ................................. 52 Figure 3.6: Hypothesized Moderating Effects ...................................................... 62 Figure 4.1: Alliance Tasks and Objectives Measures ............................................. 74 Figure 4.2: Strategic Integration Measures .......................................................... 75 Figure 4.3: Coordination Measures .................................................................... 76 Figure 4.4: Cooperation Measures ................. . ................................................... 76 Figure 4.5: Knowledge Sharing Measures ........................................................... 77 Figure 4.6: Alliance-Parent Interaction Measures ................................................. 77 Figure 4.7: Personnel Transfer Measures ............ _ ................................................. 78 Figure 4.8: Partner-to-Partner Knowledge Transfer Measures ................................. 78 Figure 4.9: Alliance-to-Parent Knowledge Transfer Measures ................................ 79 Figure 4.10: Marketing Knowledge Measures ....................................................... 81 Figure 4.11: Marketing Innovation Measures ....................................................... 82 Figure 4.12: Absorptive Capacity Measures ........................................................ 83 Figure 4.13: Tacitness Measures ....................................................................... 84 Figure 4.14: Trust Measures .............................................................................. 85 Figure 4.15: Cultural Difference Measures .......................................................... 85 Figure 5.1: Bootstrapping Procedure for Parameter Estimation ................................ 92 Figure 5.2: Sampling Error Versus Nonsampling Error .......................................... 94 Figure 5.3: Bootstrapping Procedure for Assessing Model Fit ................................. 96 Figure 5.4: Second Order Factor Model of the Marketing Knowledge Construct ......... 100 Figure 5.5: Second Order Factor Model of the Marketing Innovation Construct ......... 103 Figure 5.6: Parameter Estimates for the Relationships Between Marketing Knowledge and its Antecedents ........................................................................ 111 Figure 5.7: Parameter Estimate for the Relationship Between Marketing Knowledge and Marketing Innovation ................................................................ 114 Figure 6.1: Strategic Integration, Coordination, Cooperation, and Partner-to-Partner Knowledge Transfer ....................................................................... 135 Figure 6.2: Strategic Integration, Knowledge Sharing, Interaction and Alliance-to-Parent Knowledge Transfer .............................................. 136 Figure 6.3: Knowledge Transfers and Marketing Knowledge .................................. 138 Figure 6.4: The Direct and Indirect Effects of Knowledge Transfers on Marketing Innovation ................... . ................................................................ 139 xi CHAPTER 1 INTRODUCTION The dissertation consists of two separate parts that are related by their focus: both Part 1 and Part 2 aim to understand the process of marketing knowledge transfer in domestic and international alliances and joint ventures and how knowledge enhances innovation. Part 1 of the dissertation approaches the issue from the shareholder I perspective, whereas Part 2 does so from a managerial viewpoint. Part 1, consisting of two preliminary studies using secondary data, is discussed in Chapter 2. Part 2, using interview and survey data, is discussed in Chapter 3, 4, and 5. Chapter 6 contains discussion of the dissertation results. An overview of studies in both parts is provided in Table 1.1. The rest of this dissertation is organized as follows. Chapter 1 provides a brief background of the dissertation, the research questions and dissertation objectives, the methodological basis for answering the research objectives, and contributions of the research. The next chapter (Chapter 2) provides results of the two preliminary studies based on secondary data. These studies lead to the selection of the moderating constructs used in the subsequent survey study. Chapter 3 defines the domain of the survey study, reviews the literature and develops hypotheses based on the conceptual framework in Figure 1.1. Chapter 4 provides measurements for the constructs and the research method used in the survey. Chapter 5 discusses the analysis-results related to Part 2. Discussion of dissertation results, limitations, and directions for future research are in Chapter 6. Table 1.1: Overview of Contents in Part 1 and Part 2 Content Dissertation Dissertation Part 1 Part 2 Key Questions 0 Can knowledge 0 Can acquired acquisition in joint knowledge increase ventures and strategic firm’s marketing alliances create knowledge and is shareholder value? marketing knowledge 0 What are the key related to marketing moderators of innovation? knowledge transfer 0 What are the process? appropriate knowledge transfer mechanisms? 0 How do the moderators affect knowledge transfer mechanisms? Key Dependent o Shareholder Value 0 Marketing Knowledge Variables/Constructs Creation (Cumulative Transfer Abnormal Returns) Marketing Knowledge Marketing Innovation Methodology Secondary Data Primary Data Data Source Study 1: 0 Interviews 0 Dow Jones News 0 Survey Retrieval Study II: 0 Thomson Financial Security Data Analysis Event Studies Structural Equation Model Location in the Dissertation Chapter 2 Chapter 3,4 and 5 83965: mauve—SE Emma“; 032320. 803m -862822 owcorsoqm mauve—82 8.23:. owcorsoex Eaten bio—58¢ Comma—«C. .058qu 830835 Bream ego—>55— coufioaoou cocmcmEooU cowmuwBE omwofibm mwflofioooué 8:08me 35:30 339 .3638 3:93.94. 80563. B88032 33308.2..— .aEnS—SU £4 95me 1.1 Background It seems undeniable that knowledge and innovation are keys to wealth creation in today’s business environment (Drucker 1993; Hamel 1998). In response to changes in the business environment, firms form alliances and joint ventures with other firms as a way to adjust themselves to be more responsive into a knowledge intensive society (Drucker 1993; Inkpen 1996). Increasingly, firms create global webs of business collaborations since stand-alone competition is giving way to networked rivalry (Srivastava, Shervani and Fahey 1999). As vehicles to learn (Kogut 1988), alliances and joint ventures serve as the means to acquire new marketing knowledge and innovative capability. The notion that firms form joint ventures and alliances so as to acquire knowledge from their business partners leads to two important questions: 1) how to measure the acquired knowledge and/or its impact on performance and 2) how to manage the knowledge acquisition and transfer processes. If joint ventures and alliances are the strategic initiatives through which firms can acquire business knowledge and innovative capability, and if knowledge and innovation are keys to wealth creation, then one way to