The role of international knowledge acquisition and absorptive capacity as a predictor of international performance

Received: 19 October 2020 - Revised: 25 August 2021 - Accepted: 6 September 2021 DOI: 10.1002/cjas.1651 RESEARCH ARTICLE The role of international knowledge acquisition and absorptive capacity as a predictor of international performance Maria A. Agustí | Encarnación Ramos‐Hidalgo | Ana M. Moreno‐Menéndez Departamento de Administración de Empresas y Marketing, Universidad de Sevilla, Sevilla, Spain Correspondence María A. Agustí, Departamento de Administración de Empresas y Marketing, Universidad de Sevilla, Sevilla, Spain. Email: [email protected] Abstract The international business literature recognizes the central role of knowledge and learning as key determinants of the internationalization of firms. While the stages model underlined the influence of experiential learning, new research has shown that there are also other relevant types of knowledge acquisition, which can be structured following Huber's model (1991) as congenital, grafted, vicarious, and search knowledge acquisition. However, knowledge acquisition constitutes only a first step for learning. In this respect, absorptive capacity appears as a useful construct since it integrates knowledge acquisition, assimilation, transformation, and exploitation capacity. This paper analyzes the influence of the types of knowledge acquisition on international performance and considers absorptive capacity dimensions as moderator variables based on a sample of 200 Spanish SME exporters. KEYWORDS absorptive capacity, internationalization, knowledge, learning Résumé La literatura sobre negocios internacionales reconoce el papel crucial del conocimiento y el aprendizaje como determinantes claves dentro del proceso de internacionalización. Así, mientras que el modelo secuencia resalta la influencia del aprendizaje a través de la experiencia, nuevas evidencias han mostrado que existen otras formas de adquirir conocimiento, lo que según el modelo Huber (1991) serían: aprendizaje congénito, aprendizaje experimental, aprendizaje vicario, aprendizaje por injerto y aprendizaje a partir de la búsqueda. Sin embargo, la adquisición de conocimiento es sólo el primer paso para el aprendizaje. En este sentido, la capacidad de absorción emerge como un constructor útil en cuanto integra las capacidades de adquisición de conocimiento, asimilación, transformación y explotación. Este artículo analiza, a través de una muestra de 200 empresas, la influencia de las diferentes formas This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made. © 2021 The Authors. Canadian Journal of Administrative Sciences / Revue Canadienne des Sciences de l'Administration published by John Wiley & Sons Ltd on behalf of ASAC. Can J Adm Sci. 2021;1–16. wileyonlinelibrary.com/journal/cjas - 1 de adquisición del conocimiento sobre el rendimiento internacional, considerando la capacidad de absorción como moderadora de esta relación. MOTS‐CLÉS apprentissage, capacitéd'absorption, connaissances, internationalization JEL CLASSIFICATION D83, L25, M16 1 | INTRODUCTION Decisions throughout the internationalization process are driven by the knowledge a firm possesses or lacks with regard to the new market (Cohen & Levinthal, 1990; Huber, 1991; Klepper, 1996). Several decades ago, theorists, who relied on the behavioral model of the firm, created stages models of internationalizing that depicted internationalization as a sequential or stages process of learning in which knowledge acquisition through participation in new markets leads to increasing international commitment (Johanson & Vahlne 1977,1990). This model implies a slow and gradual internationalizing while businesses acquire and accumulate experiential knowledge in a path‐dependent process (Eriksson et al., 1997). Dissatisfied with this depiction, Oviatt and McDougall (1994) developed a new‐venture internationalization perspective that similarly noted the importance of knowledge. By using the behavioral theory assumption, which states that managers seek to avoid uncertainty, the former model explained why most companies waited to expand abroad and subsequently proceeded slowly thereafter. The latter model employed a resourcebased framework, which explains that new ventures with appropriate knowledge, stocks, and capabilities did not follow the incremental pattern but instead leapt ahead, and utilized their internationalization as a proactive, competitive strategy. While both models used experiential knowledge as a key driver and could explain why some companies went slowly and others rapidly, neither model was specifically aimed at providing more insight into whether different knowledge acquisition