Resources,Governance,and Knowledge Transfer in Spanish Footwear Clusters

Abstract

How do internal resources, relational resources, and relational mechanisms determine knowledge transfer among crucial partners in clusters? We analyze a sample of Spanish footwear manufacturers to distinguish the relative impact of internal resources, intra-cluster relationships, and governance (including power asymmetries) on knowledge transfer between leading firms and their partners. Internal resources appear to be highly beneficial for knowledge transfer among partners, while intra-cluster relationships diminish such transmissions. The governance structure of the partnership also appears to have an important influence on knowledge transfer between leading organizations and their suppliers. These results suggest valuable implications for practitioners, researchers, and policy makers at both the firm and meso levels.

Keywords

internal resources local linkages knowledge transfer clusters governance footwear Spain

Introduction

It has long been assumed that the participation of geographically clustered firms in dense local networks facilitates innovation processes (e.g., Cooke 2005). However, although with different stadiums of comprehension, new research trends transcend the dominant paradigm of externalities and generalized knowledge spillovers thanks to geographical proximity, to an in-depth analysis of firm level characteristics, network particularities, and the limits of clustering effects (Giuliani and Bell 2005; Giuliani 2007; Boschma and ter Wal 2007; Camison 2004; Hervas-Oliver and Albors-Garrigos 2009).

Knowledge transfer embodies the multifaceted nature of boundaries, cultures, and processes that make the process far from easy (Kogut and Zander 1992; Easterby-Smith, Lyles, and Tsang. 2008). In this vein, prepublished research details multiple intra or interfirm level factors that shape the existence and effectiveness of this process (van Wijk, Jansen, and Lyles 2008; Jansen, van den Bosch, and Volberda 2005; Lane, Salk, and Lyles 2001). Such complexity makes our understanding of the phenomenon still incomplete. As Easterby-Smith, Lyles, and Tsang (2008) recently highlight, there is still room for examining interfirm knowledge transfer from a multidimensional and holistic approach. In fact, with a few exceptions, scholars have failed to simultaneously consider all unidimensional antecedents, particularly in clusters (Sreckovic and Windsperger 2011).

In this vein, our article contributes to filling this research gap by developing and testing hypotheses related to knowledge transfer mechanisms and their collateral effects in clusters. It examines the firm’s capacity to offer knowledge inputs while being able to absorb knowledge from other firms with a view to achieving solid advantages through product, process, or marketing innovations. It provides a novel perspective on the detrimental effects of local interfirm relationships and approaches key aspects previously ignored in the academic debate, such as governance and power asymmetries (e.g., Christopherson and Clark 2007). Results mainly endorse our theoretical premises using data from the main Spanish footwear clusters.

The theoretical argument on which our clustered firms are analyzed is based on the integration of complementary academic bodies of literature: the strategic decision-making approach (SDMA; Cowling and Sugden 1994, 1998), and the micro-level economics of innovation, grounded on the Resource Based View or RBV (Barney 1991; Peteraf 1993) and the Social Capital approach (Dyer and Singh 1998; Burt 1992). Additionally, networks and clusters overlap throughout the article, since most definitions share the notion of clusters as localized networks of specialized organizations, whose production processes are closely linked through the exchange of goods, services, and knowledge (Van den Berg, Braun, and Van Winden 2001). Consequently, both networks and clusters emerge as relational structures regulated by the interplay of the power of the actors involved (Sacchetti and Sugden 2003; Humphrey and Schmitz 2004; Parrilli and Sacchetti 2008).

This article is organized as follows. The second section illustrates the theoretical framework and the hypotheses to be tested. The third section provides information on the Spanish clusters. In fourth section, we present data and sample issues, variables, statistical analysis, and discuss the main findings. The concluding section summarizes the novelties of the article, its limitations, and suggests implications for practitioners, public policies, and future research.

Theoretical Framework

Industrial Clusters in the Modern Economy

Since the concept of the industrial cluster was formally proposed by Porter (1998), research from diverse academic fields has analyzed and enriched the term. This has generated a plethora of terminologies and a certain semantic ambiguity (see Gordon and McCann 2000) or literature on industrial districts versus clusters (Belussi 2006).

As proposed by Markusen (1996), the “Marshallian” cluster is a geographically delimited area where the business structure is comprised of small, locally owned firms that keep investment and production decisions within its boundaries. While the “Marshallian” cluster emphasizes spatial concentration and internal productive relationships, the term industrial district also involves interaction and social embeddedness. As defined by Becattini (1990) and developed by other studies (e.g., Brusco 1982; Sammarra and Biggiero 2001), this notion introduces the idea of a unique socioeconomic territorial system characterized by a certain level of identity, shared values, and cooperation in an atmosphere of mutual trust favored by local interactions (Paniccia 1998). Due to geographical proximity, common learning and knowledge flows between different actors become frequent phenomena. Hence, the space of places and the idea of networks as vehicles of knowledge transfer and diffusion overlap (Boschma and ter Wal 2007).¹

Industrial districts, which were traditionally made up of small and medium enterprises (SMEs), have experienced an increasing trend toward large corporations with a high reputation. They also show a marked export propensity, are market-oriented and active in delocalizing production, or are highly innovative (e.g., Belussi 2004). Increasing competition and the need to achieve a stronger global position has provoked the emergence of these district leaders whose remarkable internal resources and extra-cluster connections affect the systemic configuration. In particular, local networks become increasingly hierarchical and formalized, as these new leaderships embody more market concentration and power (Boschma and Lambooy 2002; Cainelli, Evangelista, and Savona 2006). In other words, some clusters turn into “Hub-and-Spoke” structures controlled by a few dominant firms, while suppliers and related activities spring up around them (Markusen 1996). Intra-district cooperation, mostly of a vertical nature, is driven by the willingness of the hub firms and is based on contracts and commitment. Guerrieri and Pietrobelli (2004) presented these “Hub-and-Spoke” clusters as optimal systemic structures.

Leading organizations affect the development and innovative dynamics of the cluster (Wolfe and Gertler 2004; Niosi and Zhegu 2005). Previous research suggests that (a) the formation of leading firms subsequently feeds the emergence and growth of many smaller ones (Wolfe and Gertler 2004), (b) the innovative activity of large organizations attracts smaller firms (Agrawal and Cockburn 2003; Feldman, Francis, and Bercovitz 2005), and (c) leaders influence the extent to which local actors engage in external linkages and access nonlocal knowledge (Morrison 2008). The implication of this dynamic perspective is immediate: the role of governance and power in the cluster trajectory becomes relevant (Parrilli and Sachetti 2008; De Propis, Menghinello, and Sugden. 2008).

Internal Resources, Interfirm Relationships and Knowledge Transfer

Firms achieve sustained competitive advantage by implementing strategies that exploit their internal strengths and avoid internal weaknesses (Barney 1999). The RBV focuses on the key success factors of an individual firm’s behavior to achieve specific advantages through a portfolio of differential assets (Mahoney and Pandian 1992; Peteraf 1993; Teece, Pisano, and Shuen 1997). The underlying idea of this perspective is that the market competitive position determines not only the firm’s success but also the existence of a set of resources capable of making its products unique. The seminal conceptualization by Barney (1991, 101) understands firm resources in a broad sense, including “all assets, capabilities, organizational processes, firm attributes, information, knowledge etc. controlled by a firm that enable it to conceive of and implement strategies that improve its efficiency and effectiveness.”

Consequently, it becomes apparent that a firm’s competitive advantage may derive from the profile and quality of its internal resources. Only valuable, rare, inimitable, and non-substitutable resources are strategic assets that, if properly mobilized, can generate advantages and improve performance (Barney 1991, 2007). In order to link these thoughts to this article’s perspective, we consider the most recent review of studies on RBV that assign increasing weight to knowledge and intangibles as the basis for developing competitive advantage (Newbert 2007; Locket, Thompson, and Morgenstern. 2009; Galbreath 2005). From a meso-level perspective, industrial clusters also have specific higher order capabilities (Foss 1996; Maskell and Malmberg 1999). Hence, local firms may benefit from a cluster’s influence because systemic capabilities additionally extend their internal resources. Each unit progressively configures its unique set of resources based on specific internal attributes and externally acquired resources (Hervas-Oliver and Albors-Garrigos 2009).

However, local firms benefit asymmetrically from this external knowledge depending on their own resources (Molina-Morales and Martinez-Fernandez 2004). Investment in internal resources allows the firm to develop the capacity to recognize, evaluate, and assimilate new knowledge, known as absorptive capacity (Cohen and Levinthal 1990). Therefore, it can be expected that the stronger the knowledge base of the company, the greater the likelihood of benefiting from external resources. In line with the study by Cohen and Levinthal (1990), such absorptive capacity, related to prior knowledge base, improves innovation performance, as external knowledge can be better assimilated and applied to commercial ends. Subsequent refinements of this notion by Zahra and George (2002) distinguish between potential (acquisition and assimilation) and realized absorptive capacity (transformation and exploitation).