measure acquired knowledge is to measure wealth created by joint ventures and alliances (i.e., the performance impact). In this dissertation, acquired knowledge is first indirectly measured through the wealth created for shareholders (shareholder value creation) and then directly measured from managerial assessment. The first part is discussed in Chapter 2, whereas the second part is discussed in Chapter 3, 4 and 5. Researchers, as well as practitioners, see knowledge management from different viewpoints (Davenport, Long and Beers 1998). Some see knowledge management as a process supporting organizational learning (Huber 1991); others claim that the major task of a firm is to integrate knowledge residing within individual employees so as to achieving knowledge application and innovation (Grant 1996). The first focuses on how firms gather, store, and disseminate knowledge, with or without information technology. The later emphasizes transfers of knowledge among individuals, across functional units, across organizations, or across national boundary. This dissertation adopts the second viewpoint and emphasizes the transfer of knowledge across organizations and possibly across national boundaries (alliances and joint ventures, domestic and international) so as to increase marketing knowledge and achieve marketing innovation in the parent firms. In the second part of the dissertation, the meanings of marketing knowledge transfer and marketing innovation are specified in detail. The dissertation conceptualizes marketing innovation within the context of strategy innovation. Broadly, innovations can be categorized into two categories: technology innovation and strategy innovation (see Damanpour 1991). Technology innovation focuses on introducing technological solutions to business and customer problems. Strategy innovation, on the other hand, focuses on redefining the problems and creating fundamentally new and superior value (Kim and Mauborgne 1999). This dissertation adopts the later perspective of innovation in order to explore the relationship between marketing innovation and marketing knowledge. Following Srivastava, Shervani and F ahey (1999), the dissertation defines marketing as as a phenomena embedded in three core marketing processes: product development management (PDM), supply chain management (SCM) and customer relationship management (CRM). These processes emphasize customer value creation through the accomplishment of the development of new customer solutions, the enhancement of input acquisition and output transformation, and the creation of relationships to external market entities especially channel members and end users. The three processes thus address common marketing tasks and are the core objectives of marketing in most business organizations. Therefore, contextually, marketing knowledge overall. refers to marketing knowledge of PDM, SCM, and/or CRM and marketing innovation overall refers to marketing innovation in PDM, SCM and/or CRM. Various factors have been identified as potential moderators affecting the knowledge management and innovation creation process. These include characteristics of knowledge being transferred (Kogut and Zander 1993), levels of the firm’s absorptive capacity (Cohen and Levinthal 1990), and the national cultures of partner firms (Hofstede 1983; Simonin 1999a). Part 1 of the dissertation (Chapter 2) explores the impact of these moderators on shareholder value creation. Part 2 of the dissertation (Chapter 3, 4 and 5) validates the impacts of these factors on marketing knowledge transfer and marketing innovation from a managerial viewpoint. 1.2 Research Questions and Objectives In Part 1 of the dissertation (Chapter 2), the objectives were as follows. The first was to indirectly assess the effect of acquired knowledge from the shareholder perspective. If knowledge is the core capability that creates value to firms (Nonaka 1994) and if joint ventures are the ways in which firms can acquire knowledge (Kogut 1988), then one can assume that joint ventures must be knowledge creating entities. Thus, one way to measure knowledge that is acquired through joint ventures is to measure the value that joint ventures created. Although there are various methods for tapping firm valuations, approaches based on shareholder value (SHV) have received greater support than others (Srivastava, Shervani and F ahey 1998). SHV is based on the net present value (NPV) of future projected cash flows and on perceived growth potential, as opposed to being based on a mere continuation of past performance. Part 1 of the dissertation measures joint venture shareholder value creation to assess joint venture acquired knowledge. The result shows that, overall, there is a positive shareholder value creation associated with announcements of firms’ knowledge acquisition through joint ventures. The second objective of Part 1 was to explore the impacts of moderating factors on shareholder value creation. Drawing from research in organization learning and knowledge management, the dissertation examines the effects of knowledge type, the firm’s absorptive capacity and cultural difference on shareholder value creation. To explore these relationships, Part 1 of the dissertation is divided into two distinct studies that serve as preliminary studies for the survey study in Part 2 of the dissertation (Chapter 3, 4, and 5). Specifically, preliminary Study 1 (Section 2.3) examines whether different types of knowledge (market knowledge, manufacturing knowledge and technology knowledge) that firms seek to acquire from their partners lead to different effects on shareholder value (although all types of knowledge acquisition are expected to lead to positive shareholder value creation). Preliminary Study 11 (Section 2.4) investigates whether national differences and absorptive capacity moderate shareholder value creation. National differences are expected to negatively moderate the learning process, which results in decreased shareholder value. A firrn’s absorptive capacity is expected to positively moderate the firm’s learning process, which leads to increased shareholder value creation. Absorptive capacity is measured from industry relatedness, both JV- parent relatedness and partner—to-partner relatedness. Results from Part 1 suggest that, contrary to the result at the aggregate level, access to knowledge may not always result in positive shareholder value creation. Value creation, however, could be influenced by other factors such as the characteristics of transferred knowledge, absorptive capacity (measured directly rather than through industry relatedness), trust and cultural differences. Thus the dissertation includes these variables in the model of the marketing knowledge transfer process in Part 2. Part 2 of the dissertation (Chapter 3, 4, and 5) proposes a theoretical model and then uses the moderators identified in Part 1 to assess the hypothesized relationships between the mechanisms firms use to transfer marketing knowledge and the outcomes of knowledge transfer. In doing so, Chapter 3 defines