activities might be more or less useful in predicting when the companies would actually start to sell abroad (Oviatt & McDougall, 2005). Furthermore, the ever‐increasing literature on this topic is claiming a wider approach in the analysis of the role of different types of knowledge acquisition activities in the firm's internationalization process (De Clercq et al., 2012; Forsgren, 2002). However, knowledge acquisition (KA) is only one part of the learning process, especially when the firm acquires knowledge externally. Information gathered from external sources (such as promotion agencies, consultants, and public information about foreign markets) needs to be assimilated and interpreted by the firm managers for the international opportunities to be exploited. The learning process is shaped as a dynamic interaction between the stock of knowledge and new knowledge. in which managerial experience, absorptive capacity, and knowledge complementarity, among other factors, play a crucial role. Absorptive capacity (ACAP) determines the potential usefulness of new knowledge to generate new products and new behavior, or to enter into new markets (Cohen & Levinthal, 1990). According to Zahra and George (2002), knowledge acquisition is one of four dimensions of ACAP. The other three are assimilation, transformation, and exploitation capabilities. The role of ACAP in the internationalization process has barely been researched, although both the stages model and international entrepreneurship perspectives have recognized its relevance (Johanson & Vahlne, 2009; Zahra, 2005). In an effort to fill this gap, our research analyzes the role of assimilation, transformation, and exploitation capabilities as three dimensions of the ACAP construct (Zahra & George, 2002): as a predictor of international performance; and as moderator variables of the relationship between different KA types (Huber, 1991), the first ACAP dimension (Zahra & George, 2002), and international performance. In essence, with this paper, we strive to answer the following research questions: How does ACAP influence the international performance of firms? Does ACAP directly influence international performance, does it moderate the effect of international knowledge acquisition activities, or both? We propose a model in which, following Huber's framework (De Clercq et al., 2012), the five types of international KA are considered, as well as the three ACAP post‐acquisition dimensions proposed by Zahra and George (2002): assimilation, transformation, and exploitation capabilities. This model contributes by offering a holistic view of the international learning process, by simultaneously combining acquisition and absorptive capacity as explanatory variables of internationalization; by opening, 2 - AGUSTÍ ET AL. to a certain extent, the black box of learning processes regarding the internationalization process of the firm; and by guiding managers toward the relevance of the entire process of learning, beyond merely KA activities. Five hypotheses are proposed, and empirical research, based on a sample of two hundred Spanish international firms, is developed through structural equation modeling. The structure of the paper is as follows. A brief summary is first provided of the research on the role of knowledge in internationalization processes, while highlighting the role of two main constructs traditionally used in the literature: KA activities and ACAP. An overarching hypothesis is then developed that links KA types and ACAP to international performance. Our methods, including sample selection, measurements, data validity, and analysis strategy, are described and the results are presented. The final section contains the discussion, implications, and conclusion. 2 | THEORETICAL DEVELOPMENT Knowledge plays a key role as an explainer and accelerator of internationalization (Autio et al., 2000; Sapienza et al., 2006) and can occur at several levels in a company (Casillas et al., 2009). However, the consideration of knowledge as a key factor in internationalization is not new. The stages model depicted internationalization as a recursive process in which knowledge acquisition increases international commitment, and, in turn, commitment increases knowledge acquisition (Johanson & Vahlne, 1990). This perspective emphasizes the role of experiential knowledge: learning from activities reduces perceived risk and thereby encourages commitment. It suggests that how quickly a firm chooses to carry out a cross‐border activity depends on managers' perceptions of the risks and of other alternatives available (Johanson & Wiedersheim‐Paul, 1975). According to this model, companies internationalize by first exporting to nearby countries to reduce