The Effects of Formation Motives and Interfirm Diversity on the ...

5 downloads 0 Views 221KB Size Report
oriented formation motives will lead to different level of alliance performance; ..... Thus, similarities between partners help establish trust and also enhance the.
Asia Pacific Management Review (2002) 7(2), 139-166

The Effects of Formation Motives and Interfirm Diversity on the Performance of Strategic Alliance Bou-Wen Lin* and Chung-Jen Chen** (received January 2001; revision received August 2001;accepted May 2002) The number of strategic alliances appears to have increased dramatically in the past two decades. However, prior research has shown, unfortunately, that strategic alliances are fairly unstable and prone to failure. Therefore, there is a considerable interest among managers and scholars in discovering a recipe for successful alliances. Some prior studies implied that formation motives, interfirm diversity, and partner interaction may affect the strategic alliance. However, little has been done on empirically examining their effects on the alliance performance. Accordingly, this study attempts to examine the relationships between these key factors and the alliance performance. Hypotheses are derived based on three academic approaches - strategic behavior, transaction cost, and organizational learning. The findings indicate that different oriented formation motives will lead to different level of alliance performance; that formation motives and partner interaction are positively related to the alliance performance while interfirm diversity negatively affects the alliance performance. Keywords: Strategic Alliance, Formation Motives, Interfirm Diversity, Partner Interaction

1. Introduction In the last two decades, the number of strategic alliances appears to have increased dramatically [24, 28]. More alliances have been announced since 1981 than all prior years combined [2, 56]. Firms have been increasingly willing to participate in strategic alliances for a variety of motives such as attempting to gain access to specific foreign markets and distribution channels, keeping pace of fast changing technology, creating new products, and sharing R&D investment. However, prior research has shown, unfortunately, that strategic alliances are fairly unstable and prone to failure [11, 22, 31, 50, 52]. This leaves managers in a dilemma: they are eager to get the benefits from using these cooperative arrangements; however, on the other hand, they fear that what they get will be a nightmare instead of fulfillment of their expectations. Therefore, there is still a considerable interest among managers and scholars *

Institute of Technology Management, National Tsing-Hua University, 101, Section 2, Kuang Fu Road, Hsinchu, Taiwan, R.O.C., Email: [email protected] Department of Business Administration, National Cheng Kung University, 1, University Road, Tainan, Taiwan, R.O.C., Email: [email protected]

**

139

Bou-Wen Lin and Chung-Jen Chen

in discovering a recipe for successful alliances [32]. Previous alliance research has progressed along several main paths. First, some researchers have focused on the motives of alliance formation [16, 25, 31, 49, 58]. They are intent on identifying the reasons why the firms are willing to cooperate with outsiders, even with their direct competitors. However, little has been done on investigating the effect of the formation motives on the alliance performance [10]. Secondly, some scholars attempt to explore the influence of interfirm diversity on the alliances. Interfirm diversity was recognized as playing a major role in frustrating the joint efforts of interfirm-cooperation partners [31, 32, 69]. Third, some researchers have focused on the interactive nature of cooperation between organizations [14, 28, 35, 45, 62]. According to this perspective, it is the characteristics of the relationship between the firms as an ongoing pattern that should be the focal point for understanding alliance behavior and outcomes [64]. Few studies have been done on exploring the impact of the interaction process on the alliance performance [10]. There has been a lack of empirical attention to the integrative impact of above-mentioned determinants on the alliance performance. This study attempts to integrate these determinants to develop a more complete understanding of their separate and combined effects on alliance outcomes. Accordingly, the purpose of this study is to examine the interrelationships among formation motives, interfirm diversity, partner interaction, and alliance performance for a number of strategic alliances formed by U.S. firms with domestic or international partners. Three academic approaches - strategic behavior, transaction cost, and organizational learning are used to understand the interrelationships among these variables. In this study, strategic alliance was defined as an inter-firm cooperative arrangement over a given economic space and time for the attainment of some strategic objectives. In this definition, strategic alliance involves equity sharing, in particular joint venture and equity investment, and contractual agreements without equity sharing such as licensing, marketing and distribution agreement, manufacturing agreement, R&D agreement, and technology agreement. The rest of the paper is set out as follows. The next section considers the previous literature and sets out the hypotheses of this study. Following that is the methodology for the study and the characteristics of the sample reported. Then, the paper presents the results of the empirical study in achieving the goals as set out above. Discussion and conclusions are provided in the last section.