Considering knowledge as a transferable resource or asset, interorganizational relationships create opportunities for its acquisition and exploitation (Dyer and Singh 1998; Capaldo 2007). If innovation derives from new exploitations of resources or new ways of exchanging and combining resources (Moran and Ghoshal 1996), then external knowledge favors innovation performance. Access to complementary knowledge repositories is determined by the characteristics of a firm’s relationships (Yli-Renko, Autio, and Sapienza 2001).

Dense networks generate social norms and sanctions that facilitate cooperation and trust, engendering positive effects (Coleman 1990). In addition, strong ties provide organizations with exchanges of high-quality information and create mechanisms of social control that govern the interdependencies in partnerships (Uzzi 1996). Previous research confirms the positive influence of density and strongly tied networks on innovation (e.g., Yli-Renko, Autio, and Sapienza 2001; Henderson 2007). Assuming that geographical proximity promotes cohesion through pervasive interactions, colocation facilitates fine grain knowledge transfer and the superiority of clustered firms in terms of innovation (for a recent analysis regarding the Spanish case, see Boix and Galletto 2009).

From a different perspective, the structural holes approach (Burt 1992) proposes an alternative perspective based on disperse structures, information diversity, and the advantages of being the broker in relationships between disconnected actors. In the same vein, Granovetter (1973) emphasizes how weak ties enable access to new and exclusive information. We may thus expect that distant relationships, characterized by sporadic interactions and reduced mutual commitment, are suitable for exploring new knowledge.

The innovative capacity of the territory relies on a combination of local buzz, for the intra-cluster capacity for innovation, and global pipelines, in order to be able to assimilate innovation produced elsewhere (Wolfe and Gertler 2004). Thanks to their position as intermediaries between external and local actors, cluster institutions such as universities or research centers may act as catalysts and hybridizers of new knowledge, becoming providers of absorbable external knowledge at moderate cost (Molina-Morales and Martinez-Fernandez 2004).

Interfirm Relationships, Governance, and Knowledge Transfer

Knowledge flows enable firms to build innovations upon the pool of knowledge that exist outside their boundaries. For example, knowledge transfer occurs consistently among actors along the value chain. Suppliers cooperate with their customers in order to provide inputs that enhance their product offer, while customers provide innovation opportunities or polish suppliers’ proposals (Von Hippel 2005). The corollary is that the development of new capabilities varies as a function of the nature of collaborative relationships between units (MacDuffie and Helper 2006).

In the modern economy, networks are made up of heterogeneous players and encompass multiple types of ties. Thus, each network has its particular structure and governance, that is to say, the coordination mechanisms by which activities and conflicts among various actors are coordinated and managed (Hollingsworth 1993). Such governance mechanisms are critical antecedents of knowledge processes (Foss 2006).

In innovation environments and clusters, governance relies mostly on a mix of implicit and self-coordination mechanisms, such as norms, trust, reputation, power, and leadership (Grandori 1997). As per earlier considerations, even egalitarian district networks have leaders that perform key planning and coordination functions (de Man 2004). This privileged position offers opportunities to exercise power, understood as the ability to shape and restrain the choices and actions of other organizations through formal and informal mechanisms in variable proportions, depending on the circumstances.

Drawing upon local systems, de Propris, Menghinello, and Sugden (2008) highlight the distribution of power as a crucial element of network governance. Power asymmetries frequently emerge and determine the nature of the relationship (Sacchetti and Sugden 2003; Bathelt and Taylor 2002). When power is evenly distributed among network members and decision centers proliferate, as in the “Marshallian” clusters mentioned earlier, heterarchical forms of governance appear (Amin and Cohendet 2005). Conversely, in hierarchical forms of governance, such as monopsonistic clusters (de Propris 2001), the dominant actors impose their views on other actors that are forced to make concessions.

Power asymmetries are not ignored by the Global Value Chain literature. Considering a firm's resources, product profile, and knowledge flows, four types of governance patterns are proposed by Humphrey and Schmitz (2004): arm’s length, network, quasi-hierarchy, and hierarchy. Power structure is balanced in both the arm’s-length and network patterns; while uneven distribution is observed in the other two cases. Cooperation, dependence, and knowledge transfer characterize networks; conversely, power asymmetries usually demote collaboration and information sharing.² Instead of networks and quasi-hierarchies, Gereffi, Humphrey, and Sturgeon (2005) identify three structures: captive, modular, and relational. In the captive form, the lead firm virtually controls the value chain, and partners are highly dependent for complementary activities (design, logistics, or purchases). Lower explicit coordination and power asymmetries characterize the modular governance structure, where suppliers manufacture according to customers’ specifications. Finally, the network form and relational type can be assimilated.

Implications are evident for both firms and territories as profits and resources for innovation gravitate to points of concentration of power. Considering the SDMA, mutual dependence contexts, such as heterarchical and networked conceptions, lead to multiple decision centers, win–win situations, and positive development of firms and systems (Cowling and Sugden 1998; Hess 2008; de Propris, Menghinello, and Sugden 2008). Conversely, zero sum game assumptions are frequently linked to concentration of power, such as monopsonistic, captive, and hierarchical situations, preceding strategic failures and development problems (Cowling and Sugden 1998; Hess 2008).

Narrowing down the focus exclusively to network governance may lead to undervaluing other scales of analysis such as cluster governance or individual and micro-mechanisms (Foss 2006). For example, valuable information is transmitted through geographically rooted networks with specific spatial structures. The particular socioinstitutional context of the territory molds its economy and the role of firm strategies in building norms and power asymmetries within networks (Christopherson and Clark 2007; Pike 2009). In fact, it is recognized how the advantages of the cluster paradigm frequently vanish in North America due to the well-known culture of competitiveness which promotes property rights, captive suppliers, or competition based on cost rather than cooperation and shared innovation (Clark 2010).

Hypothesis

Literature on industrial clusters has often portrayed them as islands of unity and homogeneity (Cainelli and Iacobucci 2007). However, this idea of homogeneity is far from being confirmed by the facts (Boschma and ter Wal 2007; Morrison 2008). This traditional perspective possibly overemphasizes the geographical dimension, overlooking the evidence that firms largely differ in terms of knowledge base and absorptive capacity (Giuliani 2007; Giuliani and Bell 2005). In this vein, Sammarra and Biggero (2008) stress the heterogeneity and specificity of knowledge flows in innovation networks. They point out that information transfer among firms engaged in cooperation activities depends upon the nature of the knowledge to be transferred, partner-specific characteristics, and the structural features of the social networks.

Internal investments in crucial business areas contribute to a firm’s resources and absorptive capacity, favoring the firms’ involvement in knowledge transfer and allowing them to increasingly benefit from information sharing (Cohen and Levinthal 1990). The implications are immediate, knowledge appears to be unevenly diffused and selectively shared among cluster units. Firms with weak knowledge bases are usually engaged in a reduced amount of knowledge transfer, as they only offer minor contributions of common interest (Giuliani 2007). Conversely, strong internal resources enhance knowledge intensive relationships (Saliola and Zanfei 2009). Research by Morrison, Rabellotti, and Zirulia (2008) shows how leading firms even prefer to exchange knowledge with skilful nonlocal partners, instead of exchanging knowledge with nonexpert firms within the cluster. In line with these arguments, we can hypothesize (as shown in Figure 1) that:

Figure 1.

Variables and hypotheses.

Hypothesis 1: Higher internal firm resources positively influence knowledge transfer between leaders and crucial partners.

Relational resources have been linked to learning and innovation phenomena (Zaheer and Bell 2005). Extensive literature shows that spatial proximity favors interactions and embeddedness (e.g., Watts, Wood, and Wardle 2006). Therefore, cooperation and information exchanges are more likely to occur in local networks, as their actors internalize the norms that discourage opportunistic behavior and emphasize trustworthiness (Granovetter 2005). Similarly, other arguments suggest that strong ties promote fine grain information flows and act as mechanisms of social control (Kogut 1996). The corollary is immediate; spatial proximity in clusters emerges as a crucial factor for knowledge transfer and innovation.

However, from the relational perspective, industrial clusters are not homogeneous. As Morrison and Rabellotti (2009) demonstrate, different subnetworks of actors coexist in these industrial systems. Each local unit can design its own network portfolio and, consequently, show differences in terms of behavior and outcome (Zaheer and Bell 2005). Such engagement in more than one network may produce negative collateral effects. For example, cooperation with cluster lead firms implies a high cost for small units. Considering that networking takes time and effort, small cluster firms may opt to focus mostly on the crucial partner to survive. By following this path, firms run the risk of being locked out of many of their traditional dense relationships. They tend to minimize investments in local networking because the overload of loyalty and commitment in the focal relationship acts as a powerful inhibitor.