the perceived risks of entering distant markets. Thus, businesses first choose countries at the shortest psychic and geographical distance from the focal firm (Benito & Gripsrud, 1992; Johanson & Vahlne, 1977), and subsequently expand out from that point. This view suggests that risk perceptions are inversely related to knowledge, and hence knowledge is essential in explaining whether and where companies begin to trade across borders. Stages model assumptions dominated the literature on internationalization from the 1970s to the mid‐ 1990s, and experiential learning consequently became the most analyzed source of knowledge acquisition. The international entrepreneurship perspective also embraces the importance of knowledge and of learning (Oviatt & McDougall, 2005; Rialp et al., 2005). Unlike the stage perspective, this suggests that pre‐existing (i.e., congenital) and other types of knowledge are also important (Autio et al., 2000; Oviatt & McDougall, 1994; Sapienza et al., 2006). Forsgren (2002) identified alternative sources of international knowledge as vicarious, grafted, and searching activities for objective information on foreign markets. Congenital and vicarious knowledge acquisitions are also investigated in the case of international new ventures (Bruneel et al., 2010; Fernhaber et al., 2009). Areas of knowledge identified as potentially relevant for exploration include the role of pre‐existing knowledge and processes (McDougall et al., 2003; Oviatt & McDougall, 1994), the path‐dependent nature of the experiential accumulation of knowledge (Autio et al., 2000; Johanson & Vahlne, 1990), the breadth, depth, and speed of recognizing and exploiting new knowledge opportunities (Zahra et al., 2000), and the role of active participation in international networks at individual, group, and organizational levels (Coviello & Munro, 1995; Johanson & Vahlne, 2009). Combinations of different types of knowledge acquisition are also of major interest. For example, Li et al. (2004) proposed a hybrid model that combined the role of experiential learning with that of systematic learning. Levesque et al. (2009) also theoretically explored the use of vicarious versus participative learning and proposed that the contribution of each type of learning to international performance would depend on the relative value of substituting costly participative learning with vicarious learning. To sum up, knowledge constitutes one of the main determinants of internationalization behavior. Nevertheless, most research in this field focuses on a narrow range of learning types (Cohen & Levinthal, 1990; Huber, 1991). The expanding set of issues identified by researchers regarding the role of knowledge in internationalization suggests that a broader framework is needed to organize research effectively in this area. In this respect, De Clercq et al. (2012) propose that Huber's model (1991) can provide a holistic framework for the analysis of a wide range of knowledge acquisition for internationalization. 2.1 | Knowledge acquisition types Knowledge acquisition (KA) is the process by which knowledge is obtained. It is an extensive and complex process in which other factors, such as information distribution, information interpretation, and organizational memory (Huber, 1991), also intervene. A consideration of KA types promises to inform both the congenital‐ AGUSTÍ ET AL. - 3 experiential learning debate implicit in the international new‐venture and stages model perspectives (Sapienza et al., 2006), and to illuminate the role of the other knowledge types in the internationalization of firms. We now review the five types of knowledge acquisition activities in Huber's (1991) scheme: congenital, experiential, vicarious, grafted, and search. Congenital KA refers to the processes involved in knowledge acquired prior to the inception of the business: “the individuals or organization that create new organizations have knowledge about the new organization's initial environment and about the processes the organization can use to carry out its creator's intentions, and they make this knowledge available to the new organization's members” (Huber, 1991, p. 91). Huber notes that congenital KA may involve institutionalized practices and procedures and context‐specific KA by founders prior to start‐up. The international new‐venture perspective explicitly expects such pre‐existing KA to be important for young firms' intentions regarding whether to begin or not to begin to internationalize (Oviatt & McDougall, 1994). Supporting this view, Bloodgood et al. (1996) found that the extent of internationalization one year after the initial public offering was positively related to the international experience and education of its