140

Bou-Wen Lin and Chung-Jen Chen

2. Background 2.1 Theoretical background Theories of the firm are conceptualizations and models of business enterprises which explain and predict firms’ structures and behaviors [27]. However, every theory of the firm is an abstraction of the real world business enterprise which is designed to address only a particular set of firm’s characteristics and behaviors [48]. As a result, many theories of the firm supplement one another in describing different phenomenon or even offer rival explanations on the same phenomena [27]. A similar pattern evolved in the strategic alliance literature. Given the sophisticated nature of strategic alliance, researchers have tried to explore it from different theoretical perspectives such as transaction cost theory, strategic behavior theory, and organizational learning theory. The followings are the review of these theories and their application to the field of strategic alliances. Transaction cost approach (Economics-based view). The concept of transaction cost was first suggested by Coase [13]. He argued that in the real world there is no transaction without transaction cost and that an understanding of transaction cost economizing is central to the study of organizations. The transaction cost approach to the study of economic organization regards the transaction as the basic unit of analysis. A transaction occurs when a good or service is transferred across a technologically separable interface [70]. The two basic behavioral assumptions of transaction cost theory are (1) bounded rationality and (2) opportunism. Moving from initial statements made by such scholars as Williamson, several substantial developments have added to robustness of the transaction cost view in strategic alliance literature. Osborn and Hagedoorn [53] describe three such developments: (1) the expansion in the variety of forms alliances take and in the functions they serve; (2) an elaboration of some key alliance characteristics and the embeddedness of these characteristics; (3) a profound change occurred in the research designs of alliance studies even without explicit theoretical recognition of embeddedness. Strategic behavior approach (Strategy-based view). In contrast to the almost-40-year history of economics-based view, the strategy-based study began only recently [53]. Researchers adopting this perspective view alliances and networks as alternative mechanisms to markets or hierarchies for addressing specific strategic needs to improve firm’s competitive position

141

Bou-Wen Lin and Chung-Jen Chen

and thus gain maximum profits. Specific alliances are generally characterized by partnering firms’ strengths and weaknesses in helping them counter threats and capitalize on opportunities from outside environment to implement their chosen strategies [53]. Resource scarcity forces organizations to enter into more cooperative activities with other organizations [1]. Such cooperation is one way to stretch resources to meet the organization’s needs [5]. In this tradition, alliances are undertaken to secure scarce and valuable resources critical for a firm’s survival and prosperity [64]. Given functional specialization and a scarcity of resources, organizations exchange resources for mutual benefits [8, 23]. This view complements the clear resource and institutional constraints on a firm’s behavior and motives of interfirm cooperation inherent in dynamic models of competition [29]. There are considerable supports for this approach in the literature [5, 21, 33, 41, 44, 58, 67]. Organizational learning approach (Knowledge-based view). In classical economics, the sources of wealth are land, labor, and capital. Now another engine of wealth, namely knowledge, is at work [3]. It takes many forms such as technology, innovation, and know-how. Management researchers have recently begun looking at how superior knowledge can improve firms’ competitive positions [4, 20, 36, 61]. From this knowledge-based point of view, knowledge is treated as a key competitive asset and the firm is conceptualized as an institution for integrating knowledge [15, 26]. As a result, the reason of the existence of the firms for producing goods and services is that they can create conditions under which multiple individuals can integrate their specialist knowledge. Grant [27] came up with the basic characteristics of “knowledge.” They are transferability, capacity for aggregation, appropriability, specialization in knowledge acquisition, and the knowledge requirements of production. In this literature, alliances are an important part of a learning process for firms, a process in which firms can discover new opportunities in a flexible setting of a multitude of changing partnerships [12, 29]. There is a growing research interest on how firms can learn from strategic alliances to acquire and develop their competencies [17, 19, 41, 47, 65]. Sources of sustainable competitive advantages are firm’s resources that are valuable, rare, imperfectly imitable, and non-substitutable [4]. Cooperative relationships can be seen as one means of internalizing core competencies and enhancing competitiveness [60]. In summary, the transaction cost approach [38, 39, 68, 71] assumes that firms transact by the mode that minimizes the sum of production and