Additionally, in advanced stages of the industry life cycle, the functional requirements of customers are satisfied and technological innovations are exceptional (Tripsas 2007). Consequently, symbolic and aesthetic features distinguish novel from older products, becoming essential for market value perceptions. In light of the fact that competitors can imitate product designs more easily than technological innovations, firms tend to be cautious about potential leaks of valuable information, and the integrity of recipient firms facilitates transfers of knowledge (Becerra, Lunnan, and Huemer 2008). Thus, when competition increases, knowledge transfer is drastically reduced if partners have close linkages with other local units due to the risk of information diffusion or opportunistic behavior. As Figure 1 reflects, we may hypothesize that:

Hypothesis 2: Vertical linkages negatively influence knowledge transfer between crucial partners.

According to Provan and Kenis (2008), network governance has an important impact on network effectiveness. Therefore, if governance reflects the way power is exercised, power imbalance partially explains the construction of interfirm routines and the sharing of knowledge (Mason and Leek 2008). Recent meta-analysis by Van Wijk, Jansen, and Lyles (2008) supports the idea that power issues have a pronounced effect when considering interorganizational knowledge transfer. In this vein, we can state that power asymmetries between organizations exist and that they determine the intensity of knowledge flows.

As already argued, lead companies may exercise power over network members through specific governance structures. Interfirm relationships are then based on direction and control, and there is room to exploit a prevailing position in the network. In fact, powerful firms occupying central positions in networks regulate interactions of other network members (Hanneman and Riddle 2005). Along these lines, Giuliani and Bell (2005) indicate that knowledge transfer among local firms is influenced by powerful actors; while Morrison (2008) highlighted how leading firms in clusters frequently act as gatekeepers of knowledge with their partners. However, when clusters denote forms of cooperation between equals, as in the case of traditional industrial districts, there is no single actor with enough power over other network participants. Thus, relationships tend to be based on mutual dependence, shared values, and participatory forms of governance (Parrilli and Scchetti 2008).

In hierarchical structures or captive networks, local firms with weak knowledge bases are significantly dependent upon leading firms for activities such as advanced production methods, design, or marketing. Leaders are in charge of introducing new products or processes, and only the less strategic competencies are shared with their cluster partners. Otherwise, leading firms giving up the core assets will automatically reduce their power and worsen their position in the network. In this context, knowledge transfer may be episodic or limited; hence a “low road” competition scenario emerges for local producers and the whole system (Kaplinsky, Morris, and Readman 2002).

By contrast, industrial districts reduce the dependency of small local units on a lead firm’s specific assets. Nonhierarchical networks enhance collective learning processes through continuous interactions and knowledge transfer. Small local firms may reinforce their core competence thanks to pervasive knowledge transfer, and therefore improve their position in the network structure. In consequence, as Figure 1 shows, we can hypothesize that:

Hypothesis 3a: More Hierarchical (power based) governance structures positively influence knowledge transfer from lead firms to their partners.

Hypothesis 3b: More Hierarchical (power based) governance structures negatively influence knowledge transfer from partners to lead firms.

The Study Setting

Territorial Agglomerations in the Spanish Footwear Industry

This article focuses on the analysis of the relationship between internal resources, interfirm linkages, governance, and knowledge transfer in clusters. To this end, we use information from the Spanish footwear industry; a globalized manufacturing sector, mostly characterized by incremental innovations (e.g., Belso-Martinez 2010).

Data from the Spanish Footwear Manufacturers Federation (FICE 2009)³ reveal that the industry has undergone a severe crisis since the beginning of this century. During the period 2001–2008, shoe production has declined on average by 6.2 percent. Such a drop in the number of shoes manufactured is mirrored in an average fall of 4.9 percent in employment per year. Despite this negative trend, the footwear and leather sector still comprises 1,832 footwear manufacturers with 29,053 workers, accounting for 1.2 percent of the industrial gross domestic product (GDP) and 2.3 percent of national employment. In the European context, Spain is the second largest producer and exporter after Italy.

As in other European countries, the Spanish footwear industry is characterized by the prevalence of small and medium-sized enterprises. Recent figures published by FICE (2009) show that 52.6 percent of the companies employed less than ten people, while only 3.3 percent had fifty or more employees. Most of these SMEs are geographically agglomerated, and usually have high levels of specialization at particular stages of the production process. These vertically disintegrated structures have traditionally favored collective efficiency and enhanced the internationalization and global competitiveness of local manufacturers (Ybarra 2006; Tortajada, Fernandez, and Ybarra 2005; Belso-Martinez 2006). Applying different methodologies and data, the most relevant of these territorial agglomerations has been recognized as industrial districts. Particularly, Boix and Galletto (2006) use the well-known Italian methodology in two stages designed by ISTAT. They first identify local labor markets using inter-municipality commuting data and an iterative algorithm of aggregation in five steps. Second, they apply four nested specialization coefficients on occupation data to determine manufacturing local systems of small and medium-sized enterprises with strong specialization in some manufacturing sectors. Previously, ISTAT methodology also allowed Giner and Santa Maria (2002) to identify the Vinalopó agglomeration as an industrial district.

The changing economic environment has severely transformed the whole sector and the major clusters. Competition from low-cost countries and the concentration in the fashion retail industry have promoted the emergence of multiple business models and the coexistence of a wide spectrum of management practices. As FICE (2009) reveals, through cooperation, entrepreneurs have redefined strategies and restructured their organizations, moving from the traditional company focused on manufacturing (with its own brand or only production-based) to more sophisticated models such as that of the market developer (with its own brand, control of design, quality and distribution, or simply as an importer and distributor) passing through multiple combinations according to product type, target market, and business strategy.

The leather and components suppliers appear as the second important group of actors in the industry. In most cases, these firms originated as a result of spin-off processes from the footwear manufacturers themselves. Prior personal relationships and specific knowledge were the main competitive arguments of these new companies (Ybarra 2006). This part of the sector is made up of 600 companies, basically SMEs that currently employ over 11,000 people (Asociación Española de Componentes del Calzado [AEC] 2008).⁴ Over 40 percent of the Spanish components and machinery are sold abroad, and the rate of exports against imports reaches 137 percent.

The speed to market and the constant renewal of products demanded by fashion retailers and shoe manufacturers have led to major changes over the last few decades. After an intense restructuring period, a new productive model based on quick service and innovation has been implemented by this segment of the footwear industry. Technologies, designs, and materials offered by local suppliers have evolved, facilitating incremental innovation through incorporation of top quality, fashionable materials (AEC 2008). In fact, 75 percent of the Spanish components and machinery exports are sold in Europe, where top quality shoes are made.

Local institutions represent the third relevant local actor to be taken into account in our introduction to the industry. Indeed, universities, local/regional development agencies, research centers, or associations have demonstrated a noteworthy ability to consolidate the structure and enhance the economic activity of these industrial organization systems. Initially, these bodies provide highly specialized services and contribute to bringing about an atmosphere of trust that is crucial to interfirm cooperation and knowledge transfer (Tomas, Contreras, and del Saz 2000), and subsequently enable the access of local actors to nonredundant knowledge, thanks to their role as meta-organizers or intermediate agents (Molina-Morales 2008).

Data and Sample Issues

The Questionnaire

Data for this research were collected in the four leading Spanish footwear clusters: Vinalopó, Arnedo-Calahorra, Almansa-Albacete, and Balearic Islands. In a preliminary stage, a combination of semistructured questionnaires and face-to-face interviews were selectively carried out to gather primary data on multiple aspects of the footwear industry. A representative sample of fifteen local footwear manufacturers, researchers, and institutions was selected on the basis of reputation, involvement, and geographical location. Comments obtained were later used for the construction of the surveyed questionnaire, discussion, and conclusions.

Having completed the qualitative analysis, a four-page closed questionnaire was designed, based on inputs from the exploratory analysis and our literature review. After this design phase, a pilot questionnaire sent to thirty-five firms provided us with the opportunity to modify some categories and questions with the intention of obtaining better, unbiased responses.

Data and Sample Issues

The target population for the sampling was taken from the Central Company Directory (DIRCE) and included all footwear manufacturers in the four leading industrial clusters.⁵ As this public statistical source does not give detailed information at the firm level, the Dun&Bradstreet database was used to randomly select companies with more than one employee. In order to achieve a critical mass of firms per cluster and to take into account firms of different dimensions, we stratified the sample by size and geographical area. The questionnaire was submitted to business owners or top managers in the survey frame during the period January to February 2006. After controlling for respondent profile, 401 valid responses were obtained, permitting a significance level of 95.5 percent with an error margin of 5 percent in the worst case scenario (p = q = 50).

