A mediation model between dimensions of social capital

Depósito de Investigación de la Universidad de Sevilla https://idus.us.es/ This is an Accepted Manuscript of an article published by Elsevier in International Business Review, Volume 22, Issue 6, December 2013, available at: https://doi.org/10.1016/j.ibusrev.2013.02.004. Copyright 2013 Elsevier. En idUS Licencia Creative Commons CC BY-NC-ND Elsevier Editorial System(tm) for International Business Review Manuscript Draft Manuscript Number: 2522R2 Title: A MEDIATION MODEL BETWEEN DIMENSIONS OF SOCIAL CAPITAL Article Type: Full Length Article Keywords: Social capital, structural dimension, relational dimension, resource dimension, international market share, partial least squares, mediation analysis. Corresponding Author: Dr. Ignacio Castro, Corresponding Author's Institution: University of Seville First Author: Ignacio Castro Order of Authors: Ignacio Castro; José L Roldán A MEDIATION MODEL BETWEEN DIMENSIONS OF SOCIAL CAPITAL Ignacio Castro (corresponding author) Faculty of Economics and Business Administration Department of Business Administration and Marketing. University of Seville. Av. Ramón y Cajal, 1, 41018. Seville, Spain. Phone: +34-954554432 Fax: +34-954556989 [email protected] José L. Roldán Faculty of Economics and Business Administration Department of Business Administration and Marketing. University of Seville. Av. Ramón y Cajal, 1, 41018. Seville, Spain. Phone: +34-954554458 Fax: +34-954556989 [email protected] Author Biographies Dr. Ignacio Castro is an Associate Professor of Management at the Faculty of Economics and Business Administration, University of Seville. He received his Ph.D. from the University of Seville. His main research interests are inter-organisational relationships, social capital, strategic alliances and innovation. His latest research works are focussed on strategic alliances in the Public Works Industry. He has published in Construction Management and Economics and Journal Business Research. is Associate Professor of Management at the Faculty of Economics and Business Administration, University of Seville. His recent contributions have been published in European Journal of Information Systems, International Small Business Journal, Computers in Human Behavior, Handbook of Partial Least Squares, Industrial Marketing Management, International Journal of Technology Management, and Internet Research. He is currently on the editorial board of The Database for Advances in Information Systems, and was guest editor of European Journal of Information Systems’ Special Issue on Quantitative Research Methodology. His research interests include knowledge management, organisational culture and PLS methodology. Title Page (WITH author details) Acknowledgements: This research was supported by the Ministerio de Ciencia e Innovación, Spain (ECO2009-12742). We are greatful to Francisco Acedo, José Luis Galán, and the three reviewers of International Business Review for their helpful comments and suggestions. Dear Editor, We would like to thank you for giving us the opportunity to introduce minor revisions and changes, as suggested by the reviewers, throughout the article A Mediation Model Between Dimensions of Social Capital submitted to your journal International Business Research. We feel that we have digested and fully responded to the valuable suggestions from the reviewer. In response to their comments, we have introduced the following alterations to our article:  Following the suggestions of Reviewer#4, we have attempted to soften our statement in relation to the inexistence of other works on mediation in the study on social capital at an interorganizational level. Likewise, we have tried to clarify the mediating and non-moderating nature of our empirical work.  Taking the comments of Reviewer#4 into account, we have emphasized and have tried to give a certain relevance to the concept of dynamic capacity in the article, in the section on the relational dimension and, more specifically, linked to the alliance experience.  At the suggestion of Reviewer#4, we have introduced two illustrations that are appended to the former review letter and a short text with the aim of better clarifying the empirical context under analysis. We hope that the editor will appreciate the many aspects of this work that have been improved, in the light of the full and complete answers given to the reviewer’ comments and the alterations made to the text. Best wishes The authors Dear Editor RESPONSE TO COMMENTS FROM REVIEWER #4 First of all, we would once again like to offer our sincere thanks to reviewer #4, for so many valuable and constructive criticisms and comments, which have given us the best possible insight and advice to help us improve the theoretical discussion in our paper. Accordingly, we have tried to respond to all the considerations and suggestions made by the reviewer. We have reproduced the comments made by Reviewer #4 below in bold type. In each case, we provide point-by-point responses, indicating where the information may be found in the text, which we hope will make it easier to follow the changes made: #1.1. First of all I would like to congratulate with the authors because they did a very good job in replaying to the issues I have raised in my previous review, and as a matter of fact, they succeeded in convincing me in most of the points I have highlighted in the previous review. Only minor revisions and suggestions are further needed to make the paper publishable in IBR. R: The authors would like to thank the reviewer for the reassuring words and assessment of the effort made to improve the investigative work. In addition, we consider that all the points so perspicaciously raised by the reviewer have substantially improved our research. For this reason, we especially wished to acknowledge the significant contribution of the anonymous reviewers in the acknowledgements section of the article. #1.2. At pages 3 of the manuscript reading the paragraph "Accordingly, Koka.... (Hayes et al., 2011)" the reader understands that this is the first paper analysing the interaction (or the mediation) of several network dimension in an interorganisational context. This is not true. Indeed, there are two recent papers from Shilling and Phelps (2007) and Phelps (2010) analysing interactions among different network dimensions also in an inter-organisational context. However, the mentioned papers are quite different from the topic of the paper here, therefore authors should only be less definitive in their sentences and giving proper acknowledgment to the mentioned papers. Detailed Response to Reviewer#4 R: Thank you for this advice, which has led us to soften our statement. We have dropped the expression “However, no other research work has analyzed the different roles that each of the dimensions play in a global interorganizational context”, replacing it with “There is, therefore, an open area in which to examine the existing relationships between the dimensions of social capital in a global interorganizational context.” On the other hand, we would like to clarify our position concerning our contribution. To our knowledge, our paper is the first study that analyzes the direct and indirect effects between the dimensions of social capital in an interorganizational context. These indirect effects are tied up with the concept of mediation, but not with moderation (interaction). Although, in the words of Baron and Kenny (1986, p. 1173) “it is not at all uncommon for social psychological researchers to use the terms moderator and mediator interchangeably”, both concepts are very different. A mediating relationship is an intervening construct (Z) that sits between two other constructs and passes/mediates the influence of the one variable (X) through to the other (Y) (Chin, Marcolin and Newsted, 2003). Thus, an indirect effect (also called mediating effect or mediation) occurs when an independent variable, X, has an impact on a third variable, Z (mediating variable or mediator), which then influences the dependent variable, Y (Hayes, 2009; Henseler and Fassott, 2010) (Figure 1). On the other hand, a moderating effect (also called interaction effect) implies the existence of a moderator (or moderating variable) that affects the direction and/or strength of the relation between an independent (i.e., predictor) variable (X) and a dependent or criterion variable (Y) (Chin, Marcolin and Newsted, 2003; Henseler and Fassott, 2010) (Figure 1). Figure 1. An example of mediated and moderated relationship model. Source: Adapted from Hayes (2009) and Henseler and Fassott (2010). In accordance with Hayes (2009, p. 409), our model particularly represents a multiplestep multiple mediator model (Figure 2), that is, a complex version of the simple mediation model described in Figure 1. Figure 2. A Multiple-step multiple mediator model. Source: Adapted from Hayes (2009). References: Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Chin, W. W., Marcolin, B. L., & Newsted, P. R. (2003). A Partial Least Squares latent variable modeling approach for measuring interaction effects: Results from a Monte Carlo simulation study and an electronic mail emotion/adoption study. Information Systems Research, 14(2), 189–217. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408-420. Henseler, J., & Fassott, G. (2010). Testing moderating effects in PLS path models: An illustration of available procedures. In E. V. Vinzi, W. W. Chin, J. Henseler, & H. Wang (Eds.), Handbook of PLS and Marketing (713–736). Springer: Berlin. #1.3.At page 8 where authors refer alliance experience (just after Winter 2002) authors should mention that alliance experience not only generate reputation and trust, but, according to the dynamic capability stream of the literature, also the ability to better manage new inter firm relationships (routine, dynamic capabilities, and so forth). So some relevance should be given to the dynamic capability concept in here. R: We thank the reviewer for the suggestion of emphasizing the importance of the accumulation of alliance experience in relation to the concept of dynamic capability. Thus, we have introduced the following paragraph, in addition to some bibliographic references that will help to support our arguments: “Equally, many researchers (Anand and Khanna, 2000; Gulati et al., 2009; Zollo and Winter, 2002) have shown how a firm’s experience in managing alliances leads it to develop relational capabilities which are of a dynamic nature (Helfat and Winter, 2011). Accordingly, Zollo and Winter (2002) pointed to the accumulation of experience as an important learning mechanism that underlies the generation and evolution of dynamic capabilities (Zollo and Winter, 2002). So, experience of alliances constitutes a fundamental factor in the creation and development of organizational routines and, in consequence, dynamic capabilities, avoiding errors and continuously improving managerial effectiveness in inter-organizational relations (Schilke and Goerzen, 2010). Along the same lines, Blyler and Coff (2003) pointed out that social capital, and more concretely the strength of the ties and alliance experience, facilitate the acquisition, Dimensions of Social Capital and International Market Share 3 that constitute the network, the ties that connect the nodes, and the pattern or structure that results from these connections. Finally, various authors (Koka and Prescott, 2002; Nahapiet and Ghoshal, 1998; Wu, 2008) conceive of social capital as a multidimensional concept. Therefore, the above-mentioned essential elements are covered in our research through our definition of social capital as the structural characteristics of the network (structural dimension), the components of the alliance such as trust (relational dimension) (Nahapiet and Ghoshal, 1998; Tsai and Ghoshal, 1998) and the features of partner firms such as partner resources (resource dimension) (Batjargal, 2003; Rivera-Santos and Inkpen, 2009). Once the social capital is defined, a further question is how it has been analyzed. Accordingly, Koka and Prescott checked the convergent and discriminant validity of the social capital construct, but they did not study the interrelationships between its dimensions. Tsai and Ghoshal (1998), on the other hand, proposed a model of social capital with direct relationships in an intraorganizational context. There is, therefore, an open area in which to examine the existing relationships between the dimensions of social capital in a global interorganizational context. The present study proposes a multiple mediating model and tests the mediation hypotheses using bootstrapping methods (Hayes et al., 2011). In particular, we posit that the relational and resource dimensions play a mediating role in the relationship between structural dimension/international market share. This analysis leads us to a better understanding of the internal functioning of social capital and helps us to reveal its strategic configuration. This is important because managers must not only take account of individual dimensions but how one dimension affects the others dimensions. In this sense, the optimal decisions of the focal action in the choice of partners and in the development of network structure and relational patterns is likely to result in stronger partner relationships (Sarkar et al., 2009). Likewise, the relational governance of the strategic alliances of the focal actor will facilitate access and mobilization of the valuable resources of the alters (Sarkar et al., 2009). Thus, what inhibits a firm more than anything else from establishing cooperative agreements is opportunism, in other words, the fear that partners will unscrupulously protect their own self-interest. Such problems concerning safeguards are aggravated when partners have to make exchangespecific investments and the actors involved have an unequal or unbalanced endowment of resources (Rindfleisch and Heide, 1997). These risks may be minimized through the relational dimension, where innovation may be developed beyond the limits of contractual clauses, resources may be mobilized, and fine-grained tacit and explicit knowledge may be shared to a greater extent (Sarkar et al., 2009), even though both formal and informal safeguards Dimensions of Social Capital and International Market Share 4 constitute appropriate mechanisms. Along these same lines, Walker et al. (1997) contend that firms that do not manage the relational dimension of social capital well are more vulnerable to opportunistic behavior and less able to build an enduring history of effective cooperative behavior with their partners over time. Thus, these firms will spend a lot of time and energy on controlling the relation instead of looking for synergies and complementarities. In conclusion, the analysis that we propose will give us a much better understanding of the composition of social capital which will assist managers with its management and the benefits that may be gained from it. Finally, some studies have used alliance social capital in international contexts (Gulati et al., 2009; Malik, 2012; Wu, 2008). Specifically, Zhou et al. (2007) pointed out that outward internationalization requires the development of alliances with foreign business to learn about new technologies and the needs of international markets, as well as about the gains related to the scale and scope of economies achieved by revenue growth in the geographic extension of markets. We analyzed the proposed model in the global contractors industry, an empirical context in which the phenomenon of co-opetition (or cooperative competition), in order to improve the international market share, is very relevant. Accordingly, some researchers have studied the influence of social networks in achieving international performance focusing on the SMEs' structural dimension (Agndal et al., 2008; Coviello and Munro, 1997; Johanson and Mattsson, 1988; Lindstrand et al., 2011; Prashantham, 2011; Presutti et al., 2007; Zhou et al., 2007). Moreover, although the literature shows that the international market share is one of the most relevant for the international activity of MultiNational Companies (MNCs), few investigations have analyzed how the composition and management of such a critical resource as social capital impacts on this result (Kauser and Shaw, 2004). Taking account of the above insights, the objective of this study is to investigate a central question: how do the relational and resource dimensions mediate the relationship between the structural dimension and the international market share? In particular, we want to understand how different social capital dimensions generate increased value for the firm in an inter-organizational context, specifically by influencing the international market share. Secondly, we wish to explore and test the likelihood that relational and resource dimensions sequentially mediate (e.g., a three-path mediation model; cf. Hayes, 2009; Taylor et al., 2008) the relationship between the social capital’s structural dimension and the international market share. With these objectives in mind, this paper is organized as follows. We begin with a review of the literature on social capital, in order to identify and measure each of the Dimensions of Social Capital and International Market Share 5 dimensions that constitute the construct, as well as the impact of each one on the international competitive position of firms. We propose a multiple mediating model that shows the dimensions of social capital. We then present our results, and a discussion, based on the analysis of data from a sample of 225 top international contractors (Sarkar et al., 1998). This study concludes with a discussion of the results, their implications, the limitations of the study, and suggested future research lines. 