the mechanisms firms use to transfer marketing knowledge in the three marketing domains (PDM, SCM, and CRM), the knowledge transfer outcomes (marketing knowledge and marketing innovation) and the moderators, and then examines the relationships among them. Therefore, the objectives of Part 2 were as follows. The first objective was to re-conceptualize the marketing knowledge construct. Currently, marketing knowledge has been conceptualized as market information, which needs to be processed through knowledge acquisition, information distribution, information interpretation and organizational memory (Moorman and Miner 1997). This conceptualization of knowledge has been well developed in market orientation studies (J aworski and Kohli 1993). However, the dissertation argues that the marketing knowledge construct should be conceptualized to capture the extent to which firms actually understand their marketing tasks in the three previously defined marketing domains. That is, the construct should measure how much a firm knows about the tasks, rather than how much information about the tasks is processed. The dissertation proposes that marketing knowledge can be measured by systematically classifying levels or stages of understanding (Bohn 1994). These stages, ranging from complete ignorance to complete knowledge, also capture the different degrees of knowledge tacitness and learning types. This framework is fundamentally based on how to precisely map, evaluate, and compare levels of cognitive understanding. In this view, better knowledge of the three domains in marketing leads to better performance (such as more innovation in PDM, SCM, and CRM) without incremental physical investment. In contrast to most approaches for measuring knowledge, the nature of the knowledge changes qualitatively with each stage in this framework. Results from the dissertation show that measuring marketing knowledge from this approach provides good construct validity. The second objective of Part 2 was to re-conceptualize the marketing innovation construct. The dissertation argues that marketing innovation, as part of strategy innovation, should be conceptualized as the capacity to reconceive the existing industry model in ways that create new value to customers, undermine competitors, and produce new wealth for all stakeholders (Kim and Mauborgne 1999). Unlike marketing information and market orientation, marketing innovation has gained limited attention from marketing scholars; the exception is research related to product development, where the focus is on innovative ideas manifested by successful new products. The dissertation proposes a marketing innovation construct that covers not only the extent to which firms are innovative in that they radically improve their products, but also the extent to which firms are innovative in that they target non-existing demand or customers through demonstrating willingness to lose some existing customers. In addition, the marketing innovation construct also encompasses the idea of building a firrn’s capabilities through combining existing capabilities with the other companies’ capabilities, as opposed to leveraging and extending the current capabilities of the firm. This conceptualization of marketing innovation thus covers the three marketing tasks in PDM, SCM, and CRM domains. The results support the notion that marketing innovation can be conceptualized as encompassing these three business domains. The third objective was to examine the appropriate means for marketing knowledge transfers among business alliance and joint venture partners. Specifically, Part 2 examines whether parent strategic integration leads to coordination and cooperation between partners as well as to knowledge sharing, interaction, and personnel transfer between the joint venture and parent. The results support these proposed relationships. Subsequently, the dissertation examines whether coordination and cooperation increase the level of partner-to-partner marketing knowledge transfer. Additionally, the dissertation investigates whether marketing knowledge sharing (from alliance to its parent) and both interaction and personnel transfer between the alliance and its parent increase the level of alliance-to-parent marketing knowledge transfer. The results support the proposed relationships from coordination and cooperation to partner- to-partner knowledge transfer, as well as the proposed relationship between personnel transfer and alliance-to-parent knowledge transfer. The proposed relationships from 10 knowledge sharing and interaction to alliance-to—parent knowledge transfer are, however, not supported. The conceptualization of knowledge transfer along two distinctive pathways parallels the examination of the two types of industry relatedness (JV-parent and partner-partner relatedness) in Part 1 of the dissertation. Part 2 of the dissertation then examines whether either partner-to-partner marketing knowledge transfer or alliance-to-parent marketing knowledge transfer (or both) lead to enhanced marketing knowledge in the partner firm. Finally, the last link of the model examines whether the marketing knowledge a firm acquires from business partners leads to enhanced marketing innovation in that firm. Results fiom the dissertation support these proposed relationships among knowledge transfer, marketing knowledge, and marketing innovation. The fourth objective of Part 2 was to examine the impact of key moderating factors on the process of knowledge transfer described above and shown in Figure 1.1. In accordance with Part 1, the moderators studied are tacitness (a characteristic of knowledge), levels of absorptive capacity, and national cultures. In addition, Part 2 examines the moderating effect of trust in the learning process of firms forming joint ventures. Specifically, for tacitness, the dissertation hypothesizes that if the level of tacitness is low, then the relationships from cooperation and coordination to partner-to- partner knowledge transfer are stronger than if the level of tacitness is high. Similarly, if the level of tacitness is low, then the relationships from knowledge sharing, interaction and personnel transfer to alliance-to-parent knowledge transfer are stronger than if the level of tacitness is high. The results, however, do not support these proposed effects of tacitness on the relationship between learning mechanisms and knowledge transfers. 