executives prior to founding. Experiential KA is the process through which knowledge is acquired by a firm via direct experience after its birth (Huber, 1991). This may involve planned and unplanned learning, although Huber notes that unintended learning is by far the most common. (Huber names five sub‐processes, but, following the internationalization literature, only the simple general concept is examined herein). The stages model regards internationalization as an incremental accumulation of experiential international knowledge. Many studies provide evidence of the effects of accumulated experiential knowledge on internationalization (see, for instance, Johanson & Wiedersheim‐Paul, 1975; Johanson & Vahlne, 1977, 1990). The international new‐venture approach also recognizes the importance of path‐dependent experiential knowledge. The aim of this perspective, however, has been to explain why certain companies rapidly internationalize even when they have only minimal experience (Autio et al., 2000). The other three types of learning processes in Huber's typology (vicarious KA, grafted KA, and search KA) have remained relatively unexplored in the stages and newventure models. Huber (1991) refers to vicarious KA as learning through the second‐hand experience of other firms' strategies, practices, and technology. Grafted KA occurs when knowledge is brought into the firm by “acquiring and grafting on new members who possess knowledge not previously available within the organization” (Huber, 1991, p. 97). Finally, search KA is the process of obtaining knowledge by seeking and analyzing market information on external opportunities. It differs from vicarious KA in that search is more systematically or intentionally carried out on a variety of sources, not just through contact with or observation of competitors and similar firms (Huber, 1991). These five KA types have several differences. Congenital and grafted KA capture the stock of knowledge accumulated by the prior international experiences of the founders and managers. Both congenital and grafted KAs, like experiential learning, are essentially unintentional because they are shaped by individual or organizational international experiences. They are difficult to transfer between firms and business units (Johanson & Vahlne, 2009). Nevertheless, search KA is essentially intentional, since it seeks objective knowledge of international markets, which is publicly available and easy to transfer between organizations (Forsgren, 2002). Vicarious learning is mostly intentional in the way in which it is acquired. Firms intentionally look at other businesses that present a certain legitimacy regarding international behavior to learn from their experience. In brief, congenital, grafted, and experiential KAs are mainly unintentional, while search and vicarious KAs are in essence intentional. In order to link the knowledge to the international performance, we assume a behavioral theory approach, and following prior works that establish how decisions to commit resources to foreign operations are grounded on the firm's knowledge logic (Johanson & Vahlne, 2009). Consequently, as a starting point, we propose: H1 KA activities (congenital (a), grafted (b), experiential (c), vicarious (d), and search (e)) exert a positive influence on international performance. 2.2 | Absorptive capacity (ACAP) As previously explained, prior research has recognized that internationalization is a learning process per se. In this respect, much research has focused on knowledge acquisition activities. Nonetheless, learning encompasses more than simple KA (Barkema & Vermeulen, 1998; Sun & Anderson, 2010). Learning involves internal processes of assimilation, interpretation, mobilization, transformation, and exploitation, among others. Learning processes need new knowledge to interact with existing knowledge to develop new capabilities to compete in the market (Cohen & Levinthal, 1990). Consequently, ACAP is crucial as a dynamic capability embedded in a firm's 4 - AGUSTÍ ET AL. routines and processes (Lane et al., 2006; Zahra & George, 2002), which influence the firm's ability to create and deploy the knowledge to build other organizational capabilities (see for instance internationalization). In this context, Zahra and George (2002) divided ACAP into four consecutive dimensions: the acquisition, assimilation, transformation, and exploitation of knowledge. The first two are dimensions of potential capacity, while the latter two are dimensions of realized capacity. Acquisition refers to “a firm's capability to identify and acquire externally generated knowledge that is critical to its operations” (Zahra & George, 2002, p. 189). Assimilation refers to the “firm's