Almost 63 percent of the establishments in the sample were located in the Vinalopó district, with available information from DIRCE showing a similar percentage in that area. The other clusters were consciously overweighted to analyze potential differences among the four industrial systems. Each of the smaller agglomerations accounted for slightly over 12 percent. The data set reveals that 53.3 percent have one to five employees; while only 11.7 percent of the firms employ twenty-five or more workers. Thus, the structure of the industry and sample of this research mostly contains the smallest sized categories of firms. We controlled for potential bias derived from the stratification that may interfere with the accuracy of the results. No particular differences between the compared groups emerged.

Due to the research objectives, respondents were explicitly questioned as to whether any productive activity, traditionally conducted in-house, had been contracted to one or various external suppliers over the last five years. Using this conceptualization, only information provided by the 132 companies engaged in subcontracting operations during the last five years was used in the present study. Consistent with this conceptualization, 57.6 percent of the manufacturers interviewed could be considered as leading firms, while the remaining units were ascribed to the suppliers group.⁶ The average number of business areas in which subcontracting operations existed was 2 from a maximum of 4.⁷ Most of the activities subcontracted were not high-value-adding, 93.4 percent of the units externalized manufacturing processes, compared to 29.7 percent of the firms that externalized knowledge-intensive operations. Generally speaking, the geographical location of this sample also reflected the structure of the industry.

Variables

Knowledge transfer

The different hypotheses test the impact of internal resources, interfirm local linkages, and governance on knowledge transfer. Our dependent variable, named Knowledge transfer, was built using information on cooperation activities. Specifically, firms were questioned as to whether or not assistance between crucial partners existed in the following areas: (a) product design and development, (b) organizational and financial management, (c) manufacturing processes, (d) logistics and marketing, and (e) education and training.⁸ On the basis of the answers obtained, a five-level cumulative variable was constructed for the number of business areas in which the firms cooperate. This specification of the dependent variable assumes that the more areas the firm cooperates in (received assistance), the greater the amount of knowledge transfer.

The following step was to build four different factors to represent our main explanatory variables.

Internal resources: In order to gather data about a firm’s internal innovation activities, our survey requested information about product design, development innovation intensity, and marketing innovation intensity. Taking into account the study by Marsili and Salter (2006), the observed variables were operationalized as follows: % design and product development expenditures over total sales during the last three years and % marketing expenditure on total sales over the last three years. In order to achieve a more accurate picture, these objective indicators were combined with the respondent’s perception of the innovative nature of the firm’s product design and marketing activities. Using factor analysis, data obtained through the mentioned variables was condensed into a single composite factor (named Internal Resources), with a resulting Eigenvalue of 1.26 and an explained total variance of 62.75 percent (Kaiser-Meyer-Olkin [KMO] > .50; p value > .01).

Local interfirm linkages: Considering operationalization as proposed by Pla-Barber (2001), we investigated the strategic relevance, intensity, and stability of the relationships between local units (vertical and horizontal patterns). To determine the characteristics of these linkages, business owners and top managers were asked to rate: (a) the strategic relevance of the different linkages with clients and suppliers; (b) the stability and intensity of the resources shared in each type of relationship. We applied a 5-point Likert-type scale where 1 was very low and 5 very high. The internal validity and consistency of the construct (named Vertical relationships) was verified. The results obtained showed a Cronbach’s α of over .85, validating the aggregation of the four items in one factor with an Eigenvalue of 2.737 and 68.42 percent of the total variance explained (KMO > .50; p value > .01). Factor loadings ranged from .786 to .877.

Governance: To operationalize governance, we included a number of questions based on our literature review. As our purpose was to evaluate the extent to which power is asymmetrically distributed in interfirm relationships, we asked business owners and top managers to rate the formalization of interfirm relationships and (a) formalization of interfirm relationships; (b) flexibility in quality control and quality requirements; (c) flexibility in terms of delivery; (d) flexibility in the selection of suppliers and materials; and (e) amount and regularity of the orders. We used a 5-point Likert-type scale ranging from 1 (very low) to 5 (very high). As we would expect a certain amount of perceptional bias depending on the firm’s position in the network (central vs. noncentral), two different variables were created for each type of company, named Governance (Central) and Governance (noncentral). The internal validity and consistency of the constructs were verified. The results gave a Cronbach’s α of .70 and .82, respectively, validating the aggregation of items into two factors. The two factors had Eigenvalues of 1.908 and 2.355, respectively, with over 50 percent of the variance explained (KMO > .50; p value > .01). Factor loadings ranged from .527 to .826.

Control variables: Given the cross-sectional nature of our research, the final step at this stage was to establish the control variables to reduce concerns about potential endogeneity. We opted for Size, measured as the average number of employees during the last three years. Size can affect a unit’s innovation capacity, as large units have more resources and advantages in gaining support for innovation activities. Export intensity over the last three years is another variable that can affect innovation because international operations allow access to nonlocal knowledge, enhancing a firm’s innovation activity (Pla-Barber and Alegre 2007).

Local institutions links was the last control variable inserted into the models. In order to facilitate the responses, we used a similar configuration to that of the previously described local interfirm linkages. Having considered the construct applied by Molina-Morales and Martinez-Fernandez (2004), we asked respondents: (a) if they considered the role played by local institutions and sector associations to be strategically important; and (b) whether their firm maintained stable, intense, and beneficial relationships with local institutions in key business areas. Once more, a 5-point Likert-type scale was applied ranging from 1 (completely disagree) to 5 (completely agree). We attempted to improve the quality of our index by combining the answers obtained with an objective indicator such as the number of local institutions to which the firm belongs. Internal validity and consistency of the construct (named Institutional linkages) was verified. The results achieved a Cronbach’s α of over .72, validating the aggregation of items into one factor, with an Eigenvalue of 1.923 and 64.09 percent of total variance explained (KMO > .50; p value > .01). Factor loadings ranged from .450 to .956.

To verify the robustness of the variables, confirmatory analysis was conducted using qualitative techniques. Both peer debriefing (confirming analysis with a small group of academic experts and policy makers) and member checks (confirming analysis with the study's participants) corroborated the validity of the constructs. Table 1 summarizes the different variables; while Figure 1 presents the hypotheses established with respect to our dependent variable.

Table 1.

The Variables Included.

Variables	Description	Scale considerations
Internal resources	This construct was built using information obtained from three questions about: (a) design and product expenses normalized by total sales (average value of the last three years); (b) marketing expenditures normalized by total sales (average value of the last three years); (c) business owner and manager perception about firm’s innovative character of product design and marketing activities based on a 5-point Likert-type scale	Metric variable referring to the last three years
Size	Number of employees on average per year	Metric variable referring to the average of the last three years
Export intensity	Exports normalized by total sales on average per year	Metric variable referring to the average of the last three years
Institutional linkages	This construct was built mixing objective and subjective information. Two questions based on a 5-point Likert-type scale: (a) the strategic relevance of linkages with local institutions; (b) the stability and the benefits obtained from linkages with local institutions. Number of local institutions (technological research centers, business associations, professional associations), the firms belongs to	Metric variable referring the average of the last three years
Vertical linkages	This construct was built using information obtained from four questions based on a 5-point Likert-type scale: On one hand, the strategic relevance of linkages with local suppliers and customers; On the other hand, the stability and the benefits obtained from linkages with local suppliers and customers	Metric variable referring to the last three years
Governance-C	This construct was built using information obtained from five questions. Using a 5 point Likert scale, respondents were asked to rate from the central firm perspective: a) formalization of inter-firm relationship, b) flexibility on quality control and quality requirements; c) flexibility in delivery terms; d) flexibility in the selection of suppliers and raw materials; e) amount and regularity of the orders	Metric variable referring to the last three years
Governance-NC	This construct was built using information obtained from five questions. Using a 5 point Likert scale, respondents were asked to rate from the noncentral firm perspective: a) formalization of inter-firm relationship, b) flexibility on quality control and quality requirements; c) flexibility in delivery terms; d) flexibility in the selection of suppliers and raw materials; e) amount and regularity of the orders	Metric variable referring to the last three years
Knowledge transfer	This five level variable was created by accumulating the business areas in which assistance between partners existed. Respondents were asked to selected from the following list: (a) product design and development; (b) organizational and financial management; (c) manufacturing processes; (d) logistics and marketing; (e) education and management	Metric variable referring to the last three years

Table 2.

Correlation Matrix.