2. Literature Review and Research Hypotheses 2.1. International market share In an increasingly competitive and global environment, firms have to grow constantly and improve their performance. One way firms can do so is to move into international markets to expand their potential market share. Accordingly, ISAs are increasingly being used to provide firms with specific advantages, such as a fast access to distant countries at a reduced risk (Nielsen, 2007; Swoboda et al., 2011). In this sense, Guler and Guillén (2010) have recently shown how domestic (local) networks affect international expansion, contributing to the growing body of evidence that documents the influence of social networks on a firm’s performance (Dussauge et al., 2004). However, few studies – if any - examine how these networks affect established MNCs and their international performance. There are many ways to try to measure international performance. One of the many proposals made is international market share. This is one of the most relevant for the international activity of MNCs (Kauser and Shaw, 2004). A firm’s market share, a non-financial performance indicator widely accepted in both general business (Combs et al., 2007) and international business (Depperu and Cerrato, 2005) strategy, reflects its competitive position. Competitive advantage in the global industry that leads to a higher international market share is a critical aim for most firms (Hoskisson et al., 2008). To reach this objective, MNCs must strategically manage the dimensions of social capital. Hence, this requires an understanding of the interrelationships between them. In this paper, we have adopted a multiple mediating approach. This allows us to analyze how the relationship between the structural dimension and the market share is mediated by the relational and the resource dimension. Dimensions of Social Capital and International Market Share 6 2.2. The relationship between the structural dimension and international market share The structural dimension of social capital attempts to include all the interactions that exist between the different nodes that form a network. In general, network density is rarely uniform throughout a network. There are two main approaches to network position. On the one hand, Coleman (1988) posited that the greater the cohesion within the network in which the focal firm is situated and the more central its position in the network, the greater the potential benefits that a firm can gain (Ahuja, 2000a; Ahuja, 2000b; Cowan and Jonard, 2006; Dyer and Nobeoka, 2000; Gulati, 1995a; Houghton et al., 2009; Partanen et al., 2008). On the other hand, Burt (1992) suggested that firms that occupy favorable brokering or entrepreneurial positions within a network can secure important advantages, by exploiting their privileged relationships with isolated colleagues (Houghton et al., 2009; Sapsed et al., 2007). Thus, the existence of structural holes implies an asymmetry of access to information within the network (Burt, 1992; Shipilov, 2006; Ven, 1986). Although some authors consider these as two independent approaches, we regard them as complementary, because it makes sense to view them as being linked and inversely related or at both ends of the continuum (Koka and Prescott, 2008; Vanhaverbeke et al., 2009). We shall focus on centrality in this work, while drawing on relevant research into interfirm networks from international industries that support Coleman’s position (Ahuja, 2000a; Lorenzoni and Ornati, 1988; Lorenzoni and Baden-Fuller, 1995). A centrally-located firm is generally considered to be well positioned in a network (Coleman, 1988; Inkpen and Tsang, 2005; Wasserman and Faust, 1994), as this position mainly implies two benefits. Firstly, a centrally-located firm can access a variety of resources including tacit knowledge and explicit information (Bell, 2005; Powell et al., 1996). Central firms are predicted to have significant access to benefits because they possess direct and indirect ties with multiple firms and many of their partner firms are prominent in the network. Access arguments view network ties as conduits providing timely information, knowledge, and resources (Batjargal, 2003). Secondly, a central firm can achieve affiliation benefits (Coleman, 1988). Observers tend to consider that firms have certain attributes on the basis of their associates. Network centrality thus provides a mechanism for signaling legitimacy and reputation (Podolny, 2001). These theoretical considerations on central network positions that enhance firm performance have, in general, been confirmed by empirical findings. Ahuja (2000a) found that centrally-located firms with a large number of direct and indirect ties achieved better results than firms in less favorable positions. Other researchers have stated that firms that occupy more central positions or that Dimensions of Social Capital and International Market Share 7 show greater cohesion within their ego-networks achieve higher market shares (Baum and Ingram, 2002; Gnyawali and Madhavan, 2001; Hoffmann, 2007). Besides, Vanhaverbeke et al. (2009) argued that a network closure strategy seems to work well for firms that are principally centered on the innovation of core technologies (exploitation) –such as the case of MNCs which enjoy relatively mature and well-developed operating routines (Johanson and Vahlne, 2009; Prashantham and Floyd, 2012) and that operate in a mature sector i.e. the construction sector-, and therefore require no structural holes, unlike SMEs (Zhou et al., 2007). In accordance with these ideas: H1. The more closure there is in the structure of the alliance portfolio, the greater centrality of the focal actor and the greater its international market share of MNCs. 2.3. The mediating role of the relational dimension Many researchers (Tsai and Ghoshal, 1998) propose that the structural dimension is the prime generator of social capital and influences the other components. In this paper, we seek to show that the other components (relational and resource dimensions) play a mediating role. A relationship between actors can be characterized in terms of the strength of their social ties, their level of trust (Dhanaraj et al., 2004; Lee, 2009) In accordance with Granovetter’s idea of weak versus strong ties, social capital’s relational dimension aims to show how economic actions are affected by the quality of relations (Granovetter, 1985; Granovetter, 1992). Therefore, this dimension refers to the characteristics and attributes of relationships such as trust and other complex incentives (quality of the relationships) which are mainly derived from the company’s history and reputation (Granovetter, 1992; Gulati et al., 2000). Previous studies have observed that when two actors interact over time, their relationships become stronger and the actors are more likely to view each other as trustworthy (Granovetter, 1985; Gulati and Gargiulo, 1999; Tsai, 2001). Gulati (1995a) posited that past strategic alliances affect the firms’ future cooperative activity. This occurs because the tendency to share resources also rises (Griffith and Harvey, 2004; Puhakka, 2006), as trust increases and becomes embedded in the relationships. In this vein, various researchers (Doz, 1996; Gulati, 1995a; Gulati, 1995b; Koka and Prescott, 2002) consider that trust is the critical factor in the creation and transference of knowledge and other resources. This is because it decreases the likelihood of developing opportunistic relationships and allows greater fluidity in the sharing and combination of resources (Eapen, 2012). Multiple Dimensions of Social Capital and International Market Share 8 experiences at managing cooperative relationships across firm boundaries generate opportunities and the leveraging of alliance capabilities with other partners (Koka and Prescott, 2008). Therefore initial alliances lead to reputation effects and trustworthiness that facilitate the creation of more ties, not only with each other, but also with other partners. This enhances network cohesion and strengthens their position (Powell et al., 1996; Tsai and Ghoshal, 1998). Equally, many researchers (Anand and Khanna, 2000; Gulati et al., 2009; Zollo and Winter, 2002) have shown how a firm’s experience in managing alliances leads it to develop relational capabilities which are of a dynamic nature (Helfat