142

Bou-Wen Lin and Chung-Jen Chen

transaction costs. However, strategic behavior approach [16] posits that firms transact by the mode which maximizes profits through improving a firm’s competitive position vis-a-vis rivals. Transaction cost and strategic behavior motivation explanations provide compelling economic reasons for strategic alliance. The third rational explanation, organizational learning approach [30, 41], views strategic alliance as a means by which firms learn or seek to retain their capabilities. In this view, firms consist of a knowledge base, which is not easily diffused across the boundaries of the firm. Strategic alliances are a vehicle by which tacit knowledge is transferred. From this perspective, an alliance is encouraged under two conditions: one or both firms desire to acquire the other’s organizational know-how; or one firm wishes to maintain an organizational capability while benefiting from another firm’s current knowledge or cost advantage. The three perspectives of transaction cost, strategic behavior, and organizational learning provide distinct explanations for strategic alliance behavior. Transaction cost approach analyzes strategic alliance as an efficient solution to the hazards of economic transactions. Strategic behavior approach places strategic alliance in the context of competitive rivalry and collusive agreements to enhance market power. Finally, organizational learning approach views strategic alliance as a vehicle which organizational knowledge is exchanged and imitated. Table 1 summarizes basic differences of these three theories in strategic alliance. Table 1 Differences of the three theories in strategic alliance Field Assumption Transaction cost Firms transact by the mode which theory minimizes the sum of production and transaction costs Strategic Firms transact by the mode which behavior theory maximizes profits through improving a firm’s competitive position vis-a-vis rivals. Organizational Firms seek to develop their capabilities learning theory through gaining knowledge inside or outside

Viewpoint Strategic alliance is an efficient solution to the hazards of economic transactions. Strategic alliance is a competitive rivalry and collusive agreement to enhance market power. Strategic alliance is a vehicle by which organizational knowledge is exchanged and imitated.

2.2 Formation motives and alliance performance Why do firms want to participate in strategic alliances, even with their direct competitors? Many prior studies have illustrated the motives of alliance formation [7, 18, 25, 31, 43, 49, 59, 63]. Scholars taking the strategic behavior viewpoint argued that firms would do interfirm cooperation to gain

143

Bou-Wen Lin and Chung-Jen Chen

access to necessary resources from outside to enhance their competence [32]. Accordingly, the use of strategic alliances can be motivated by a number of reasons such as sharing risk, shaping competition, facilitating international expansion, improving market position, and building vertical links [16, 25, 58]. Some scholars pointed out that, from the transaction cost standpoint, firms will have a stronger incentive to use strategic alliances as a means of reducing transaction costs involved in technology transfer/exchange of patents, risk sharing, and creation of vertical links [16, 25]. From organizational learning viewpoint, firms will have strong motivation to create strategic alliances to absorb necessary knowledge from the partners. Therefore, the use of strategic alliances can be motivated by a number of reasons such as technology transfer/exchange of patents [16, 25]. These studies, based on different theories, basically attempted to capture all the possible reasons why firms would cooperate with others. In this sense, prior research provides quite detailed information. However, prior research has left two important questions unanswered. Will the firms, with different motivation joining the alliance, get different levels of alliance performance? Will the degree of motivation affect the alliance performance? Firms with competition-oriented motivation such as competing against common competitors will cooperate more closely and seriously since the partners face the same threat and the alliance may immediately strengthen their competitiveness. On the other hand, alliances with the purpose of technology development may have trouble at the end because of interest conflicts among the partners. To the extent that is the case, it would be expected that alliance performance would vary with the motivation settings. Moreover, no matter what type of motivation for joining the alliance, as the motivation is strong enough, firms will overcome any obstacles in front of the alliances to achieve the goals. In light of the reasoning, the following hypotheses are proposed: Hypothesis 1: Different oriented formation motives will lead to different level of alliance performance. Hypothesis 2: The relative importance of formation motives is positively related to alliance performance. 2.3 Interfirm diversity and partner interaction The interactions between partnering firms in strategic alliances bring together people who may have different patterns of behaving and believing, and different cognitive blueprints for interpreting the world [6]. Thus