Knowledge transfers	1,000
Size	*.182	1,000
Export intensity	***.316	***.298	1,000
Internal resources	**.211	***.248	**.185	1,000
Vertical linkages	***–.238	.003	.107	**.203	1,000
Institutional linkages	**.218	*.174	**.232	***.278	***.362	1,000
Governance-C	**–.198	–.064	–.026	–.129	.122	–.046	1,000
Governance-NC	.124	–.010	–.003	–.049	**.184	.109	.077	1,000
M	2.20	18.73	19.99	4.07	3.86	3.84	3.92
SD	1.30	25.41	31.45	.85	2.03	.90	.61
N	132	132	132	132	132	132	132	132

Significance level: ***0.01; **0.05; *0.1.

Statistical Inference

As in previous empirical research (e.g., Tokatli 2007), the role of lead firms and supplier organizations was analyzed with regard to subcontracting operations. We assumed that lead firms should be large buyers in the selected clusters, with a strong market position, good reputation within the footwear industry and proximity to final customers. The Mann-Whitney U test confirmed the differences expected in size and export intensity (p value < .01) between the two groups.

A regression model was applied to evaluate the contribution of each independent variable (representing a firm’s internal resources, vertical relationships, governance, institutions, size, and export intensity) to knowledge transfer. Preliminary diagnostics were performed to verify the assumptions and to assess the accuracy of our analysis. The assumptions were confirmed mainly with the help of residuals of fitted models. Additionally, the correlation matrix did not show correlations >.7, see Table 1, and Cronbach’s α coefficients proved that the reliability assumption for the models was also satisfied. The equation model was expressed as follows:

K n o w l e d g e T r a n s f e r s_{i} = constant + β_{1} S i z e_{i} + β_{2} E x p o r t i n t e n s i t y_{I} + β_{3} I n t e r n a l R e s o u r c e s_{i} + β_{4} V e r t i c a l l i n k a g e s_{i} + β_{5} I n s t i t u t i o n a l l i n k a g e s_{i} + β_{6} I n t e r a c t i o n_{i} + β_{7} G o v e r n a n c e - C_{i} + β_{8} G o v e r n a n c e - N C_{i} + ε_{i} g_{i} .

Table 3 presents the regression models obtained using the stepwise procedure to avoid multicollinearity problems and spurious variables. Our baseline model tests the impact of the control variables size, export intensity, and institutional linkages (β₁, β₂, and β₅) on knowledge transfer. The second model reflects the influence of internal resources (β₃), vertical linkages (β₄), and the control variables (β₁, β₂ and β₅) on the dependent variable. The final model also controls for the role of governance (β₇ and β₈) on knowledge transfer. Along the lines of previous literature (e.g., Hervas-Oliver and Albors-Garrigos 2009), multiplicative variables were also used to test the interaction effects (β₆). The F-tests conducted show that all linear regressions are significant at the 1 percent level. In addition, it should be mentioned that adjusted R ² values are moderate (.110, .239, and .255, respectively), although most of the coefficients are significant. The explanatory power of the model rises by 8.1 percent when the intermediate set of variables is augmented with the governance specific set. It thus appears that the explanatory power is partly explained by the governance variables.

Table 3.

Regression Analysis.

Model	Variables	Adjusted R ²	F	β (sig.)
1	Constant	.110	7.792***	1.939
	Export intensity (β₂)			.011***
	Institutional linkages (β₆)			.199*
2	Constant	.239	8.375***	1.973***
	Export intensity (β₂)			.011***
	Internal resources (β₃)			.300**
	Vertical linkages (β₄)			–.572**
	Institutional linkages (β₆)			.248*
3	Constant	.255	7.438***	1.954***
	Export intensity (β₂)			.012***
	Internal resources (β₃)			.268**
	Vertical linkages (β₄)			–.512**
	Institutional linkages (β₅)			.237*
	Governance-C (β₇)			–.260*

Note: The variables not shown in models 1 (Size), 2 (Size, Interactions), and 3 (Size, Governance-NC, Interactions) were excluded because their lack of contribution to the model in the adjusted R ² through the stepwise procedure in the multiple regression analysis.

Significance level: ***0.01.; **0.05; *0.1.

Models 2 and 3 predict that internal resources are positively associated with knowledge transfer between crucial partners (p < .05). Such an outcome provides strong support for the first assumption (Hypothesis 1). Furthermore, our second proposition suggests that intra-cluster relationships (Vertical relationships) are negatively associated with knowledge transfer. Both models confirm Hypothesis 2 at p value < .01.

The governance effect was subsequently included in model 3, and for this specific purpose two variables were added to the initial configuration. The model was run with all the internal resources, vertical relationships, control variables, interaction terms, and the two governance variables. The statistically significant negative influence of Governance-C (perception from customers) at p < .1 showed how rigid and power-driven governance structures promote knowledge transfer between crucial partners. Therefore, Hypothesis 3a is confirmed. Conversely, the positive coefficient achieved by Governance-NC (perception from suppliers) is not statistically significant (Hypothesis 3b). Therefore, we only find partial support for Hypothesis 3.

The institutions variable (Institutions relationships) is also statistically significant at p < .05 (model 2) and p < .1 (models 1 and 3), indicating a positive effect on knowledge transfer between partners. Furthermore, the Size variable was not statistically relevant, while Export intensity was highly significant in all models (p < .01). This final result provides some insights into the positive effect of a firm’s access to nonlocal sources of knowledge on valuable information transfer and innovation.

Discussion

Internal firm resources and knowledge transfer. It is apparent from the confirmed Hypothesis 1 that when partners have strong knowledge bases, interfirm relationships are more likely to be characterized by high exchanges of design, marketing, or technology. As knowledge transfers can take place in both directions, this result endorses the premise that actors involved in knowledge transfer must have something worthwhile to offer, but also the internal resources to recognize the potential value of shared knowledge. For instance, relevant information about fashion trends, product forms, designs, or colors can flow from customer to supplier (and vice versa). Consequently, actors need the capacity to absorb this key information, and apply it in successful product lines. Several experts interviewed in the preliminary stage of our research, indicated that past experiences or human resources contribute to the development of this absorptive capacity.

This outcome is also in line with Saliola and Zanfei (2009) who suggest that the capacity “to handle the technology” is linked to the transfer of value-added activities. Consequently, leading footwear firms characterized by sophisticated products, advanced organizational systems, or important marketing activities are more open to sharing knowledge with their crucial partners in order to maintain their privileged position. Taking into account our qualitative evidence, we can state that such transfer allows leading firms to partially or entirely shift sophisticated activities to skilful suppliers, permitting significant cost reductions and reorientation of resources. Usually, expert partners assume the desirability of transferring knowledge in order to maintain their privileged relationship with the lead firm.

Intra-cluster relationships and knowledge transfer. The results obtained are in line with the new research trend that rejects the uniform positive impact of interfirm relationships on many aspects of firm performance. The acceptance of Hypothesis 2 leads us to suggest that dense relationships in clusters may not be systematically beneficial for firms. It seems that the particularities of the firm’s relationship portfolio can shape the actor’s knowledge transfer in different ways.

Generally speaking, local firms have ties to another organization that can lend resources and valuable information that are essential to innovation. However, when footwear units enjoy a preferential business relationship, tight linkages with other local suppliers and customers may harm knowledge flows between crucial partners. Expert interviews provide us with some insights as to why both leaders and expert suppliers may become reluctant to transfer knowledge if their partners are excessively embedded in intra-cluster relationships: (a) lead firms may be concerned about the dissemination of valuable knowledge (management practices, new product launches, design developments, or fashion trends) through the pervasive local interactions of their skilful suppliers; (b) leaders’ contacts with other local actors (particularly suppliers) may generate uncertainty and even destabilize their main business relationship; (c) cultivating many local relationships demands time and effort that can detract from the main partnership, diminishing attention devoted to the main partner.

Governance and knowledge transfers. Two different scenarios arise when discussing results on Hypothesis 3: the perspective of the lead firms (Hypothesis 3a) and the skilful suppliers (Hypothesis 3b). Consistent with our hypothesis, it appears that if leaders feel themselves to be involved in a relatively flexible relationship with their crucial suppliers, intensive knowledge transfers are less likely to occur. The negative impact of Governance-C can be interpreted as a signal that the capacity of leaders to control supplier behavior emerges as a relevant issue. Knowledge transfer from lead firms requires governance modes through which leaders feel secure from opportunistic behavior on the part of their partners. Lead firms will only be induced to share knowledge related to core business activities when suppliers’ upgrading processes cannot meet their competitive position. Therefore, to this extent, knowledge transfer requires hierarchical forms of governance and the necessary power asymmetries.