routines and processes that allow it to analyze, process, interpret, and understand the information obtained from external sources” (Zahra & George, 2002, p. 189). Transformation captures the “firm's capability to develop the routines that facilitate combining existing knowledge and the newly acquired and assimilated knowledge” (Zahra & George, 2002, p, 190). And finally, exploitation is a firm's capability “based on the routines that allow firms to refine, extend, and leverage existing competencies or to create new ones by incorporating acquired and transformed knowledge into its operations” (Zahra & George, 2002, p, 190). In Zahra and George's conceptualization, KAs (Huber, 1991) are only the first stage of ACAP, and their impact on the performance of firms depends on the remaining internal dimension of ACAP (Cohen & Levinthal, 1990; Zahra & George, 2002). These four dimensions are complementary and interdependent. Nonetheless, independently considered, the firm's capability of acquiring, assimilating, transforming, and exploiting external knowledge exerts a positive impact on innovative outputs, and other outcomes that pertain to the creation of a competitive advantage (Jansen et al., 2005; Zahra & George, 2002). The previous hypothesis recognizes the influence of international KA activities on international performance, and therefore a similar impact of the other three ACAP dimensions on international performance can be proposed: H2 ACAP dimensions (assimilation (1), transformation (2), and exploitation (3)) exert a positive influence on international performance. 2.3 | KA activities and ACAP capabilities The learning process is usually considered as a pathdependent process. KA activities shape the future stock of available knowledge, but the stock of knowledge accumulated over time simultaneously guides the searching activities for new knowledge. Cohen and Levinthal (1990) stated that a stock of past knowledge is essential in order to assimilate new knowledge. New knowledge can easily be absorbed when it is complementary to knowledge existing inside the firm. Similarly, Zahra and George (2002) proposed that past experience influences the development of potential ACAP due to three related effects. First, past experience influences the development of future acquisition capabilities. Second, experience is connected to organizational memory, which influences future actions and performance (Moorman & Miner, 1996). And finally, experience has a significant impact on managerial cognition, which influences the firm's ability to manage information. Following Huber's typology of KA activities, the stock of knowledge accumulated by the individual experience of founders and/or managers in the past shapes how the firm will assimilate and manage future knowledge. In addition, experiential knowledge, developed through a learning‐by‐doing process over time has a potential effect on the firm's ability to assimilate, transform, and exploit new knowledge. As a consequence, we propose: H3 congenital (a), grafted (b), and experiential (c) KA activities exert a positive influence on assimilation (1), transformation (2), and exploitation (3) capabilities. ACAP is defined as a process itself (Zahra & George, 2002). Herein, KA is an antecedent of knowledge assimilation; knowledge assimilation influences knowledge transformation; and knowledge transformation affects knowledge exploitation. Acquisition, assimilation, transformation, and exploitation are consecutive steps of a continuous learning process that configures ACAP as a dynamic capability (Wu & Vahlne, 2020). H3 assumes part of this process perspective of ACAP. Following this logic, Zahra and George (2002) state that potential ACAP is an antecedent of realized ACAP, and both steps integrate the whole process of the firm's ACAP. Knowledge assimilation refers to the firm's capability to analyze, process, interpret, and understand the information obtained from external sources. The firm's higher assimilation capability improves its capacity to combine existing knowledge and the newly acquired and assimilated knowledge. A firm's transformation capability depends on how external information has previously been assimilated inside the firm. Similarly, the organizational capability of exploiting knowledge by incorporating internal knowledge into its operations depends on how external knowledge has been assimilated and transformed (Cohen & Levinthal, 1990; Zahra & George, 2002). Only knowledge that has already been AGUSTÍ ET AL. - 5 adequately internalized is ready to use. Following this process perspective of ACAP, we propose: H4 The assimilation capability of the firm exerts a positive influence on the firm's transformation capability (a), and