11 For absorptive capacity, the dissertation expects that if the level of absorptive capacity is high, then the relationships from coordination and cooperation to partner-to- partner knowledge transfer are stronger than if the level of absorptive capacity is low. Similarly, if the level of absorptive capacity is high, then the relationships from knowledge sharing, interaction, and personnel transfer to alliance-to-parent knowledge transfer are stronger than if the level of absorptive capacity is low. Additionally, the dissertation argues that absorptive capability plays a role in transforming marketing knowledge into marketing innovation. That is, if the level of absorptive capacity is high, then the relationship between marketing knowledge and marketing innovation is stronger than if the level of absorptive capacity is low. The dissertation found that only the effects of absorptive capacity on the relationships from coordination and cooperation to partner- to-partner knowledge transfer are supported. The rest of the proposed effects of absorptive capacity are not supported. Trust and cultural distance are partner-related variables, so they are expected to moderate only the learning mechanisms that are related to partner-to-partner knowledge transfer. That is, if the level of trust is high, then the relationships fiom coordination and cooperation to partner-to-partner knowledge transfer are stronger than if the level of trust is low. However, if the level of cultural distance is low, then the relationships from coordination and cooperation to partner-to-partner knowledge transfer are stronger than if the level of cultural distance is high. The findings show that only the moderating effects of trust on the relationships from coordination and cooperation to partner-to-partner knowledge transfer are supported. The proposed effects of cultural difference and the other proposed effects of trust are not found. 12 1.3 Methodological Basis The proposed relationships and the results discussed in the previous sections derive from two distinct approaches, one from the shareholder’s perspective and one from a managerial perspective. These approaches demand different methodological bases. In the first part of the research (Part 1), secondary data was used to conduct event analysis. Secondary data sources (Dow Jones News Retrieval and Thomson Financial Security Data) were consulted to identify joint venture announcements. After the parent firms were identified, their stock prices were retrieved from the CRSP (Center for Research in Security Prices) database and event analysis was performed accordingly. Event analysis or event studies (Brown and Warner 1985) are natural experiments that assess the impact of an event on a firm’s market value using expected stock returns as benchmarks. The event (e. g., a company announcement) contains new information, which is then incorporated in the stock price by investors. The stock price response reflects investors’ assessments of the new information, which is, in this dissertation, related to the formation of new alliances and joint ventures. Researchers have used event studies to examine the effects of various types of announcements including announcements of celebrity endorsement (Agrawal and Kamakura 1995), new product launches (Lane and Jacobson 1995), service changes (Nayyar 1995), and alliances and joint ventures (Koh and Venkatraman 1991; Reuer and Koza 2000). Also, researchers have shown that stock market responses to announcements provide a reliable indication of long-term performance and managerial assessments (Koh and Venkatraman 1991). In the second part of the research (Part 2), managers in knowledge management areas, marketing managers, as well as managers responsible for the alliance projects were 13 interviewed in order to identify the key components of knowledge management and to develop measurement scales. Subsequently, questionnaires were sent out to managers responsible for marketing in alliances or joint ventures. The data were then analyzed using Structural Equation Modeling (SEM) and regression. To overcome the inherent problems associated with small sample size, the analyses were conducted using samples generated from bootstrapping. Details of the bootstrapping procedure are discussed in Section 5.2. To validate the marketing knowledge construct and other constructs, confirmatory factor analysis using SEM was employed (Anderson and Gerbig 1988; Bollen 1989). To test the proposed relationships, SEM was used. To test the effects of moderators, Seemingly Unrelated Regression (Zellner 1962) and ordinary regression were used. Seemingly Unrelated Regression (SUR) is a method to estimate the parameters of a set of regression equations, whose dependent variable is shared by more than one independent variable. Details of the confirmatory factor analysis and hypothesis testing are discussed in Section 5.3 and 5.4. 1.4 Contributions The research seeks to explain marketing knowledge and marketing innovation as outcomes of learning from strategic alliances. Since the literature proposes that knowledge management and creation (i.e., innovation) are keys to firm success, the results should be of great interest to practitioners. In addition, the thesis extends the domain of theoretical work on the knowledge-based view of the firm (Grant 1996) and strategy innovation (Hamel 1998; Kim and Mauborgne 1999), and addresses key 14 measurement operationalization issues. Thus, the contributions of the research fall in five key areas. First, Part 1 of the dissertation contributes to research by determining conditions in which a performance effect (as a result of knowledge transfer) is detectable in terms of shareholder value creation. For managers, Part 1 offers an insight into partner selection since Part 1 considers shareholder value creation as a result of forming joint ventures with various types of partners: those in related or unrelated industries, as well as similar or dissimilar cultures. Part 1 of the dissertation thus provides opportunities for managers to assess the ramifications of their decisions on the firm’s capital gains before the joint venture is actually formed. Theoretically, Part 1 extends research that views joint ventures as ways to acquire knowledge from the partners through examining the impacts of knowledge type, cultural difference and absorptive capacity in light of shareholder value creation. Second, the dissertation contributes to the understanding of how marketing knowledge and marketing innovation can be assessed by proposing measures that can tap the three key domains of marketing, i.e., new product development, supply chain management, and customer relationship management (Srivastava, Shervani and Fahey 1999). For managers, this should provide an insight into how marketing knowledge and marketing innovation (as performance outcomes) could be assessed. The measurement model provides managers with practical psychometric scales that can be used to assess their knowledge and innovation in the three domains of marketing. For academia, the constructs provide new ways to conceptualize marketing knowledge and marketing innovation. Through these new lenses, researchers should be able to uncover new 15 relationships concerning the production and utilization of knowledge in a knowledge- intensive business environment. The new conceptualization of marketing knowledge makes assessment of the relationship between knowledge and innovation possible because it provides a clear distinction between marketing knowledge that is transferred (i.e., information) and understanding in the three marketing domains (i.e., knowledge). Third, the dissertation contributes to the understanding of how