Contrary to expectations, Governance-NC shows positive insignificant influence on knowledge transfer. The evidence obtained probably suggests that even relationships lead to higher levels of knowledge transfer from the suppliers’ perspective. Although particular caution is needed when discussing this outcome, we may speculate and say that it is related to the relative weakness of these traditional intra-cluster networks. Mutual dependence and reciprocity are at the base of dense local relationships. However, the more control lead firms exert over their suppliers, the less interdependent partners are likely to feel. Consequently, SMEs become progressively reluctant to transfer knowledge; although they cannot ignore their fragile status and are forced to assume the risk and benefits of sharing valuable information.

Local institutions and knowledge transfer. We also explore the role of local institutions on knowledge transfer among lead firms and partners. In line with Parra-Requena, Molina-Morales, and Garcia-Villaverde. (2010), the results endorse the relevance of these actors in knowledge flow dynamics in Spanish footwear clusters. The agglomeration of firms boosts a range of institutions that enable technological upgrades, providing advanced services, or enhance innovation activities, acting as gatekeepers of knowledge. However, in light of both the theoretical literature and the results of this article, perhaps these perspectives are insufficient to explain differences in knowledge transfer, and other institutional roles may provide relevant clues.

Considering the study by Coleman (1990), social capital refers to elements such as trust, common norms, habits, or rules that bind the different local actors, facilitating knowledge circulation at the cluster level. In advanced stages of cluster evolution, the resultant dense collaborative networks emerge as a prerequisite for successful exchanges and collective actions (Maskell and Malmberg 2007). Meetings and joint projects mediated through cluster institutions bring multiple local actors together, contributing to the reinforcement of shared values and the atmosphere of trust necessary for interfirm knowledge transfer. Consequently, the positive coefficient of this variable reveals the extent to which a cluster’s institutions enhance knowledge transfer and innovativeness.

Conclusions, Limitations, and Policy Implications

Questions concerning the negative effects and constraining nature of certain network ties, the growing weight of lead firms in clusters and the implications in terms of governance for knowledge transfer clearly emerge in analyses of cluster transformation and innovation. In light of the findings from this study, several important implications have appeared related to the aforementioned questions. We suggest that knowledge transfer between crucial partners is a result of the interaction between the firms’ internal resources, intra-cluster linkages, and local institutions. Some purely intra-cluster factors help explain knowledge transfer dynamics. In this vein, supporting institutions still provide the glue for local interfirm relationships and enhance knowledge transfer between partners. Thus, our findings legitimize crucial partners as targets for public actions. Public policies should not only include the provision of advanced technical services or the local–global mediating role but also contemplate programs that improve knowledge transfer between crucial partners. Interventions need to go beyond merely promoting good matches between organizations, and have to be underpinned by principles, such as consolidating the proper atmosphere, considering governance, or evaluating a partner’s capabilities.

The analysis supports the claim that clearly warns against assigning purely beneficial effects to dense intra-cluster linkages. In fact, negative collateral consequences are likely to emerge when norms for appropriate business behavior become weak and doubts about partners’ trustworthiness appear. This may cause uncertainty in terms of loyalty, and a good deal of frustration. Consequently, managers should be aware that if the cooperative atmosphere is not stable, there might be a need for strengthening norms, values, and communication channels essential for a fluent and effective interaction between partners. Additionally, skilful suppliers may be locked out of their traditional local networks by their lead firm because of demands for concessions and the underlying business dependency. In light of this fact, the implication is clear: once the supplying organization is nearly isolated in the cluster, its innovation performance progressively declines (Macduffie and Helper 2006).

In the footwear industry, no firms can perform the activities of all the parts of the value chain. This evidence raises the relevance of high-quality exchanges of knowledge for firm survival and growth. Traditionally, poor absorptive capacity of the recipient and unwillingness of the donor to share profitable knowledge have been recognized as powerful inhibiting factors of transfer. The simultaneous need for technological strengths and absorptive capacity in both donor and recipient is another insight provided by our research. Intensive knowledge transfer is more likely to occur when both partners possess the necessary business experience and technological capabilities to make them happen and make both sides “win.”

In mature clusters, many suppliers and lead firms can act as donors and recipients of valuable flows. Skilful suppliers can acquire higher managerial, organizational, market, or product development knowledge in aspects such as fashion trends, mass global markets, off-shoring activities, or logistics. However, the poor absorptive capacity of the supplying organization can impede “successful” knowledge transfer. On the other hand, due to their product design and technological capabilities, skilful suppliers may also transfer valuable information to lead firms. Release of knowledge from lead firms or multinationals to suppliers has been widely documented in the academic literature. However, it must be highlighted that, in industrialized clusters, lead firms with the proper absorptive capacity can also benefit from the knowledge transferred by their skilful suppliers.

In the later stages of the cluster life cycle, previous internal investments and interactions at both local and nonlocal level consolidate a supplier’s knowledge base, permitting information flows to the lead company and reinforcing its knowledge base. The key issue is that both lead firms’ and supplier organizations’ internal resources are key factors in explaining interfirm knowledge transfer and, subsequently, the innovation capacity of the collective action. Both leader and supplier internal resources not only allow the emission of information flows but also determine how the partner’s resources are accessed, exploited, and combined. Additionally, complementarities between the partners’ resources encourage information flows and reinforce the synergies achieved by both actors.

Considering that cluster literature has largely ignored the notion of power, our findings highlight some valuable contributions related to the governance structure and power asymmetries between partners. The more hierarchical and asymmetric the relationship is, the greater the knowledge transfer from the lead firm will be. As concerns about a supplier’s business behavior arise, the lead firm uses its dominant position to ensure that the knowledge transferred is not implemented or diffused in an opportunistic manner. In other words, the emergence of lead firms propels less balanced relationships; due to the fact that leaders expect to capture all the benefits derived from the information flows and their applications. Conversely, the supplier’s perspective does not mirror the process, as more asymmetrical relationships probably moderate the cooperation intensity. The corollary is immediate; traditional “Marshallian” configurations may deteriorate as more hierarchical governance structures materialize. To some extent, this evidence probably underlies the progressive emergence of Hub-and-Scope structures in well-known “Marshallian” clusters. Future research needs to further examine and corroborate this line of reasoning.

This research breaks new ground in addressing complicated phenomena. As such, the study is subject to several limitations. Measurement of the constructs is limited by single-source data, which relies on footwear manufacturers’ reports. Even though we validated the measurements using different methods, additional data collected from a variety of sources are warranted to further examine this issue. In particular, research should overcome the weaknesses of our dependent stacked variable which assumes that knowledge transfers are equally important in the different areas analyzed. Indicators on the frequency, the reasons, the profile, and the channels through which knowledge is transferred would provide refined tools capable of improving the picture obtained. In addition, we are conscious of the particularities and the need to extend the sample used from the sector dimension in order to increase the generalizability of our findings. In addition, the dynamic nature of the phenomenon analyzed invites future research into how knowledge transfers evolve over time. It is possible that success in past knowledge sharing experiences will exercise considerable influence. Obtaining longitudinal data across multiple transfer experiences would help shed light on the dynamics and corroborate the causal ordering among constructs, thus representing a fruitful research direction. Nevertheless, in our opinion, these limitations do not devalue the robustness and relevance of these findings.

Converse to several prepublished studies, this article finds that participation in intra-cluster relationships may lead to conflicts or create dysfunctionalities. For example, increased formality adds costs and interdependencies. Therefore, researchers should attempt to identify the contextual or firm-level factors that would provide more insights into when linkages may positively or negatively affect knowledge transfer. Additionally, the explanatory power of the models is relatively moderate. Further consideration of interaction effects would probably contribute to circumvent this question. Accordingly, development of consistent theoretical frameworks and hypotheses that test mediating and moderating effects will represent a privileged research path in the future. Finally, disentangling the different dimensions of the knowledge transferred and evaluating the particular role played by each sort of local institution represent crucial targets to achieve a nuance picture of knowledge circulation in Spanish clusters.

Footnotes

Acknowledgments

This research has benefited from the comments and suggestions of the editor and the three anonymous referees.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Spanish Ministry of Economy and Competitiveness [project number ECO2010-20557].

Notes

References

Agrawal

Cockburn

. 2003 “The Anchor Tenant Hypothesis: Examining the Role of Large, Local, R&D-intensive Firms in University Knowledge Transfer.” International Journal of Industrial Organization 21:1227–53.

Amin

Cohendet

. 2005. “Geographies of Knowledge Formation in Firms.” Industry and Innovation 12:465–86.

AEC (Asociación Española de Componentes del Calzado). 2008. INFOAEC Report. Elche, Spain: AEC. www.aec.es.

Barney

1991. “Firm Resources and Sustained Competitive Advantage.” Journal of Management 17:99–120.