the firm's transformation capability exerts a positive influence on its exploitation capability (b). The stages model asserts that experiential knowledge provides the driving force in the internationalization process of firms (Johanson & Vahlne, 1977,2009). According to behavioral theory (Cyert & March, 1963), the basic assumption is that learning‐by‐doing is more valuable for managers than external and objective knowledge (Forsgren, 2002). Due to the tacit character of market knowledge (Johanson & Vahlne, 1977), knowledge is highly dependent on individuals, and is therefore difficult to transfer to other firms or contexts (Penrose, 1959). External knowledge needs to be internalized and integrated into a stock of individual and experiential knowledge in order for it to be exploited. In this respect, ACAP is perceived as a dynamic capability that enhances the effect of external knowledge on its potential exploitation in international markets. External and intentional knowledge, like vicarious and search KA activities, has to be absorbed by the firm. It has to be assimilated, transformed, and exploited through a combination process of different complementary types of knowledge (knowledge internally stored inside the organization and externally‐acquired knowledge). Consequently, assimilation, transformation, and exploitation capabilities enhance the effect of external and intentional KA activities on international performance. For this reason, we propose: H5 The ACAP dimensions positively moderate the influence of external KA activities, like vicarious (a) and search (b) KA on international performance. Figure 1summarizes the model proposed. 3 | METHODOLOGY 3.1 | Sample The sample was selected from the directory of Spanish exporters and importers, put together by the Superior Council of Chambers of Commerce (Consejo de Cámaras de Comercio), which uses information provided by Public Finance. Firms were selected from five different sectors, characterized by a high proportion of exporters (chemical, food and beverages, textile, manufacturing, and commerce). The total population was made up of 3,158 firms, from which approximately 50% were exporters. Questionnaires were sent to 1,524 firms, 221 of which no longer existed or had erroneous contact information. We achieved 200 valid responses (15.35% of 1,303) by using interviewers. The answers given by the firms were analyzed in terms of response timing, and no significant differences were found between those that answered first and the rest. The characteristics of the sample are shown in Table 1. FIGURE 1 Model and hypotheses TABLE 1Sample characteristics NPercentage Age Less than 10 years 29 14.50 From 11 to 25 years 63 31.50 From 26 to 50 years 71 35.50 From 51 to 100 years 27 13.50 More than 100 years 10 5.00 Total 200 100.00 Size Fewer than 10 employees 23 11.50 From 10 to 49 employees 103 51.50 From 50 to 249 employees 52 26.00 250 or more employees 22 11.00 Total 200 100.00 Sector Food and beverages 64 32.00 Chemicals 34 17.00 Manufacturing 64 32.00 Textile 18 9.00 Commerce 20 10.00 Total 200 100.00 6 - AGUSTÍ ET AL. 3.2 | Variables 3.2.1 | International performance The final endogenous variable has been measured in terms of export intensity (the ratio between exports and total sales), which is traditionally used as an indicator of export commitment (Filatotchev et al., 2008; Lages et al., 2008). 3.2.2 | Knowledge acquisition types Five multi‐item scales were created to measure the five KA types. Each item was measured on a seven‐point Likert scale. We created the scales from a review of the literature and several interviews with three experts on internationalization, two members of two agencies for internationalization promotion, and one private consultant on internationalization (see Appendix Afor the items). The questionnaire used by Casillas et al. (2015) showed coincidence with our objectives but omitted the influences of the export promotions agency (Acedo & Casillas, 2007). Thus, this questionnaire was modified accordingly and, once the questions had been written, they were personally passed to seven export managers in order to monitor that the items were well understood, and to three different managers in an export promotion agency to verify the adequacy of the modifications. In order to test whether our modifications lead to adequate results, the intraclass correlation coefficient (ICC), cross loadings, and Cronbach's alpha coefficient were employed to test the reliability between test and retest and internal consistency of all items, respectively. Furthermore, the questionnaires were split into subsamples, thereby obtaining adequate results with no significant difference in the application of this instrument. Our analyses showed that, as expected, five factors emerged, and each item was grouped in the factor proposed (one factor for each KA type [Huber, 1991], similar to that in Casillas et al. [2015]). 