knowledge among business partners should be transferred. Part 1 provides preliminary results suggesting a conceptualization of learning in joint ventures along two distinct pathways. These results are validated in Part 2, where marketing knowledge and marketing innovation are hypothesized to be engendered by two sub-processes of knowledge flows, namely ‘ partner-to-partner transfer (acquiring knowledge from the partner) and alliance-to-parent firm transfer (transfer knowledge back to the parent); each flow requires a different underlining mechanism. For managers, this part of the dissertation provides insights into the mechanisms by which firms can effectively manage joint ventures and alliances so as to maximize knowledge transfer and consequently to maximize marketing knowledge and marketing innovation in the three marketing domains. For academia, this extends theoretical understanding of the antecedents and consequences of marketing knowledge flows across organization. Traditionally, researchers considered only acquiring knowledge from partners, whereas knowledge transfer back to the parent was automatically assumed. Forth, the dissertation contributes to the understanding of how trust, the characteristics of knowledge (i.e., tacitness), levels of absorptive capacity, and cultural differences moderate the marketing knowledge transfer process. Part 1 of the dissertation 16 explores the impact of selected moderators on shareholder value, whereas Part 2 of the dissertation studies these impacts through managerial assessment. For managers, understanding moderators contributes to the area of marketing alliance/joint venture management: the findings should provide insights into what managers should do in light of different knowledge types, and different levels of trust, cultural difference, and absorptive capacity. For researchers, this contributes to the better understanding of the boundary conditions of joint venture theory (where a joint venture/alliance is viewed as a vehicle to learn). Finally, the dissertation contributes to the understanding of how marketing innovation is related to marketing knowledge. Part 2 of the dissertation provides insight into the antecedents of strategy innovation and the relationship between knowledge gained from business partners and marketing innovation in that firm. Managerially, this provides a direction for managers pursuing innovative ideas in PDM, SCM, and CRM as it sheds light on what leads to innovative ideas in three marketing domains. If marketing knowledge is an antecedent for marketing innovation, managers may need to invest in gaining more knowledge about the product/market they are operating in so as to arrive at more innovative ideas. For academia, this provides a theoretical understanding if innovation comes from within a system (Grossman and Helpman 1991). This should be able to provide insights into sources of strategy innovations in the three marketing domains. 17 CHAPTER 2 MARKET-BASED EVALUATION Chapter 2 discusses Part 1 of the dissertation, which consists of two distinct studies that serve as preliminary studies for the survey study in Part 2 of the dissertation (Chapter 3, 4, and 5). In Section 2.1, the chapter discusses shareholder value creation, which is a market-based evaluation, and how it can be used to capture the economic value created by announcements of joint venture formation. Section 2.2 discusses the event study methodology, which is employed in both preliminary studies. Preliminary Study I (Section 2.3) examines whether different types of knowledge that firms seek to acquire from their partners (measured from motivations of the alliance formation) create different patterns of changes in shareholder value. Preliminary Study 11 (Section 2.4) investigates whether absorptive capacity (measured as industry relatedness) and national differences affect shareholder value creation. Finally, the chapter discusses the findings from the two studies (Section 2.5 and 2.6). Preliminary Study I and preliminary Study 11 differ not only in their study objectives, but also in their secondary data sources. Data for preliminary Study I were obtained from Dow Jones News Retrieval, whereas data for preliminary Study 11 were obtained from Thomson Financial Security Database. This is because, in preliminary Study 1, motivations of joint venture formation must be identified from the content of the announcements, and thus the actual announcements (e. g., news pieces) must be obtained. Dow Jones News Retrieval provides these actual announcements. On the other hand, in preliminary Study 11, the other characteristics of firms forming joint ventures (such as the 18 industries they are in) must be identified. This information is more easily and accurately obtained from commercially available sources such as Thomson Financial Security Database. Thus, two different data sources are used for the two preliminary studies. The next section describes shareholder value creation, the dependent variable for both preliminary studies. 2.1 Shareholder Value Creation There is considerable debate regarding how economic value is created by strategic initiatives and how it should be measured (Srivastava, Shervani and Fahey 1998). Although there are various valuation methods, approaches based on shareholder value (SHV) have received greater support than others (Srivastava, Shervani and Fahey 1998). SHV is created by a business process and is based on the net present value (NPV) of future projected cash flows. Although the concept is discouraged by some researchers, due to possible difficulty in projecting a firm’s future performance, it is well accepted by the dominant financial perspective (Srivastava, Shervani and Fahey 1998). Srivastava, Shervani and Fahey (1998) argue that, viewed fi'om the financial perspective, market value created by strategic initiatives (such as alliance and joint venture formations) is best reflected by the NPV of all future cash flows expected to accrue to the firm. This is because NPV firm valuation is based on perceived growth potential and associated risks, as opposed to a mere continuation of past performance. The challenge, therefore, is to demonstrate and measure the value created by resources devoted to marketing activities in terms of the impacts both on current outcomes and on perceptions of future financial performance. 19 This chapter discusses two studies using SHV to measure the expected future performance of joint ventures. These two studies serve as preliminary studies for the subsequent survey research. The objective of the studies is to identify key moderating factors that may impact joint venture performance. The assumption for the preliminary studies is that factors affecting the future expected value of a marketing joint venture are the same as those affecting marketing knowledge transfer and, subsequently, marketing innovation of the firm. This is based on the premise that knowledge and innovation are keys to wealth creation (Hamel 1998). The rest of this chapter first discusses the methodology used in both studies. The discussion of hypothesis development, sample selection, measurement and results of preliminary Study I is discussed first, followed by those of preliminary Study 11. The hypotheses for preliminary Study I are labeled as H 1.1 to H 1.3, whereas the hypotheses for preliminary Study 11 are labeled as H 11.1 to H 11.4. The final section of this chapter discusses the findings from the two studies. 