Barney

1999. “How a Firm’s Capabilities affect Boundary Decisions.” Sloan Management Review 40:137–46.

Barney

2007. Gaining and Sustaining Competitive Advantage. Upper Saddle River, NJ: Prentice Hall.

Bathelt

Taylor

. 2002. “Clusters, Power and Place: Inequality and Local Growth in Time–space.” Geografiska Annaler B 84:93–109.

Becattini

1990. “The Marshallian Industrial District as a Socio-economic Notion.” In Industrial Districts and Inter-firm co-operation in Italy, edited by Pyke

Bcattini

Sengenberger

, 37–51. Geneva, Switzerland: International Institute for Labour Studies.

Becerra

Lunnan

Huemer

. 2008. “Trustworthiness, Risk, and the Transfer of Tacit and Explicit Knowledge between Alliance Partners.” Journal of Management Studies 45:675–98.

10.

Belso-Martinez

2006. “Do Industrial Districts Influence Export Performance and Export Intensity? Evidence for Spanish SMEs' Internationalization Process.” European Planning Studies 14: 791–810.

11.

Belso-Martinez

2010. “Outsourcing Decisions, Product Innovation and the Spatial Dimension: Evidence from the Spanish Footwear Industry.” Urban Studies 47:3057–77.

12.

Belussi

(2004): “In Search of a Useful Theory of Spatial Clustering.” Paper presented at the DRUID Conference on Industrial dynamics, innovation and development. Elsinore, Denmark, June 14–16.

13.

Belussi

2006. “In Search of a Useful Theory of Spatial Clustering.” In Clusters and Regional Development, edited by Asheim

Cooke

Martin

, 69–89. London: Routledge.

14.

Boix

Galletto

. 2006. “Sistemas Locales De Trabajo Y Distritos Industriales Marshallianos En España.” Economía Industrial 359:165–84.

15.

Boix

Galletto

. 2009. “Innovation and Industrial Districts: A First Approach to the Measurement and Determinants of the I-district Effect”. Regional Studies 43:1117–33.

16.

Boschma

Lambooy

. 2002. “Knowledge, Market Structure, and Economic Coordination: Dynamics of Industrial Districts.” Growth and Change 33:291–311.

17.

Boschma

ter Wal

. 2007. “Knowledge Networks and Innovative Performance in an Industrial District: The Case of a Footwear District in the South of Italy.” Industry and Innovation 14:177–99.

18.

Brusco

1982. “The Emilian Model: Productive Decentralization and Social Integration.” Cambridge Journal of Economics 6:167–84.

19.

Burt

1992. Structural Holes: The Social Structure of Competition. Cambridge, UK: Harvard University Press.

20.

Cainelli

Evangelista

Savona

. 2006. “Innovation and Economic Performance in Services: A Firm-level Analysis.” Cambridge Journal of Economics 30:435–58.

21.

Cainelli

Iacobucci

. 2007. Agglomeration, Technology and Business Groups. Cheltenham, UK: Edward Elgar.

22.

Camison

2004. “Shared, Competitive, and Comparative Advantages: A Competence-based View of Industrial-district Competitiveness.” Environment and Planning A 36:2227–56.

23.

Capaldo

2007. “Network Structure and Innovation: The Leveraging of a Dual Network as a Distinctive Relational Capability.” Strategic Management Journal 28:585–608.

24.

Clark

2010. “Are the Geographies of Innovation and Production Converging or Diverging? An Assessment of High-tech Employment in Regional Economies in the US.” Paper presented at the Association of American Geographers Annual Conference, Washington DC, April 14–16.

25.

Christopherson

Clark

. 2007. “Power in Firm Networks: What it means for Regional Innovation Systems.” Regional Studies 41:1223–36.

26.

Cohen

Levinthal

. 1990. “Absorptive Capacity: A New Perspective on Learning and Innovation.” Administrative Science Quarterly 35:128–52.

27.

Coleman

1990. Foundations of Social Theory. Cambridge, UK: Harvard University Press.

28.

Cooke

2005 “Regionally Asymmetric Knowledge Capabilities and Open Innovation Exploring ‘Globalisation 2’: A New Model of Industry Organisation.” Research Policy 34:1128–49.

29.

Cowling

Sugden

. 1994. “Human Resources and Industrial Development: New Social Roles.” International Journal of Technology Management 9:287–96.

30.

Cowling

Sugden

.1998. “The Essence of the Modern Corporation: Markets, Strategic Decision-making and the Theory of the Firm.” The Manchester School 66:59–86.

31.

De Man

2004. The Network Economy—Strategy, Structure and Management. Cheltenham, UK: Edward Elgar.

32.

De Propris

2001. “Systemic Flexibility, Production Fragmentation and Cluster Governance.” European Planning Studies 9:739–53.

33.

De Propris

Menghinello

Sugden

. 2008. “The Internationalisation of Production Systems: Embeddedness, Openness and Governance.” Entrepreneurship and Regional Development 20:493–515.

34.

Dyer

Singh

. 1998. “The Relational View: Cooperative Strategy and Sources of Inter-organizational Competitive Advantage.” Academy of Management Review 23:660–79.

35.

Easterby-Smith

Lyles

Tsang

. 2008. “Inter-organizational Knowledge Transfer: Current Themes and Future Prospects.” Journal of Management Studies 45:677–90.

36.

FICE (Federación de Industrias del Calzado Españolas). 2009. Footwear Sector Annual Report. Madrid, Spain: FICE. www.fice.es.

37.

Feldman

Francis

Bercovitz

. 2005. “Creating a Cluster while Building a Firm: Entrepreneurs and the Formation of Industrial Clusters.” Regional Studies 39:129–41.

38.

Foss

1996. “Higher-order Industrial Capabilities and Competitive Advantage: Some Analytical Suggestions.” Journal of Industry Studies 3:1–20.

39.

Foss

2006. “Knowledge and Organization in the Multinational Enterprise: Some Foundational Issues.” Journal of Management and Governance 10:3–20.

40.

Galbreath

2005. “Which Resources Matter the Most to Firm Success? An Exploratory Study of Resource-based Theory.” Technovation 25:979–87.

41.

Gereffi

Humphrey

Sturgeon

. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12:78–104.

42.

Giner

María

M. Santa

. 2002. “Territorial Systems of Small Firms in Spain: An Analysis of Productive and Organizational Characteristics in Industrial Districts.” Entrepreneurship and Regional Development 14:211–28.

43.

Giuliani

2007. “The Selective Nature of Knowledge Networks in Clusters: Evidence from the Wine Industry.” Journal of Economic Geography 7:139–68.

44.

Giuliani

Bell

. 2005. “The Micro-determinants of Meso-level Learning and Innovation. Evidence from a Chilean cluster.” Research Policy 34:47–68.

45.

Gordon

McCann

. 2000. “Industrial Clusters: Complexes, Agglomerations and/or Social Networks?” Urban Studies 37:513–32.

46.

Grandori

1997. “An Organizational Assessment of Interfirm Coordination Modes.” Organization Studies 18:897–925.

47.

Granovetter

1973. “The Strength of Weak Ties.” American Journal of Sociology 78:1360–80.

48.

Granovetter

2005. “The Impact of Social Structure on Economic Outcomes.” Journal of Economic Perspectives 19:33–50.

49.

Guerrieri

Pietrobelli

. 2004. “Industrial Districts’ Evolution and Technological Regimes: Italy and Taiwan.” Technovation 24:899–914.

50.

Hanneman

Riddle

. 2005. Introduction to Social Network Methods. Riverside, CA: University of California.

51.

Henderson

2007. “Understanding Knowledge Spill-overs.” Regional Science and Urban Economics 37:497–508.

52.

Hervas-Oliver

Albors-Garrigos

. 2009. “The Role of the Firm’s Internal and Relational Capabilities in Clusters: When Distance and Embeddedness are not Enough to Explain Innovation.” Journal of Economic Geography 9:263–83.

53.

Hess

2008. “Governance, Value Chains and Networks: An Afterword.” Economy and Society 37:452–59.

54.

Hollingsworth

1993. “Variation among Nations in the Logic of Manufacturing Sectors and International Competitiveness.” In Technology and the Wealth of Nations: The Dynamics of Constructed Advantage, edited by Foray

Freeman

. London, UK: Pinter.

55.

Humphrey

Schmitz

. 2004. “Governance in Global Value Chains.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, edited by Schmitz

, 95–109. Cheltenahm, UK: Edward Elgar.

56.

Jansen

van den Bosch

Volberda

. 2005. “Managing Potential and Realized Absorptive Capacity: How Do Organizational Antecedents Matter?” The Academy of Management Journal 48:999–1015.