3.2.3 | Absorptive capacity Three different Likert scales were used (seven points), one for each ACAP dimension (assimilation, transformation, and exploitation), as proposed by Jansen et al. (2005). A confirmatory factor analysis was also performed to ensure that the measures were reliable, and three different factors were found. Finally, we have included three control variables. The first is the age at entry, measured as the number of years between the foundation and the firm's first international activities (Gabrielsson et al., 2008; Rialp et al., 2005). The second control variable is the firm's age (Autio et al., 2000; Sapienza, et al., 2006), and, finally, research and development intensity (Autio et al., 2000). 3.3 | Analysis This model, despite being recursive in its nature as a learning process, is formed as a dynamic interaction between the stock of knowledge and new knowledge acquisition. Thus, “although in a state of constant flux, the firm can at any point in time be described in terms of its capabilities, resource positions […] which in turn affect the processes in the next time period” (Vahlne & Johanson, 2017, p. 1089). By this approach, we try to infer the dynamic property of the model by linking state and change variables, transforming them as independent and dependent variables (Vahlne & Johanson, 2017) to deepen our insights into the constructs and relations commonly studied empirically. This approach has commonly been used both in the study of the internationalization process (Welch & Paavilainen‐Mantymäki, 2014) and in learningrelated studies (Casillas et al., 2015). A structural equation modeling (SEM) is proposed to assess the relationships between the constructs, along with the predictive power of the research model. The partial least squares (PLS) technique was employed, since this tool is primarily intended for causal‐predictive analysis in cases where the problems explored are complex and theoretical knowledge remains scarce. However, PLS is also an appropriate technique to use in a theory development situation (Hair, Risher, et al., 2019; Hair, Sarstedt, et al., 2019). Despite the substantial amount of literature regarding these issues, our model proposes that there is a link between the different constructs, which justifies the exploration of a theoretical extension of previously established theories (Chin, 2010; Hair, Risher, et al., 2019). Furthermore, PLS is particularly suitable for complex models that have a high number of constructs, indicators, and/or relationships, as is the case presented here (Chin, 2010; Hair, Risher, et al., 2019, Hair, Sarstedt, et al., 2019). The model was estimated with SmartPLS 3.2.7 (Ringle et al., 2015). The different constructs were estimated in Mode A (correlation weights), which is advisable when the indicators are correlated (Becker et al., 2013). Common method bias was controlled by performing a full collinearity test based on variance inflation factors (VIFs) (Kock, 2015) designed to evaluate both vertical and lateral collinearity. Each of the VIF values obtained for each construct lies below the 3.3 threshold that could lead to a statistical problem (Kock, 2015). AGUSTÍ ET AL. - 7 4 | RESULTS Table 2summarizes the main descriptive statistics and the correlation matrix. The analysis and interpretation of a PLS model involves a two‐stage process. First, we assessed the reliability and validity of the measurement model; and second, carried out the evaluation of the structural model. This sequence ensures that the measures of the constructs are valid and reliable before attempting to draw any conclusions from the relationships existing between said constructs (Barclay et al., 1995). The main parameters corresponding to the measurement model appear in Table 3. The first step consists of analyzing the reflective constructs through the reliability of scales (Cronbach's alpha). Table 3first shows the lower value for loadings (reflective items) and weights (formative items) of each of the constructs to analyze their individual reliability. All items attain values above 0.7. The reliability of the reflective constructs is represented by the composite reliability (which should remain higher than 0.7, according to Fornell & Lacker [1981]). The table also includes the convergent validity of latent variables, measured using the average variance extracted (AVE), which must be over 0.5 (Fornell & Lacker, 1981). As Table 3shows, all constructs surpass this threshold. Finally, the discriminant validity must be analyzed. This