2.2 Event Studies: Model and Methodology The two preliminary studies employ event study methodology. Event studies are natural experiments that assess the impact of an event on a firm’s market value using expected stock returns as a benchmark. The event (e. g., a company announcement) contains new information that is then incorporated in the stock price by shareholders. Changes in the stock returns thus reflect shareholder assessment of strategic initiatives being announced. In order to investigate the effect of an announcement, the difference between actual and predicted stock returns on the announcement day is computed and abnormal return (AR) for the announcement day is yielded. In order to assess the effects 20 over time, the abnormal returns are summed over different time intervals around the announcement day. The result of the summation is cumulative abnormal returns (CARS), which tap shareholder value creation. Section 2.2.1 discusses in detail the calculation to obtain CARS. In the present study, abnormal returns created by joint venture announcement represent increased shareholder value created by joint venture formation. Researchers have previously used event studies to examine the shareholder value creation associated with JV announcements (e.g., McConnell and Nantell 1985; Koh and Venkatraman 1991; Merchant and Schendel 2000; Reuer and Koza 2000). Although some researchers question the use of stock market responses to announcements with regard to strategy implementation (Ravenscrafi and Scherer 1987), others have shown that stock-market responses to announcements provide a reliable indication of long-term performance (Healy, Palepu and Ruback 1992) and managerial assessments (Koh and Venkatraman 1991) 2.2.1 Cumulative Abnormal Returns Shareholder value creation is the dependent variable in both preliminary Study I and preliminary Study II of Part 1 of this dissertation. Shareholder value creation is the measurement of ex ante JV performance and is measured using cumulative abnormal stock returns associated with the joint venture announcement. This cumulative abnormal return reflects the shareholders’ assessment of the future joint venture performance and is reflected in the stock market response. The procedure of measuring the cumulative abnormal stock returns (CARS) is described by event study methodology. 21 Following Brown and Warner (1985), each of preliminary Study I and preliminary Study 11 use a market model in the event study, in which the day on which the joint venture was announced is considered the event date (i.e., day 0 or t=0). The trading days prior to the announcement are day -l , day -2, and so on. The days following the announcement are referred to as day +1, day +2, and so on. Ordinary least squares is used to estimate parameters of the market model during a l4l-day estimation period (t = -150 to t = ~10). That is, rit = at + flirmt +eit where r,-, equals firm i ’5 return, 6, represent the systematic risk, rm, is the market return, and e“ is the residual on day t. The daily return of the S&P 500 index is used as a proxy for the market return. A firm’s risk-adjusted abnormal return (ARit) on day t is: ARit = rit_di—18irmt where ARi, is the abnormal return of firm i on day t, which reveals the impact of new information about firm i on day t. The coefficients (with hats) were determined from each firrrr’s estimation-period regression. Although the announcement date is considered day 0, it is likely that some firms released their JV announcements on the previous day before the close of the stock market or near the close of the stock market on day 0. Following Nayyar (1995), a 3-day announcement period (Day —1, Day 0, and Day +1) is used, and therefore each firm’s cumulative abnormal return (CAR) for the announcement period is computed as: +1 CAR,- = 2M” t=—1 22 This measure is used as the dependent variable in both preliminary Study I (described in Section 2.3, subsections 1 to 4) and preliminary Study 11 (described in Section 2.4, subsections 1 to 4). 2.3 Study 1: Characteristics of Knowledge 2.3.1 Hypothesis Development Firms expand abroad to acquire resources, to generate sales, or to diversify markets and suppliers through learning opportunities (Daniels and Radebaugh 1998). These learning prospects range from the opportunities to access tangible resources to intangible resources. Makhija and Ganesh (1997) contended that codifiability of knowledge associated with the business domain being learned is important to learning success. The more codifiable the knowledge, the easier it is to acquire, analyze, and disseminate. The less codifiable knowledge, on the other hand, the more difficult it is difficult to analyze and transmit. However, additional costs are incurred when doing business abroad, due to the unfamiliarity of the local environment (with, e.g., cultural, legal, political and economic differences) and the coordinating activities across geographic distances (Hymer 1976). Nevertheless, multinational companies (MNEs) can benefit from gaining market knowledge through the partnership. This market knowledge includes knowledge about the host-country’s institutional environment, local suppliers and local customers. Forming a JV in a foreign country can lead to use of the firm’s distribution channels, their patents and licenses, and their skilled personnel; as well the opportunity exists to internalize the local partner’s government relations and marketing know-how (Makhija 23 and Ganesh 1997). All of these can improve performance and increase firm value. Therefore, H 1.]: W3 outside the home country (outside the US.) are expected to achieve higher cumulative abnormal returns than JVs within a parent firm 's home country (within the US). Domestic and international JVs provide firms with important means to acquire knowledge and facilitate organizational learning (Makhija and Ganesh 1997). Access to technical knowledge can lead either to the refinement of existing products and technology or to new breakthroughs that create sustainable competitive advantage. Similarly, access to manufacturing knowledge will enable firms to increase production efficiency and effectiveness. Therefore, it is expected that a firm that acquires technical knowledge and manufacturing knowledge in a JV will create ex ante value. Thus: H 1.2: Firms that acquire manufacturing knowledge in a JV are expected to achieve higher CARS than firms that do not acquire manufacturing knowledge. H 1.3: Firms that acquire technology knowledge in a JV are expected to achieve higher CARS than firms that do not acquire technology knowledge. 