and Winter, 2011). Accordingly, Zollo and Winter (2002) pointed to the accumulation of experience as an important learning mechanism that underlies the generation and evolution of dynamic capabilities (Zollo and Winter, 2002). So, experience of alliances constitutes a fundamental factor in the creation and development of organizational routines and, in consequence, dynamic capabilities, avoiding errors and continuously improving managerial effectiveness in inter-organizational relations (Schilke and Goerzen, 2010). Along the same lines, Blyler and Coff (2003) pointed out that social capital, and more concretely the strength of the ties and alliance experience, facilitate the acquisition, integration, recombination and release of resources for firms, - the key tasks that Eisenhardt and Martin (2000) have identified. Therefore, firms that have had significant alliance experience are more likely to enjoy access to valuable partner resources (Koka and Prescott, 2002), which will improve the value of those gained in their subsequent alliances. In the internationalization process of foreign expansion, firms require swift access to the many resources embedded in that network (i.e., information, knowledge, reputation, tangible resources, etc.). This is especially so in global industries where competitiveness is very intense. Accordingly, trust plays a critical mediating role in facilitating the access and exchange of resources and capabilities through diverse network channels (W. Tsai and Ghoshal, 1998). Other researchers show that trust is very important to acquire the more tacit and ambiguous knowledge and competencies that positively influence a firm’s international performance (Costa e Silva et al., 2012; Liu et al., 2010; Mu et al., 2008; Nielsen, 2007). Finally, in a case study on Siemens, Hoffmann (2007) argued that the more a firm implements a shaping and stabilizing strategy - characterized by intensive strengthening of inter-organizational relationships -, the greater the growth of its international market share. Therefore, a firm that occupies a central position in the network will have multiple conduits through which the resources will flow. However, if these relationships are characterized by trust and strength, Dimensions of Social Capital and International Market Share 9 this trust will be an important social lubricant (Rivera-Santos, 2009, p. 208) for the flow of valuable resources via different network ties (Tsai and Ghoshal, 1998). In this case, organizations can improve their international market share more easily. All in all, it is logical to think that the focal actor’s social structure can affect both trust and perceived trustworthiness (Theingi et al., 2008; Tsai and Ghoshal, 1998), raising the likelihood of increasing a firm’s international market share. Accordingly: H2. The relational dimension positively mediates the relationship between the structural dimension and the firm’s international market share. 2.4. The mediating role of the resource dimension Some researchers (Ahuja et al., 2012; Wassmer, 2010) consider that one of the fundamental pillars of the architecture of any interfirm network resides in the nodes that constitute the network and, in particular, in their characteristic attributes. This could build up value for the focal actor, as these characteristics are sometimes considered as more important than the absolute number of alliances (Jiang et al., 2010; Stuart, 2000). So, Mu et al. (2008) conceive of social capital as resources, knowledge, and capability accessed through a social network, and resources mobilized through ties in the network. Therefore, the resource dimension of social capital refers to the extent to which the contacts within the network possess valuable resources (Batjargal, 2003; Lin, 1999). While the traditional RBV has analyzed the resources found within the boundaries of a firm that and are owned and controlled by the focal firm (Barney, 1991; Wernerfelt, 1984), some researchers have proposed an extension of the RBV of the firm, so as to take the external network resources and capabilities into account (Huggins and Johnston, 2010; Lavie, 2006; Wassmer and Dussauge, 2011). Agndal et al. (2008) referred to this dimension as the economic dimension of social capital which will enable the focal actor to access both intangible resources, such as knowledge, and tangible resources, such as the means of production controlled by its partners (Hakansson and Snehota, 1995; Jansen et al., 2011). For their part, Gulati (1999) refers to network resources as those resources which the focal actor can access owing to the participation of a firm in interorganizational networks. According to Lavie (2008: 548) “network resources are assets that are owned by the firm’s partners but can potentially be accessed by the firm through its ties to these partners”. There are two critical points to Dimensions of Social Capital and International Market Share 10 know if firms are to be able to mobilize their contacts’ resources. Firstly, they must be aware of the resources that exist within the network (Batt and Purchase, 2004; Westerlund and Svahn, 2008). Secondly, there must be a social structure, as the relationships form conduits through which resources such as knowledge or financial support will flow (Partanen et al., 2008; Puhakka, 2006; Tsai and Ghoshal, 1998; Westerlund and Svahn, 2008). Thus, the creation of inter-organizational networks can be motivated by a firm's intention to access and to make use of partner resources (Håkansson et al., 1993). In this case, this dimension becomes a critical explanatory factor of inter-organizational networks (Hsieh and Tsai, 2007). For some authors, the main benefit of social capital is to gain access to valuable resources (Arenius, 2002; Autio et al., 2005; Huggins and Johnston, 2010). In particular, firms that are centrallylocated and hence have more ‘neighbors’ in a network (structural dimension) are more effective at gathering knowledge (Cowan and Jonard, 2009), can access a variety of resources and those neighbors which are valuable (Bell, 2005; Koka and Prescott, 2008; Powell et al., 1996) and share the fluidity of resources (Ahuja, 2000a). Gnyawali and Madhavan (2001) stated, in the international literature, that improved access through inter-organizational relationships and resource-advantage frequently enable companies to increase or to maintain their international market share. Accordingly, MNCs have made extensive use of international cooperative agreements, partnerships and joint ventures, which reduce the high level of risk involved in their leveraged strategies, in combination with internalized operations to achieve the desired coordination of resources, activities and routines (Mathews and Zander, 2007). Finally, the relation between resource dimension and international performance has been widely studied in the literature that argues the RBV (Chen, 1999; Lim et al., 1993; Woodcock et al., 1994). We therefore propose that the resource dimension is a critical mediator in the relationship between centrality and international market share. Accordingly, firms occupying favorable central positions in international networks will be able to access higher and critical resources. They will therefore increase their international performance. All things considered, it is logical to think that the focal actor’s social structure can affect access to valuable resources (Batjargal, 2003; Theingi et al., 2008; Tsai and Ghoshal, 1998; Vapola et al., 2010), thus raising the likelihood of increasing a firm’s international market share. Based on this logic and previous research, we hypothesize the following: Dimensions of Social Capital and International Market Share 11 H3. The relationship between the structural dimension and international market share is mediated positively by the resource dimension. In this work, we propose a multiple mediating model in which the structural dimension, in particular centrality, is the prime generator of social capital (Tsai and Ghoshal, 1998) and the relational and resource dimensions are critical mediators - although the former precedes the latter. On this point, Sarkar et al. (2009) affirmed that the decisions taken by the focal actor to configure the alliance portfolio structure are likely to result in stronger relationships with partners because the focal firm is likely to select those alters that are best adapted to its objectives and expectations and that strengthen its central position in the network. In this way, the decisions taken by the focal actor with respect to questions of the structural dimension would impact in a positive way on the strengthening of relations and, therefore, on the relational dimension. On the