144

Bou-Wen Lin and Chung-Jen Chen

interfirm diversity may have important influence on the interaction process. From the organizational learning point of view, for an organization to effectively capture the potential learning embedded in an alliance relationship, it needs to have a common frame of reference. As dissimilar partners may extend time and energy to establish standard managerial routines to facilitate communication, they may incur higher costs and mistrust than similar partners [54]. Also, from transaction cost perspective, opportunism may arise from the dissimilarities between the partners. It is difficult to build trust when there is an atmosphere of opportunism existed in the partnership. Interaction between partners accordingly will not be effective since partners are skeptical on each other. The foregoing arguments lead to the following hypothesis: Hypothesis 3: Partner interaction is more favorable if the degree of interfirm diversity is smaller. 2.4 Interfirm diversity and alliance performance Do differences among sponsoring firms influence the efficacy of their interfirm-cooperation? This is one of the topics that captured the interest of scholars and theorists. Organizational learning theory suggests that similarities between partners may affect alliance performance because they facilitate the appropriability of tacit and articulated knowledge [34, 55, 66]. Thus, similarities between partners help establish trust and also enhance the appropriability of knowledge, in turn increasing the likelihood of a successful alliance [64]. From transaction cost viewpoint, interfirm diversity may lead to partner's opportunistic behavior and thus negatively affect alliance performance [52, 54]. Scholars adopting strategic behavior perspective argued that firms enter partnerships when they perceive critical strategic interdependence with other firms. However, the compatibility of partners may serve as barriers to achieve their strategic objectives [32, 40]. Following this line of reasoning, the following hypotheses are proposed: Hypothesis 4: Interfirm diversity is negatively related to alliance performance. 2.5 Partner interaction and alliance performance From the viewpoint of organizational learning, to absorb the knowledge from the partners smoothly, existing organizational incompatibilities must be compensated. Otherwise, the inherent procedural, structural, and cultural differences between organizations become insurmountable obstacles to a successful cooperation. If partners lack the understanding of each other’s

145

Bou-Wen Lin and Chung-Jen Chen

operating requirements or if they are unwilling to make concessions and meet on a middle ground for cooperation, misunderstandings will result and a lack of support for the relationship will give rise to frustration and disillusionment with the partnership [51]. Unlike arm’s transactions, in which initial commitments govern, alliances require ongoing mutual adjustments. To ensure an effective match, forming an alliance should include adjusting both organizations to the new priorities [46]. According to transaction cost theory, performance generally depends on transaction costs, which in turn reflect level of trust. Without trust, transaction costs tend to be high because more monitoring and safeguards against opportunistic behavior are needed [9]. Buckley and Casson [7] suggested that cooperation is efficient when a given amount of mutual forbearance generates the largest possible amount of mutual trust. Successful alliances rarely depend on formal contracts for compliance. Instead, trust is the necessary ingredient for successful strategic arrangements [42]. Also, poor communications and mutual distrust can make the transfer of management practices and technologies very costly [57]. Collectively, these arguments suggest the following hypothesis: Hypothesis 5: A favorable partner interaction between alliance partners is positively related to alliance performance. 3. Methods 3.1 Data collection and sample The unit of analysis is the strategic alliance. This study involves strategic alliances formed by U.S. firms with domestic or foreign partners. A list of 566 alliance cases was mainly obtained from Predicasts’ Funk & Scott Index of Corporate Change. Other secondary sources such as Mergers and Acquisitions, the Wall Journal Index, Yearbook on Corporate Mergers, and Joint Ventures and Corporate Policy were also used. Based on the recommendation by Glaister and Buckley [25], it is expected that the resulting sampling frame represented a reasonable approximation of the overall population of these strategic alliances, and that any selection bias would be minimal. A survey questionnaire was developed to collect data for testing the validity of the model and research hypotheses. Variables in the questionnaire include alliance background information, formation motives, interfirm diversity, partner interaction, and alliance performance. All of the variables

146

Bou-Wen Lin and Chung-Jen Chen

are measured with multi-items. 5-point Likert scales are used. If there was more than one alliance case for the subject company to report, the company was asked to choose the most significant one. Appendix I presents the questionnaire items of this study. 566 questionnaires were mailed to the American firms in the U.S. Follow-up letters, emails and phone calls were done after two weeks. Of the 566 questionnaires mailed, 22 were undeliverable. 98 responses were received and three of them were incomplete. The remaining 95 valid and complete questionnaires were used for the quantitative analysis. It represented a useable response rate of 17%. Table 2 shows the characteristics of the survey sample. The sample is composed of 95 strategic alliances of which 30 (31.6%) are international alliances and 65 (68.4%) are domestic alliances. 52 (54.7%) of the strategic alliances were formed in the same industry and the rest, 43, (45.3%) were formed between different industries. The duration of 61 (64.2%) alliance cases is 5 years or less and that of 34 (35.8%) is more than 5 years. The number of equity-based alliances is 58 (61.1%) and the rest 37 (38.9%) belongs to contract-based alliances. Finally, the distributions of annual sales and the number of employees of the companies are listed below. Table 2 Characteristics of the survey sample Item