2.3.2 Sample Selection Dow Jones News Retrieval was used to identify announcements of joint ventures over the period 1980 - 1999. The content of each announcement was then analyzed in detail to identify the motivation for the joint venture formation. The study focused on JV 5 between two partners, at least one of which is headquartered in the United States. This search produced 316 observations (i.e., U.S. partners). Subsequently, the Center for Security Price Research (CRSP) was used to obtain daily stock returns for the sample firms for the period 1980-1998. In addition, an online source (Yahoo’s finance web site) was used to obtain stock returns of firms that announced joint ventures in the year 1999 24 because the 1999 stock price data were unavailable from CRSP at the time the study was conducted. If a firrn’s stock return data was unavailable from CRSP or Yahoo’s finance web site, it was removed from the sample. Two final refinements were then made to the data set. First, the observations that had merger, acquisition, or earnings announcements, as well as those for which analysts made recommendations during the 3-day announcement period (the day of, the day before, and the day after the announcement) were excluded. Second, since the valuation of service firms is usually based on a different market model, service-related firms were taken out. The final sample yielded 240 observations. 2.3.3 Measurement: Classification of Announcement Content The content in each of 240 announcements was analyzed to determine the objectives of the joint venture formation. Since a firm might join a joint venture for more than one type of learning, these observations are not mutually exclusive. The objectives of the learning were coded into dichotomous variables. Local market access is a dichotomous variable that equals one if a US. firm is gaining market access in a host country via the joint venture, zero otherwise. Manufacturing learning is a dichotomous variable that equals one if the US. firm acquires manufacturing know-how in the joint venture, otherwise zero. Technology learning is a dichotomous variable that equals one if the US. firm acquires technology know-how in the joint venture, otherwise zero. Eighty-two firms were identified to have local market access objectives, 23 firms were identified to have manufacturing learning objectives, and 68 firms were identified to have technology learning motivations. Shareholder value creation, the dependent variable, 25 was measured using cumulative abnormal stock returns associated with the joint venture announcement (see Section 2.2. 1). 2.3.4 Analysis and Results After computing the CARS for each observation, the significance of the CARS in each group was tested. The difference in CARS across the groups was then tested using ANOVA. Table 2.1 presents results showing a positive average valuation effect associated with JV formation of 1.38% (p < 0.01). This is consistent with the results of prior JV studies (e.g., McConnell and Nantell 1985; Koh and Venkatraman 1991; Reuer and Koza 2000). JVS within the United States had CARS of 1.77% (p < 0.01). In contrast, JVS outside of the United States had CARS of 0.64% (p < 0.10). The far right column of Table 2.1 presents the f-Statistics for the difference in mean CARS for each of the pairings. In contrast to Hypothesis 1.1, the findings reveal that JV S in the US. achieved higher CARS than JV S outside the United States. This result was significant at the 1% level. The results indicate that the CARS for manufacturing learners (0.08%) were nonsignificant, but that non-manufacturing learners reported CARS of 1.52% (p < 0.01). The difference in mean CARS is significant at p<0.10 level. Furthermore, the difference in CARS of the technology learners versus non-learners was not Significant (1.36% versus 1.39%, ms), which did not support the predicted direction. Three way ANOVA was also conducted to confirm the above analysis. The results showed the same conclusion in that local market access was significant (p<0.05), manufacturing learning access was significant (p<0.10), and technology learning access was nonsignificant (p = 0.957). Neither the interaction effect between local market 26 access and technology learning access nor the interaction effect between local market access and manufacturing learning access was significant (p = 0.357 and 0.930, respectively). The three factor interaction was also nonsignificant (p = 0.271). TABLE 2.]: Analysis of CARS by Different Types of Knowledge Acquisition Joint Venture Mean Different Difference Characteristic CARS from zero in Means N (“/o) (t-Stat) (f-stat) Total Sample 240 1.38 551*” Access to Local Market JV 3 outside the US. 82 0.64 1.84* JVS within the US. 158 1.77 532*" 4.76*** Access to Manufacturing Learning JV 3 with Manufacturing Learning 23 0.08 0.91 JVS with No Manufacturing Learning 217 1.52 5.72*** 290* Access to Technology Learning JVS with Technology Learning 68 1.36 264*" JVS with No Technology Learning 172 1.39 4.87*** 0.01 ***=significant at 1% level; **=significant at 5% level; *=significant at 10% level. 2.4 Study 11: Absorptive Capacity and Cultural Differences 2.4.1 Hypothesis Development JV -Parent Relatedness & Organizational Learning When a joint venture is characterized as a vehicle to learn, the extent to which the venture can acquire and transfer knowledge to its parent becomes a primary factor in determining its success. The ability of a firm to acquire and transfer knowledge depends largely on its experience and familiarity with the knowledge being developed by the JV (Cohen and Levinthal 1990). Cohen and Levinthal (1990) argued that familiarity, or relatedness, increases the firm’s organizational learning capability because learning, at the most basic level, requires a common language among the joint venture partners. For 27 example, a JV partner from the defense industry is unlikely to share common technical language with a JV partner in the retailing industry. The prior related knowledge of a partner, with respect to the knowledge being developed in the JV, would increase its ability to recognize the value of new information, and then assimilate and apply it throughout the parent organization. Thus, to achieve successful knowledge transfer from the joint venture to the parent, there must be a common ground or relatedness between the joint venture and the parent. In other words, a parent in the same industry as the JV is expected to be able to transfer knowledge from the JV to the parent organization more successfully than a parent in an industry dissimilar to the JV. Therefore, H 11.]: Parent organizations classified in the same industries as the joint ventures (JV -parent related) are expected to achieve higher cumulative abnormal returns than parent organizations in industries dissimilar to their joint ventures (J V-parent unrelated). Codifiability of knowledge is also important to organizational learning capability (Makhija and Ganesh 1997). Knowledge that is highly codifiable can be more easily structured and thus transferred. In contrast, knowledge that is not easily codified is more challenging to acquire