other hand, Gulati (1995a) posited that past strategic alliances affect the firm's future cooperative activity. This occurs because as trust increases and becomes embedded in the relationships, the firm's tendency to access and to share resources increases (Puhakka, 2006). Thus, trust can be considered in terms of a risk-reward relationship, where predictable actions by a one party allow the relationship to operate more effectively. When trust is placed in someone, others are more willing to commit to relationships (Griffith and Harvey, 2004). In this sense, various researchers (Gulati, 1995b; Koka and Prescott, 2002) consider that the relational dimension is the critical factor in the creation and transference of knowledge and other resources, as it decreases the likelihood of developing opportunistic relationships and leads to more fluid sharing and combinations of resources (Liu, 2012; Liu and Chaminade, 2010; Tsai and Ghoshal, 1998). Lee (2009) pointed to the importance of trust as the most widely studied characteristic of the relational dimension and one that is linked to greater access and exchange of resources that are critical to obtain high performance. Moreover, a trustworthy focal actor is more likely to be a popular exchange partner and, therefore, it is more likely that the most valuable firms will choose to cooperate with it. In this same sense, there appears to be general support in the literature for the proposition that trust is important in the management of ISA relationships. Trust leads to a shared understanding between the alliance partners (Dhanaraj et al., 2004; Dyer and Nobeoka, 2000) and allows greater access to resources (Uzzi and Lancaster, 2003). Consequently, we also propose that a firm will be able to access really valuable partner resources, if trust has been previously developed in interorganizational relationships. This trust functions as a “social Dimensions of Social Capital and International Market Share 12 lubricant”, by facilitating access to partner resources (Rivera-Santos, 2009, p. 208) and, should therefore precede the resource dimension in the model. In a general analysis of the internationalization of all 14 firms, some authors have shown that the wider and the more cohesive the structural social capital of the individuals, the more trust and social interaction the relational social capital holds, and the more financial resources the firms will acquire (Lindstrand et al., 2011). In view of the abovementioned theory and empirical evidence, the structural dimension will be related to the international market share through, first of all, the relational dimension and, then, the resource dimension. Integrating these two models with mediation via the relational dimension and with the resource dimension generates a three-path mediation model (Figure 1) (Hayes, 2009; Taylor et al., 2008). To sum up, we test whether the relational and resource dimensions sequentially mediate the relationship between centrality and international market share. In other words, firm centrality impacts on the quality of the relations and this will influence the quality of the nodes through access to valuable resources, which will permit the focal actor to improve its competitive position at an international level. Thus: H4. The relational and resource dimensions sequentially mediate the relationship between the structural dimension and international market share. 3. Methodology 3.1. Sample Bearing in mind the proposed research objectives, a specific industry with a high globalization level has been selected. The public works sector has specifically been chosen for the following reasons. Firstly, because it is a global sector made up of large MNCs able to undertake projects in any part of the world. Secondly, this industry has recently experienced growing internationalization, mainly through alliances or cooperative actions (Bresnen and Marshall, 2000; Crespin-Mazet and Ghauri, 2007).Thirdly, our aim is to study the international market share at a global level and large contractors operate in a similar institutional context. This minimizes the bias caused by governmental influences and the strength of the consortium, both technically and financially, which is a critical issue in this industry (Ahadzi and Bowles, 2004). Finally, it makes sense to analyze social capital in a specific sector or activity where the number of actors is relatively easy to define or control (Ahuja, 2000a; Koka and Prescott, 2002). The International Civil Works industry is rather small, since a company needs to reach a Dimensions of Social Capital and International Market Share 19 quantifies the amount of variance that a construct captures from its indicators relative to the amount due to measurement error (Chin, 1998). AVE values should be greater than 0.50, which means that 50 per cent or more of the variance of the indicators should be accounted for. Consistent with this suggestion, AVE measures for all latent variables are greater than 0.8437 (Table 3). Discriminant validity indicates the extent to which a given construct differs from other constructs (Roldán and Sánchez-Franco, 2012). There are two approaches to assess discriminant validity in PLS. On the one hand, Fornell and Larcker (1981) suggest that the AVE should be greater than the variance between the construct and other constructs in the model (i.e., the squared correlation between two constructs). Thus, the square root of the AVE of each latent variable (LV) should be greater than its correlations with any other LV in the assessment. This condition is satisfied for all latent constructs in relation to their other variables (Table 3). On the other hand, the second approach suggests that no item should load more highly on another construct than it does on the construct it intends to measure (Hair et al., 2011). In addition, each construct should load higher with its assigned indicators than other items. This cross-loading analysis may be performed by calculating the correlations between the construct scores and the standardized data of the indicators (Gefen et al., 2011), a criterion which is also met in this study (Table 2). ------------------------------------------- Place Table 3 here Table 3. Construct reliability, convergent and discriminant validity coefficients ------------------------------------------- 4.2. Structural model The algebraic sign, magnitude and significance of the structural path coefficients, the R2 values, and the Q2 test for predictive relevance allow an evaluation of the structural model. Consistent with Hair et al. (2011), bootstrapping (5000 resamples) was used to generate standard errors and t-statistics. This enabled us to assess the statistical significance of the path coefficients. At the same time, the bootstrapping confidence interval of standardized regression coefficients was given. “If a confidence interval for an estimated path coefficient w does not include zero, the hypothesis that w equals zero is rejected” (Henseler et al., 2009, p. 306). Specifically, we applied a percentile approach, which has the advantage of being completely distribution free (Chin, 2010). Five of six direct effects described in Figure 2b are significant since these surpass the minimum level indicated by a Student's t-distribution with one tail and n-1 (n = number of Dimensions of Social Capital and International Market Share 20 resamples) degrees of freedom (Table 4). Likewise, this result is also achieved using the percentile bootstrap 95% confidence interval. The analysis of these outcomes shows that H1 is not supported. The direct effect of the structural dimension on the international market share (c’) is not significant and its confidence interval includes zero. In addition, our research model appears to have an appropriate predictive power for all the dependent variables (Table 4). Hence, the resource dimension attains the largest explained variance (0.648). We also evaluate the model with the cross-validated redundancy index (Q2) for endogenous reflective constructs (Chin, 2010). Chin (1998) suggests this measure to examine the predictive relevance of the theoretical/structural model. A Q2 greater than 0 implies that the model has predictive relevance, whereas a Q2 of less than 0 suggests that this is lacking in the model. The results summarized in Table 4 confirm that the structural model has satisfactory predictive relevance for the three dependent variables: the relational and resource dimensions and international market share. ------------------------------------------- Place Figure 2 here Fig. 2. Structural model: Three-path mediation model ------------------------------------------- ------------------------------------------- Place Table 4 here Table 