57.

Jessop

1998. “The Rise of Governance and the Risks of Failure: The Case of Economic Development.” International Social Science Journal 155:29–45.

58.

Kaplinsky

Morris

Readman

. 2002. “The Globalisation of Product Markets and Immiserising Growth: Lessons from the South African Furniture Industry.” World Development 30:1159–78.

59.

Kogut

1996. “Direct Investment, Experimentation, and Corporate Governance in Transition Economies.” In Corporate Governance in Central Europe and Russia, edited by Frydman

Gray

Rapaczynski

, 293–332. London, UK: Central European University Press.

60.

Kogut

Zander

. 1992. “Knowledge of the Firm, Capabilities, and the Replication of Technology.” Organization Science 3:383–97.

61.

Lane

Salk

Lyles

. 2001. “Absorptive Capacity, Learning, and Performance in International Joint Ventures.” Strategic Management Journal 22:1139–61.

62.

Locket

Thompson

Morgenstern

. 2009. “The Development of the Resource- based View of the Firm: A Critical Appraisal.” International Journal of Management Reviews 11:9–28.

63.

MacDuffie

Helper

. 2006. “Collaboration in Supply Chains: With and Without Trust”. In The Firm as Collaborative Community, edited by Heckscher

Adler

, 417–68. New York, NY: Oxford University Press.

64.

Mahoney

Pandian

. 1992. “The Resource-based View within the Conversation of Strategic Management.” Strategic Management Journal 13:363–80.

65.

Markusen

1996. “Sticky Places in Slippery Space: A Typology of Industrial Districts.” Economic Geography 72:203–313.

66.

Marsili

Salter

. 2006. “The Dark Matter of Innovation: Design and Innovative Performance in Dutch Manufacturing.” Technology Analysis and Strategic Management 18:515–34.

67.

Maskell

Malmberg

. 1999. “Localised Learning and Industrial Competitiveness.” Cambridge Journal of Economics 23:167–85.

68.

Maskell

Malmberg

. 2007. “Myopia, Knowledge Development and Cluster Evolution.” Journal of Economic Geography 7:603–618.

69.

Mason

Leek

. 2008. “Learning to Build a Supply Network: An Exploration of Dynamic Business Models.” Journal of Management Studies 45:774–99.

70.

Molina-Morales

2008. “Los Distritos Industriales En La Europa Mediterránea. Las Diferencias Entre España e Italia.” Mediterráneo Económico 13:183–202.

71.

Molina-Morales

Martínez-Fernández

. 2004. “How Much difference is there between Industrial District Firms? A Net Value Creation Approach.” Research Policy 33:473–86.

72.

Moran

Ghoshal

. 1996. “Value Creation by Firms.” Academy of Management Best Paper Proceedings:41–45.

73.

Morrison

2008. “Gatekeepers of Knowledge within Industrial Districts: Who They are, How They interact.” Regional Studies 42:817–35.

74.

Morrison

Rabellotti

. 2009. “Knowledge and Information Networks in an Italian Wine Cluster.” European Planning Studies 17:983–1006.

75.

Morrison

Rabellotti

Zirulia

. 2008. “Local and Global Links and the Diffusion of Knowledge in Clusters: A Simulation Model.” Paper submitted to 25th DRUID celebration conference, Copenhagen, June 17–20.

76.

Newbert

2007. “Empirical Research on the Resource-based View of the Firm: An Assessment and Suggestions for Future Research.” Strategic Management Journal 28:121–46.

77.

Niosi

Zhegu

. 2005 “Aerospace Clusters: Local or Global Knowledge Spillovers?” Industry and Innovation 12:1–25.

78.

Paniccia

1998. “One, a Hundred, Thousands of Industrial Districts. Organizational Variety of Local Networks of Small and Medium-sized Enterprises.” Organization Studies 4:667–99.

79.

Parra-Requena

Molina-Morales

Garcia-Villaverde

P. M.

. 2010. “The Mediating Effect of Cognitive Social Capital on Knowledge Acquisition in Clustered Firms.” Growth and change 41:59–84.

80.

Parrilli

Sacchetti

. 2008. “Linking Learning with Governance in Networks and Clusters: Key Issues for Analysis and Policy.” Entrepreneurship and Regional Development 20:387–408.

81.

Peteraf

1993. “The Cornerstones of Competitive Advantage: A Resource Based View.” Strategic Management Journal 14:179–91.

82.

Pike

2009. Whither Regional Studies? London, UK: Routledge.

83.

Pla-Barber

2001. “The Internationalisation of Foreign Distribution and Production Activities. New Empirical Evidence from Spain.” International Business Review 10:455–74.

84.

Pla-Barber

Alegre

. 2007. “Analysing the Link between Export Intensity, Innovation and Firm Size in a Science-Based Industry”. International Business Review 16:275–93.

85.

Porter

1998. “Clusters and the New Economics of Competition.” Harvard Business Review, 76:77–90.

86.

Provan

Kenis

. 2008. “Modes of Network Governance: Structure, Management and Effectiveness.” Journal of Administration Research and Theory 18:229–52.

87.

Sacchetti

Sugden

. 2003. “The Governance of Networks and Economic Power: The Nature and Impact of Subcontracting Relationships.” Journal of Economic Surveys 17:669–91.

88.

Saliola

Zanfei

. 2009. “Multinational Firms, Global Value Chains and the Organization of Knowledge Transfer.” Research Policy 38:369–81.

89.

Sammarra

Biggiero

. 2001. “Identity and Identification in Industrial Districts.” Journal of Management and Governance 5:61–82.

90.

Sammarra

Biggiero

. 2008. “Heterogeneity and Specificity of Inter-firm Knowledge Flows in Innovation Networks.” Journal of Management Studies 45:785–814.

91.

Sreckovic

Windsperger

. 2011. “Organization of Knowledge Transfer in Clusters: A Knowledge-based View.” In New Developments in the Theory of Networks, edited by Tuunanen

Windsperger

Cliquet

Hendrikse

, 299–315. Berlin, Germany: Springer-Verlag.

92.

Teece

Pisano

Shuen

. 1997. “Dynamic Capabilities and Strategic Management.” Strategic Management Journal 18:509–33.

93.

Tokatli

2007. “Asymmetrical Power Relations and Upgrading among Suppliers of Global Clothing Brands: Hugo Boss in Turkey.” Journal of Economic Geography 7:67–92.

94.

Tomas

Contreras

del Saz

. 2000. “Institutos Tecnológicos Y Política Sectorial En Distritos Industriales. El caso del calzado.” Economía Industrial 334:25–34.

95.

Tortajada

Fernandez

Ybarra

. 2005. “Evolución De La Industria Española Del Calzado: Factores Relevantes En Las Últimas Décadas.” Economía Industrial 355:211–27.

96.

Tripsas

2007. “Customer Preference Discontinuities: A Trigger for Radical Technological Change.” Managerial and Decision Economics 29:79–97.

97.

Uzzi

1996. “The Sources and Consequences of Embeddedness for the Economic Performance of Organizations.” American Sociological Review 61:674–98.

98.

Van den Berg

Braun

Van Winden

. 2001. “Growth Clusters in European Cities: An Integral Approach.” Urban Studies 38:186–206.

99.

Van Wijk

Jansen

Lyles

. 2008. “Inter- and Intra-organizational Knowledge Transfer: A Meta-analytic Review and Assessment of its Antecedents and Consequences.” Journal of Management Studies 45:815–38.

100.

Watts

Wood

Wardle

. 2006. “Owner-managers, Clusters and Local Embeddedness: Small Firms in the Sheffield (UK) Metal-working Cluster.” Entrepreneurship and Regional Development 18:185–205.

101.

Williamson

1996. The Mechanisms of Governance. New York: Oxford University Press.

102.

Wolfe

Gertler

. 2004. “Clusters from the Inside and Out: Local Dynamics and Global Linkages.” Urban Studies 41:1071–93.

103.

Von Hippel

2005. Democratizing Innovation. Cambridge: MIT Press.

104.

Ybarra

2006. “Los Distritos Industriales en El Desarrollo Local Valenciano.” XIV International Economic History Congress, Helsinki, August 21–25.

105.

Yli-Renko

Autio

Sapienza

. 2001. “Social Capital, Knowledge Acquisition, and Knowledge Exploitation in Young Technology-based Firms.” Strategic Management Journal 22:587–613.

106.

Zahra

George

. 2002. “Absorptive Capacity: A Review, Reconceptualization, and Extension.” Academy of Management Review 27:185–203.

107.

Zaheer

Bell

. 2005. “Benefiting from Network Position: Firm Capabilities, Structural Holes and Performance.” Strategic Management Journal 26:809–25.