Description

Nationality Domestic International Alliance Equity-based Forms Contract-based Annual Less than $100M Sales $100M-500M $500M-1B $1-5B More than $5B

Frequency Percent 65 30 58 37 13 14 24 23 21

68.4 31.6 61.1 38.9 13.7 14.7 25.3 24.2 22.1

Item Industry

Description

Same Different Duration ≦ 5 years > 5 years Number of Less than 100 Employees 100-500 500-3000 3000-10000 More than 10000

Frequency Percent 52 43 61 34 13 12 16 31 23

54.7 45.3 64.2 35.8 13.7 12.6 16.8 32.6 24.3

The industry categories of the alliances are as follows: food & kindred products (5.3% of the total), apparel & related products (7.4%), chemicals products (10.5%), primary metal products (6.3%), machinery (4.2%), electronic & electric equipment (9.5%), transportation equipment (5.3%), instruments & related products (2.1%), communications (13.6%), utilities (14.7%), financial services (5.3%), software (8.4%), and other services (7.4%). In total 48 (50.6%) of the alliances are in the manufacturing sector and 47 (49.4%) are in the tertiary sector.

147

Bou-Wen Lin and Chung-Jen Chen

3.2 Measures Formation motives. Based on prior literature, eleven major motives for alliance formation are identified. They are “sharing R&D costs and risks”, “exchange of complementary technology”, “exchange of patents/territories”, “competing against common competitors”, “maintaining market position”, “reducing competition”, “facilitating international or domestic expansion”, “gaining presence in new market”, “conforming to foreign or domestic government policy”, “gaining economies of scale” and “lowering production cost.” Respondents were asked about the importance of these motives for the alliance formation. Responses were assessed using 5-point Likert scales. Interfirm diversity. Four indicators are included in this study for the interfirm diversity construct. They are cultural difference, strategic divergence, managerial and organizational difference, and size difference. Two items are included in the cultural difference indicator. They are perception of national cultural difference and perception of organizational cultural difference. The size difference variable measures the size difference between subject company and the major partner in terms of the number of employees and of the annual sales. The strategic divergence and managerial and organizational difference are reflected by perceived differences in strategic directions between subject company and the major partner and perceived differences in management practices and organizational structure between subject company and the major partner respectively. Partner interaction. The major indicators regarding partner interaction are trust, communication, conflict, and adjustment. The managers’ perceptions of these interaction indicators were collected through survey questions. The trust indicator is measured by perceived degree of trust of the subject company on its partner. The perceived degree of conflicts between subject company and its partner presents the degree of conflict. The communication indicator is reflected by perceived degree of communication between subject company and its partner. The adjustment indicator is measured by perceived degree of adjustment subject company made. Alliance performance. In this study, the alliance performance indicator, perceived satisfaction, includes three items using five-point Likert scale to report performance as perceived by the subject firm in the alliance. These three items reflect the degree to which the alliance has met the goals of the subject firm, the degree to which it has met the expectation of the subject firm, and its contribution to the subject firm’s core competencies.

148

Bou-Wen Lin and Chung-Jen Chen

4. Results 4.1 Factor analysis of key constructs Factor analysis is normally used to uncover any dimensions or structure underlying the data. Its purpose is to highlight key features that might otherwise be obscured by detail and in this way to simplify and clarify analysis. A principal component analysis followed by varimax rotation was used in this study. The followings describe the factor analysis for those key variables including formation motives, interfirm diversity, partner interaction, and alliance performance. Table 3 shows that factor analysis of formation motive items produced four underlying factors from eleven formation motive items. The results of factor analysis explain a total of 73.16 percent of the observed variance. Four measures in market & economics development factor include facilitating international or domestic expansion, gaining presence in new market, gaining economies of scale, and lowering production cost. Three measures in the shaping competition factor are competing against common competitors, maintaining market position, and reducing competition. Three measures including sharing R&D costs and risks, exchange of complementary technology, and exchange of patents/territories are categorized into the technology development factor. The fourth factor, regulation, includes only one measure which is conforming to government policy. The factor scores of the formation motive factors for each sample were used in the following statistical analysis. For interfirm diversity construct, Table 3 shows that factor analysis of the six interfirm diversity items produced two underlying factors. Measures of national culture difference, organizational culture difference, strategic divergence, and managerial and organizational difference are categorized into one factor called culture and management difference. Two size related measures are in the size difference factor. The results of factor analysis explain a total of 70.42 percent of the observed variance. The factor scores of the interfirm diversity factors for each sample were used in the following statistical analysis. For partner interaction construct, factor analysis of the four partner interaction items produces only one underlying factors called interaction. The results of factor analysis explain a total of 57.00 percent of the observed variance. The factor scores of the partner interaction factor for each sample were used in the following statistical analysis.