and transfer through the parent organization (Makhija and Ganesh 1997; Simonin 1999b). Examples of knowledge that are difficult to codify include new technological breakthroughs, and to a lesser degree, manufacturing, marketing and management processes, especially for those in the high technology industries (Makhija and Ganesh 1997). In high technology industries, there are rapid changes in technology development and thus the necessity of quick preemption strategies (Hagedoom and Schakenraad 1994). Consequently, in a situation in which transferred knowledge is less codifiable, such as when a parent (or both parent and the venture) are in the high 28 technology industries, JV-parent relatedness may play a more pronounced role in achieving successful acquisition and transfer of knowledge. Hence, H II. 2: The difference between CARS of parent organizations classified in the same industries as the joint ventures (JV-parent related) and those classified in dissimilar industries to their joint ventures (JV-parent unrelated) will be greater in high-technology industries than in low— technology industry. Parent-Parent Relatedness Another important aspect of JV strategy is the extent to which partners are engaged in similar businesses (Rumelt 1974; Singh and Montgomery 1987; Merchant and Schendel 2000). Differences between partners adversely affect JV performance because there is little overlap within which to integrate partners’ Skills and capabilities. Moreover, lack of relatedness may accentuate differences in the relative strategic importance of the JV to each partner. In the acquisition context, for instance, Singh and Montgomery (1987) contended, “while the specialized resources in related acquisitions may result in increased efficiencies in technological and product market activities, or increased market-specific market power, the efficiency and power gains in unrelated acquisitions are of a more general variety” (p. 380). Their empirical results supported this claim. Therefore, unrelated partners are expected to be less effective in exploiting technical knowledge and to be viewed less favorably by informed investors in the stock market. In contrast, related partners (i.e., related parents) are expected to provide more strategic and organizational compatibility. Thus, H 11.3: JVs between partners in related industries (parent-parent related) are expected to achieve higher cumulative abnormal returns than JVs among partners in unrelated industries (parent-parent unrelated). 29 Cultural Differences Culture is an important aspect of cross border activity that can significantly influence JV success (Parkhe 1991). Makino and Beamish (1998) suggested that JVS between partners with similar national cultures should experience higher survival rates and performance levels than JVS between partners with dissimilar cultures. Cultural differences between partners can imply different management styles and knowledge management practices. Cultural differences between countries can lead to misunderstandings about the local market, or to prolonged or reduced knowledge acquisition in the host country market, both of which can adversely affect a foreign firm’s performance in the host country (e.g., Parkhe 1991). Cultural differences also affect a firm’s ability to operate with a foreign partner in the joint venture (Barkema, Shenkar, Vermeulen and Bell 1997) and may influence the firm’s learning capabilities (Makhija and Ganesh 1997). Cultural differences may create ambiguities and mistrust in the relationship, which can cause conflict or even terminate the JV (Barkema, Bell and Pennings 1996). It is expected that partner cultural differences will adversely influence organizational learning and hence negatively affect cumulative abnormal returns. Therefore, it is hypothesized that: H II. 4: JVS between partners with no cultural differences are expected to achieve higher cumulative abnormal returns than JVS between partners with cultural difi'erences. 2.4.2 Sample Selection Thompson Financial Security Data (TFSD) was used to identify JV announcements involving manufacturing firms over the period 1997 - 1999. Only JVS between two partners, of which at least one is headquartered in the United States, were 30 selected. This search from TFSD produced 1,300 joint ventures involving 1,665 US companies. After the joint ventures were identified, the Center for Security Price Research (CRSP) database was used to obtain daily stock returns. Observations were excluded if the firm’s stock was not publicly traded, the announcements contained duplicate or missing data, or there were multiple announcements by one firm on the same day. The refinements resulted in 1,015 final observations. 2.4.3 Measurement The focus on relatedness and knowledge transfer required the classification of parent and JV knowledge into high or low codifiability. This research attempted to classify parent and JV as being in a high technology (low codifiability) or a low technology (high codifiablity) industry. Following Calantone and Schatzel (2000), the research classified high technology industries as those engaged in technology innovation, new product development, or both. High technology industries included: chemicals and allied products (major group SIC 28), industrial and commercial machinery and transportation (major group SIC 35), electronic and other electrical equipment and components, except computer equipment (major group SIC 36), transportation equipment (major group SIC 37), measuring, analyzing, and controlling instruments; photographic, medical and optical goods (major group SIC 38). The rest of the industry groups were classified as low technology industries. JV-parent relatedness was measured as a dichotomous variable that equals one if the joint venture and the parent were both in the same industry group, otherwise zero. For example, if Thompson Financial Security Data (TF SD) identified that the parent and the JV have the same first two digits for their SIC code, then they were considered related 31 (and JV-parent relatedness was assigned a value of one). If TFSD identified that the parent and the JV have different first two digits for their SIC codes, then they were considered unrelated (and assigned a value of zero), even though both of them might be in the same category as far as high technology versus low technology is concerned. Similarly, Parent-parent relatedness is a dichotomous variable that equals one if the JV partners were in the same industry group, zero if they were in different industry groups based on the first two digits of their respective SIC codes reported by TF SD. Cultural difference was measured using the Makino and Beamish (1998) typology (Table 2.2), which reflects differences in location of the JV 5 and partners’ countries of origin. The differences in culture were also measured by Cultural distance, which is based on Hofstede (1983) culture dimensions (e.g., Kogut and Singh 1988; Barkema, Bell and Pennings 1996). Specifically, this study uses Kogut and Singh (1988) cultural difference equation: 4 CD,- =2<