4. Effects on endogenous variables ------------------------------------------- We have applied the analytical approach described by Preacher and Hayes (2008) and Taylor et al. (2008) to test our mediation hypotheses (H2-H4). The indirect effects are specified and contrasted with the mediators (i.e., RelD and ResD) (Table 5). We also examined the total (c) and direct (H1: c’) effects of the independent variable (i.e., the structural dimension) on the dependent variable (i.e., the international market share). Following Williams and MacKinnon’s (2008) suggestions, we chose the bootstrapping procedure to test the indirect effects. Chin (2010) proposed a two-step procedure for testing mediation in PLS: (1) Use the specific model in question including both the direct and the indirect paths and perform N bootstrap resampling and explicitly calculate the product of the direct paths that form the indirect path under assessment. (2) Estimate the significance using percentile bootstrap (Williams and MacKinnon, 2008). This generates a 95% confidence interval (CI) for mediators: RelD (H2), ResD (H3), RelD and ResD (H4). If the interval for a mediation hypothesis does not contain Dimensions of Social Capital and International Market Share 21 zero, it means that the indirect effect is significantly different from zero with 95% confidence. As Figure 2a and Table 5 show, the structural dimension has a significant total effect on the international market share. When mediators are introduced (Figure 2b), SD no longer has a significant direct effect on IMS (H1: c’). This means that the relational and resource dimensions fully mediate the influence of the structural dimension on the international market share (Baron and Kenny, 1986). Indeed, as previously mentioned, H1 is not supported. However, support is found for H2 to H4, which means that the three indirect effects of SD on IMS included in our research model are significant. Consequently, these analyses show that the relational dimension mediates the relationship between the structural dimension and the international market share (H2: a1b1). The results also show that the resource dimension mediates the path between SD and IMS (H3: a2b2), which is the most important indirect effect. Finally, we find that the structural dimension is positively associated with higher relational and resource dimensions, which relates to higher levels of international market share (H4: a1a3b2). ------------------------------------------- Place Table 5 here Table 5. Summary of mediating effect tests ------------------------------------------- 5. Discussion In this study, we aim to examine the social capital construct in greater depth through the application of a statistical mediation analysis. The structural dimension could be placed at the beginning of the process, whilst the relational and the resource dimensions occupy mediating positions between centrality (structural dimension) and international market share. Our results show that social capital dimensions are interrelated, in such a way that the relational and the resource dimensions fully mediate the effect of centrality on the international market share. Accordingly, the results have both theoretical and practical implications. 5.1. Theoretical implications This article is intended to respond to calls from different scholars for the need to broaden our theoretical understanding of social capital (Adler and Kwon, 2002; Koka and Prescott, 2002). Based on this work, we have distinguished three dimensions which occupy successive stages or positions of the social capital process, that is, the structural, relational and resource dimensions. When we consider the model with the total effect (Figure 2a), our results indicate that the more central the position of a firm in the global contractors network (i.e., the Dimensions of Social Capital and International Market Share 22 structural dimension), the greater the international market share achieved by the focal actor. However, the importance of the direct effect of the structural dimension decreases considerably in favor of the relational and resource dimensions when we analyze the full model (Figure 2b). On the one hand, the relational dimension - a consequence of strengthening a firm’s relationships and developing trust (Gulati, 1995b) - is more important than the mere establishment of an inter-organizational alliance. On the other hand, the resource dimension is the variable which has been more directly associated with firm performance (Batjargal, 2003; Tsai and Ghoshal, 1998). Accordingly, our research shows that firms that have privileged access to partners’ resources can achieve a greater international market share than their competitors (Autio et al., 2000). Therefore, when undergoing the internationalization process, firms must obtain, share, assimilate and combine resources – primarily knowledge– in order to compete and to grow in international markets. In fact, the resource dimension represents the best predictor of the international market share of the global contractors' network in our research model (it explains 18.77% of its explained variance), surpassing the direct influence of structural and relational dimensions on the final endogenous variable. Moreover, this study has helped to demystify the complex relationships between these different dimensions of social capital. Indeed, the structural model shows that all path coefficients between SD, RelD and ResD are significant. Consequently, such social capital dimensions are interrelated not only in a sequential process, but also in parallel (Tsai and Ghoshal, 1998); which is to say that the SD is significantly associated with RelD and ResD. More profoundly, we have shown that the relational and resource dimensions sequentially mediate the relationship between centrality and international market share. Our results further show that this is a full mediation. Accordingly, it demonstrates the relevance of strategic management of an alliance portfolio by the firm, in order to develop trusty relationships with valuable partners with the aim of exploiting the social capital and of achieving an international competitive advantage. Certainly, in the presence of the relational and resource dimensions, the direct impact of centrality on the international market share becomes insignificant. 5.2. Managerial implications The increasing importance of strategic alliances in the global business environment leads to the question of how MNCs should manage their alliance portfolios (Wassmer, 2010). On these lines, Mathews and Zander (2007) affirm that the competitive advantage of a firm is not due to Dimensions of Social Capital and International Market Share 23 owned assets that are exploited in international markets so much as it is due to the firm’s ability to manage and to access the assets of others through inter-firm networks. A good understanding of social capital is needed to respond to this question, which means analyzing its dimensions and the interrelationships between them. In our empirical context, the results show that a firm’s network position is important provided that trust is embedded in its network’s ties and its partners have valuable resources which can be accessed by the focal firm. Therefore, managers take account of the way in which a certain dimension influences the other dimensions, in order to configure and manage their alliance portfolios. Managers should therefore follow a holistic approach (Wassmer, 2010) in the management of their alliance portfolio, which implies considering all the relations of the focal actor and all the dimensions and aspects of the configuration of their capital. This implies that managers should take into account, when selecting the partners and developing the relational structure or pattern, which partners are better adjusted to their strategic objectives and values. It is therefore more likely that they will develop strong and trusty relationships with their partners, in order to avoid opportunistic behaviours and to promote the exchange of resources and they should, moreover, understand which resources are necessary and valuable for their firms and, therefore, which are the most suitable partners. In conclusion, managers have to maximize network benefits and minimize network risks, and to reach this objective the managers should understand and value the interrelations existing between the different dimensions of social capital. To sum up, by identifying the dimensions of social capital and showing that relational and resource dimensions play a critical mediating role, managers will be able to manage their alliances or inter-organizational relationships more strategically and hopefully obtain better performance from social networks. 