149

Bou-Wen Lin and Chung-Jen Chen

Table 3 Results of factor analysis and Cronbach’s alpha Constructs Strategic Motive

Interfirm Diversity

Factors and Items

Total Alpha Variance 2.28 20.79% 0.70

Loadings Eigenvalue

Market & Economics Development Facilitating international or domestic expansion Gaining presence in new market Economies of scale Lower production cost Shaping Competition Competing against common competitors Maintaining market position Reducing competition Technology Development Sharing R&D costs & risks Exchange of complementary technology Exchange of patents/territories Regulation Conforming to government policy Culture & Management Difference There were major differences in strategic directions between your company and the partner. There were major differences in management practices and organizational structure between your company and the partner. There are major national cultural differences between your country and the partner’s country. There are major organizational cultural differences between your company and the partner.

0.836 0.845 0.547 0.637 2.22

40.97% 0.70

2.10

60.08% 0.74

1.43

73.16%

2.21

36.92% 0.73

2.01

70.42% 0.97

2.28

57.00% 0.68

2.05

68.33% 0.83

0.769 0.855 0.673 0.888 0.901 0.575 --

0.910 0.639 0.719 0.751 0.849

Size Difference What is the difference between your company and the partner in terms of annual sales? What is the difference between your company and the partner in terms of employees? Partner Partner Interaction Interaction During the alliance period, you feel that the partner was trustworthy. During the alliance period, the communication between your company and the partner was effective. During the alliance period, conflicts occurred very frequently between your company and the partner. During the alliance period, your company made huge adjustments to make the alliance succeed. Performance Alliance Performance This alliance has realized the planned objectives your company set out to achieve. The performance of this alliance is better than what you expected. This alliance has contributed to the core competence and competitive advantages of your company.

150

0.980 0.978

0.657 0.677 0.789 0.605

0.894 0.851 0.727

Bou-Wen Lin and Chung-Jen Chen

For the alliance performance construct, the results of factor analysis indicate that all of the three performance indicators belong to one factor as shown in Table 3. The results of factor analysis explain a total of 68.33 percent of the observed variance. The factor scores of the alliance performance factor for each sample were used as the scores of the overall alliance performance in the following statistical analysis. The results of alpha coefficients of major factors are shown in Table 3. Most of the alpha reliability coefficients are effective and higher than 0.70, except the one for partner interaction factor, which is 0.68. 4.2 Rank of formation motives The rank order of the formation motives based on the means of the importance is shown in Table 4. For the full set of strategic alliances, the market & economics development factor, with means 3.48, is the most important factor for firms to create strategic alliances. The means of the four motive items in the market & economics development factor all exceed the median value 3. The second important motive factor is the shaping competition factor with means 3.06. The means of two of the three motive items in this factor are larger than median value 3. The technology development motive factor with means 2.79 is the third important factor for firms to create strategic alliances. Its three motive items rank 7, 8, and 9 respectively and the means are all less than median value 3. Regulation factor is the least important factor with means 2.24. Table 4 Rank of formation motives by mean of importance Motive Factor (Mean/SD) Market & Economics Development (3.48/1.21)

Motive

Rank

Mean

SD

Facilitate international or domestic expansion Gain presence in new market Economies of scale Lower production cost

1 3 5 4

3.81 3.58 3.19 3.35

1.21 1.31 1.13 1.17

Shaping Competition (3.06/1.05)

Compete against common competitors Maintain market position Reduce competition

6 2 10

3.02 3.64 2.52

1.19 0.93 1.03

Technology Development (2.79/1.19)

Share R&D costs and risks Exchange of complementary technology Exchange of patents/territories

7 8 9

2.95 2.87 2.56

1.21 1.27 1.09

Regulation (2.24/1.06)

Conform to government policy

11

2.24

1.06

151

Bou-Wen Lin and Chung-Jen Chen

4.3 Creating formation motive clusters The next step of the analysis focuses on creating homogenous motivation clusters by using cluster analysis. Three clusters are identified as shown in Table 5. To understand these clusters, multivariate analysis (MANOVA), analysis of variance (ANOVA), and Scheffe tests of difference in group means were conducted. MANOVA shows that the three clusters are significantly different (F= 39.97, p