6. Limitations The present study recognizes a series of limitations in its results and conclusions. First, an initial limitation is related to the notion of causality. Although evidence is provided of the model's causality, this has not really been tested. Our study has considered a soft modeling approach directed more towards prediction than causality. While causation guarantees the ability to control events, predictability allows only a limited degree of control (Falk and Miller, 1992). Second, another limitation is determined by the technique used for the proposed model: structural equations, which assume a linearity of relationships between latent variables Dimensions of Social Capital and International Market Share 24 (Hair et al., 1998). Third, a cross-sectional research design was employed. However, interorganizational networks are dynamic and change over time. Accordingly, future research should likewise measure the constructs analyzed at several points of time, taking into account the dynamics to configure the different dimensions of social capital. Fourth, the limitations of using sectoral characteristics, while having the advantage of allowing the network to be analyzed at a global level, show little cohesion. Testing the conclusions in other sectors where there is a stronger connection between members could be an important line of research which would allow us to advance in our understanding of the effect of social capital on a firm’s international performance. Finally, the advantages and disadvantages of social capital depend on the characteristics of the empirical context that is analyzed. References Acquaah, M. (2007). Managerial social capital, strategic orientation, and organizational performance in an emerging economy. Strategic Management Journal, 28(12), 1235. Adler, P. S., & Kwon, S. (2002). Social capital: Prospects for a new concept. Academy of Management Review, 27(1), 17. Agndal, H., Chetty, S., & Wilson, H. (2008). Social capital dynamics and foreign market entry. International Business Review, 17(6), 663-675. Ahadzi, M., & Bowles, G. (2004). Public-private partnerships and contract negotiations: An empirical study. Construction Management & Economics, 22(9), 967-978. Ahuja, G. (2000a). Collaboration networks, structural holes, and innovation: A longitudinal study. Administrative Science Quarterly, 45(3), 425. Ahuja, G. (2000b). The duality of collaboration: Inducements and opportunities in the formation of interfirm linkages. Strategic Management Journal, 21(3), 317. Ahuja, G., Soda, G., & Zaheer, A. (2012). Introduction to the special issue: The genesis and dynamics of organizational networks. Organization Science, 23(2), 434. Akintoye, A., & Main, J. (2007). Collaborative relationships in construction: The UK contractors' perception. Engineering Construction & Architectural Management (09699988), 14(6), 597-617. Alwin, D. F., & Hauser, R. M. (1975). The decomposition of effects in path analysis. American Sociological Review, 40(1), 37-47. Anand, B. N., & Khanna, T. (2000). Do firms learn to create value? the case of alliances. Strategic Management Journal, 21(3), 295. Arenius, P. (2002). Creation of firm-level social capital, its exploitation, and the process of early internationalization. (Unpublished Ph.D. thesis, Helsinki University of Technology, Espoo, Dimensions of Social Capital and International Market Share 25 Autio, E., Sapienza, H. J., & Almeida, J. G. (2000). Effects of age at entry , knowledge intensity, and imitability on international growth. Academy of Management Journal, 43(5), 909-924. Autio, E., Sapienza, H. J., & Arenius, P. (2005). International social capital, technology sharing, and foreign market learning in internationalizing entrepreneurial firms. In Dean A. Shepherd and Jerome A. Katz (Ed.), Advances in entrepreneurship, firm emergence and growth (pp. 9-42) No longer published by Elsevier. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173. Barroso, C., Cepeda, G., & Roldán, J. L. (2010). Applying maximum likelihood and PLS on different sample sizes: Studies on SERVQUAL model and employee behaviour model. In V. Esposito Vinzi, W. W. Chin, J. Henseler & H. Wang (Eds.), Handbook of partial least least squares: Concepts, methods and applications (pp. 427-447). Berlin: Springer-Verlag. Batjargal, B. (2003). Social capital and entrepreneurial performance in Russia: A longitudinal study. Organization Studies, 24(4), 535. Batt, P. J., & Purchase, S. (2004). Managing collaboration within networks and relationships. Industrial Marketing Management, 33(3), 169. Baum, J. A. C., & Ingram, P. (2002). Interorganizational learning and network organization: Toward a behavioral theory of the interfirm. In M. Augier, & J. G. March (Eds.), The economics of choice, change, and organization: Essays in memory of Richard M. Cyert (pp. 191-218). Cheltenham: Edward Elgar. Bell, G. G. (2005). Clusters, networks, and firm innovativeness. Strategic Management Journal, 26(3), 287. Borgatti, S. P., Everett, M., & Freeman, L. C. (2002). Ucinet 6 for windows: Software for social network analysis. analytic technologies: Natick. Blyler, M., & Coff, R. W. (2003). Dynamic capabilities, social capital, and rent appropriation: Ties that split pies. Strategic Management Journal, 24(7), 677-686. Bresnen, M., & Marshall, N. (2000). Partnering in construction: A critical review of issues, problems and dilemmas. Construction Management & Economics, 18(2), 229-237. Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press. Carmines, E. G., & Zeller, R. A. (1979). Reliability and validity assessment. (). Beverly Hills and London: Sage University Paper Series on Quantitative Applications in the Social Sciences. Carpenter, M. A., Li, M., & Jiang, H. (2012). Social network research in organizational contexts: A systematic review of methodological issues and choices. Journal of Management, 38(4), 1328-136. Chen, H. (1999). International performance of multinationals: A hybrid model. Journal of World Business, 34(2), 157. Dimensions of Social Capital and International Market Share 26 Chen, C., & Messner, J. I. (2009). Entry mode taxonomy for international construction markets. Journal of Management in Engineering, 25(1), 3-11. Chin, W. W. (1998). The partial least squares approach to structural equation modelling. In G. A. Marcoulides (Ed.), Modern methods for business research (pp. 295-336). Mahwah, NJ: Lawrence Erlbaum Associates. Chin, W. W. (2010). How to write up and report PLS analyses. In V. Esposito Vinzi, W. W. Chin, J. Henseler & H. Wang (Eds.), Handbook of partial least squares: Concepts, methods and applications (pp. 655-690). Berlin: Springer-Verlag. Coleman, J. S. (1988). Social capital in the creation of human capital. The American Journal of Sociology, 94, S95. Combs, C. F., Yasin, M. M., & Lisboa, J. V. (2007). The dimensionality and utilization of performance measures in a manufacturing operational context: Organizational change implications. Cross Cultural Management: An International Journal, 14(4), 286-306. Costa e Silva, S., Bradley, F., & Sousa, C. M. P. (2012). Empirical test of the trust–performance link in an international alliances context. International Business Review, 21(2), 293-306. Coviello, N., & Munro, H. (1997). Network relationships and the internationalisation process of small software firms. International Business Review, 6(4), 361. Cowan, R., & Jonard, N. (2006). Innovation networks and the distribution of knowledge. 1st International Conference on Economic Sciences with Heterogeneous Interacting Agents (Formerly WEHIA), Bologna, June 2006; DRUID Summer Conference 2006. Copenhagen. Cowan, R., & Jonard, N. (2009). Knowledge portfolios and the organization of innovation networks. Academy of Management Review, 34(2), 320-342. Crespin-Mazet, F., & Ghauri, P. (2007). Co-development as a marketing strategy in the construction industry. Industrial Marketing Management, 36(2), 158-172. Depperu, D., & Cerrato, D. (2005). Analyzing international competitiveness at the Firm Level: Concepts and measures. Working Paper, 32. Dhanaraj, C., Lyles, M. A., Steensma, H. K., & Tihanyi, L. (2004). Managing tacit and explicit knowledge transfer in IJVs: The role of relational embeddedness and the impact on performance. Journal of International Business Studies, 35(5), 428-442. Doz, Y. L. (1996). The evolution of cooperation in strategic alliances: Initial conditions or learning processes? Strategic Management Journal, 17(SPECIAL ISSUE), 55. Dussauge, P., Garrette, B., & Mitchell, W. (2004). Asymmetric performance: The market share impact of scale and link alliances in the global auto industry. Strategic Management Journal, 25(7), 701. Dyer, J. H., & Nobeoka, K. (2000). Creating and managing a high-performance knowledge-sharing network: The Toyota case. Strategic Management Journal, 21(3), 345. Eapen, A. (2012). Social structure and technology spillovers from foreign to domestic firms. Journal of International Business Studies, 43(3), 244-263. Dimensions of Social Capital and International Market Share 27 Eisenhardt, K. M., & Martin, J. A. (2000). Dynamic capabilities: What are they? Strategic Management Journal, 21(10-11), 1105-1121. Falk, R.F., Miller, N.B. (1992). A primer for soft modeling. Akron, OH.: University of Akron. Fombrun, C., & Shanley, M. (1990). What's in a name? reputation building and corporate strategy. Academy of Management Journal, 33(2), 233. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. JMR, Journal of Marketing Research (Pre-1986), 18(1), 39. Gefen, D., Rigdon, E. E., & Straub, D. (2011). An updated and extension to SEM guidelines for administrative and social science research. MIS Quarterly, 35, iii-xiv. Gnyawali, D. R., & Madhavan, R. (2001). Cooperative networks and competitive dynamics: A structural embeddedness perspective. Academy of Management Review, 26(3), 431. Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. The American Journal of Sociology, 91(3), 481-510. Granovetter, M. (1992). Problems of explanation in economic sociology. In N. Nohria, & R. G. Eccles (Eds.), Networks and organizations: Structure, form, and action (pp. 25-56). Boston: Harvard Business School Press. Griffith, D. A., & Harvey, M. G. (2004). The influence of individual and firm level social capital of marketing managers in a firm's global network. Journal of World Business, 39(3), 244-254. Gulati, R. (1995a). Social structure and alliance formation patterns: A longitudinal analysis. Administrative Science Quarterly, 40(4), 619. Gulati, R. (1995b). Does familiarity breed trust? the implications of repeated ties for contractual choice in alliances. Academy of Management Journal, 38(1), 85-112. Gulati, R. (1999). Network location and learning: The influence of network resources and firm capabilities on alliance formation. Strategic Management Journal, 20(5), 397. Gulati, R., & Gargiulo, M. (1999). Where do interorganizational networks come from? The American Journal of Sociology, 104(5), 1439. Gulati, R., Lavie, D., & Singh, H. (2009). The nature of partnering experience and the gains from alliances. Strategic Management Journal, 30(11), 1213-1233. Gulati, R., Nohria, N., & Zaheer, A. (2000). Strategic networks. Strategic Management Journal, 21(3), 203. Guler, I., & Guillén, M. F. (2010). Institutions and the internationalization of US venture capital firms. Journal of International Business Studies, 41(2), 185-205. Hair, J. F., Anderson, R. E., & Tatham, E. L. (1998). Multivariate data analysis. Englewood Cliffs, NJ: Prentice-Hall. Hair, J. F., Ringle, C., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice, 19(2), 139. Dimensions of Social Capital and International Market Share 28 Hair, J. F., Sarstedt, M., Ringle, C. M., & Mena, J. A. (2012). An assessment of the use of partial least squares structural equation modeling in marketing research. Journal of the Academy of Marketing Science, 40, 414-433. Hakansson, H., & Snehota, I. (1995). Developing relationships in business networks. London and New York: Routledge. Håkansson, P., Kjellberg, H., & Lundgren, A. (1993). Strategic alliances in global biotechnology — a network approach. International Business Review, 2(1), 65-82. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408-420. Hayes, A. F., Preacher, K. J., & Myers, T. A. (2011). Mediation and the estimation of indirect effects in political communication research. In E. P. Bucy, & R. L. Holbert (Eds.), (pp. 434-465). New York: Routledge. Helfat, C. E., & Winter, S. G. (2011). Untangling dynamic and operational capabilities: Strategy for the (N)ever-changing world. Strategic Management Journal, 32(11), 1243-1250. Hoffmann, W. H. (2007). Strategies for managing a portfolio of alliances. Strategic Management Journal, 28(8), 827. Hoskisson, R. E., Hitt, M. A., Ireland, R. D., & Harrison, J. S. (2008). Competitive for advantage. USA: Thomson South-Western. Houghton, S. M., Smith, A. D., & Hood, J. N. (2009). The influence of social capital on strategic choice: An examination of the effects of external and internal network relationships on strategic complexity. Journal of Business Research, 62(12), 1255-1261. Hsieh, M., & Tsai, K. (2007). Technological capability, social capital and the launch strategy for innovative products. Industrial Marketing Management, 36(4), 493-502. Huggins, R., & Johnston, A. (2010). Knowledge flow and inter-firm networks: The influence of network resources, spatial proximity and firm size. Entrepreneurship and Regional Development, 22(5), 457- 484. Hulland, J. (1999). Use of partial least squares (PLS) in strategic management research: A review of four recent studies. Strategic Management Journal, 20(2), 195. Inkpen, A. C., & Tsang, E. W. K. (2005). Social capital, networks, and knowledge transfer. Academy of Management Review, 30(1), 146. Jansen, R. J. G., Curşeu, P. L., Vermeulen, P. A. M., Geurts, J. L. A., & Gibcus, P. (2011). Social capital as a decision aid in strategic decision-making in service organizations. Management Decision, 49(5), 734-747. Jiang, R. J., Tao, Q. T., & Santoro, M. D. (2010). Alliance portfolio diversity and firm performance. Strategic Management Journal, 31(10), 1136-1144. Johanson, J., & Mattsson, L. (1988). Internationalization in industrial systems – a network approach. In N. Hood, & J. E. Vahlne (Eds.), Strategies in global competition (pp. 287-314). New York: Croom Helm. Dimensions of Social Capital and International Market Share 35 RM1T 0.3163 0.9385 0.4149 0.2468 RES_A 0.7539 0.4147 0.9860 0.5035 RES_AB 0.8162 0.5247 0.9875 0.4869 IMS1 0.3754 0.2991 0.4477 0.9304 IMS2 0.4557 0.3052 0.4904 0.9435 Table 3 Construct reliability, convergent and discriminant validity coefficients Composite reliability Cronbach’s alpha AVE (SD) (RelD) (ResD) (IMS) Structural D. (SD) 0.9416 0.9107 0.8437 0.9185 Relational D. (RelD) 0.9531 0.9044 0.9105 0.4697 0.9542 Resource D. (ResD) 0.9867 0.9730 0.9737 0.7965 0.4776 0.9868 (IMS) 0.9350 0.8613 0.8779 0.4455 0.3225 0.5016 0.937 SD: Structural dimension. RelD: Relational dimension. ResD: Resource dimension. IMS: International market share. Diagonal elements (bold) are the square root of the variance shared between the constructs and their measures (average variance extracted). Off-diagonal elements are the correlations among constructs. For discriminant validity, diagonal elements should be larger than off-diagonal elements. AVE: Average variance extracted. n.a.: non-applicable Table 4 Effects on endogenous variables Effects on endogenous variables Direct effect t-value (bootstrap) Percentile 95% confidence intervals Explained variance Relational dimension (R2 = 0.221 / Q2 = 0.186)  Structural D. (a1) 0.4697*** 10.7487 [0.3871;0.5568] Sig 22.06% Dimensions of Social Capital and International Market Share 36 Resource dimension (R2 = 0.648 / Q2 = 0.106)  Structural D. (a2) 0.7342*** 28.0197 [0.6835;0.7842] Sig 58.48%  Relational D. (a3) 0.1328** 2.9001 [0.0418;0.2207] Sig 6.34% Internat. market share (R2= 0.264 / Q2= 0.046)  H1: Structural D. (c') 0.1024ns 1.0934 [-0.0563;0.3123] Nsig 4.56%  Relational D. (b1) 0.0956* 2.0349 [0.0052; 0.1872] Sig 3.08%  Resource D. (b2) 0.3743*** 3.8891 [0.1742;0.5468] Sig 18.77% *** p< 0.001, ** p< 0.01, * p < 0.05, ns: not significant (based on t(4999), one-tailed test) t(0.05, 4999) = 1.645158499, t(0.01. 4999) = 2.327094067, t(0.001, 4999) = 3.091863446 Sig. denotes a significant direct effect at 0.05; Nsig. denotes a nonsignificant direct effect at 0.05 Table 5 Summary of mediating effect tests Total effect of SD on IMS (c) Direct effect of SD on IMS Indirect effects of SD on IMS Coefficient t value Coefficient t value Point estimate Percentile bootstrap 95% confidence interval Lower Upper Dimensions of Social Capital and International Market Share 37 0.4455*** 5.8078 H1 = c’ 0.1024ns 1.0934 Total 0.3431 0.1795 0.4884 H2 = a1b1 (via RelD) 0.0449 0.0023 0.0929 H3 = a2b2 (via ResD) 0.2748 0.1285 0.4095 H4 = a1a3b2 (via RelD + ResD) 0.0233 0.0049 0.0500 SD: Structural dimension. RelD: Relational dimension. ResD: Resource dimension. IMS: International market share. *** p< 0.001, ** p< 0.01, * p < 0.05, ns: not significant (based on t(4999), one-tailed test) t(0.05, 4999) = 1.645158499, t(0.01, 4999) = 2.327094067, t(0.001, 4999) = 3.091863446 Figure 1 Click here to download high resolution image Figure 2 Click here to download high resolution image Illustration 1 Click here to download high resolution image Illustration 2 Click here to download high resolution image