measures whether the constructs do indeed differ. For this purpose, the AVE should be greater than the variant TABLE 3Construct statistics AVE Composite reliability Cronbach's alpha Communality International performance – – – 0.734 Congenital KA 3.504 1.223 0.851 0.897 Grafted KA 3.373 1.215 0.866 0.920 Experiential KA 1.991 1.146 0.852 0.936 Vicarious KA 1.783 1.175 0.796 0.950 Search KA 1.940 1.138 0.859 0.956 Assimilation 1.090 1.028 0.914 0.990 Transformation 1.266 1.037 0.898 0.958 Exploitation 1.239 1.034 0.912 0.949 Age 0.941 0.941 1.000 1.000 R&D intensity 2.943 2.943 1.000 1.000 Age at entry 0.544 0.544 1.000 1.000 TABLE 2Descriptive statistics and correlation matrix Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 1. International performance 41.88 29.550 1.000 2. Congenital KA 3.356 2.070 0.230 1.000 3. Grafted KA 3.968 1.964 0.272 0.646 1.000 4. Experiential KA 5.031 1.520 0.473 0.435 0.455 1.000 5. Vicarious KA 5.082 1.482 0.420 0.248 0.292 0.496 1.000 6. Search KA 4.495 1.581 0.413 0.260 0.262 0.588 0.548 1.000 7. Assimilation 4.923 1.543 0.216 0.257 0.210 0.294 0.240 0.383 1.000 8. Transformation 4.193 1.392 0.260 0.325 0.310 0.498 0.418 0.514 0.671 1.000 9. Exploitation 4.048 1.620 0.232 0.289 0.205 0.424 0.391 0.480 0.592 0.716 1.000 10. Ln Firm age 3.320 2.482 0.156 −0.271 −0.102 0.031 0.018 0.068 −0.080 −0.046 −0.093 1.000 11. Ln Age at entry 2.443 2.105 0.220 −0.038 0.036 0.168 0.108 0.101 −0.024 0.071 0.050 0.609 1.000 12. R&D intensity 0.104 0.215 0.231 −0.043 0.104 0.076 0.022 0.108 0.023 0.007 0.056 0.003 −0.061 1.000 8 - AGUSTÍ ET AL. shared between two constructs in the model. For suitable discriminant validity, the diagonal elements should be significantly greater than the off‐diagonal elements in the corresponding rows and columns (Barclay et al., 1995). All of our constructs satisfy this condition. Once the measurement model has been validated, the second stage then involves the estimation of the structural model. Structural models attempt to respond to two basic questions (Falk & Miller, 1992) regarding: the amount of the variance of the endogenous variables that can be explained by the latent variables that predict said variance; and the extent to which the predictive variables contribute toward the explained variance of the endogenous variables. In order to answer these questions, two basic indicators are usually employed: R 2 and the standardized path coefficient, β. As a measure of the predictive power, the R 2 can be interpreted in the same way as those obtained in multiple regression analysis. On this matter, Falk and Miller (1992) establish that suitable values are those that are equal to or greater than 0.1. Table 4and Figure 2summarize the main parameters (R 2 as a measure of the variance explained, and path coefficients, β). Bootstrapping (1,000 resamples) was applied to generate standard errors and t‐statistics (Chin, 1998; Davidson & MacKinnon, 2000) to estimate the standard error and t‐values of the parameters. Results show the high explicative power of the endogenous constructs. In the case of the final construct (international performance), R 2 shows that the model explains 45.09% of the variance. The variance explained by two of the three absorptive capacity constructs (transformation and exploitation capabilities) is even higher. In both cases, the predictors explain more than 50% of their variances (51.98% for the knowledge transformation capability, and 62.21% for the knowledge exploitation capability). However, the model explains only 17.32% of the variance of knowledge assimilation capability. Once the explicative power of the model has been analyzed, we can center on the standardized path coefficients in order to test the five hypotheses proposed (results are summarized in Table 5). H1 posits that international KA positively affects international TABLE 4R‐square of endogenous variables R square Assimilation 0.1732 Transformation 0.5198 Exploitation 0.6221 International performance 0.4509 FIGURE 2 Structural equation modeling and hypothesis tests. †p< 0.1; *p< 0.05; **p< 0.01; ***p< 0.001 AGUSTÍ ET AL. - 9 Vahlne, J. E., & Johanson, J. (2017). From internationalization to evolution: The Uppsala model at 40 years. Journal of International Business Studies,48(9), 1087–1102. Welch, C., & Paavilainen‐Mäntymäki, E. (2014). Putting process (back) in: Research on the internationalization process of the firm. International Journal of Management Reviews,16(1), 2–23. Wu, A., & Voss, H. (2015). When does absorptive capacity matter for international performance of firms? Evidence from China. International Business Review,24(2), 344–351. Wu, J., & Vahlne, J. E. (2020). Dynamic capabilities of emerging market multinational enterprises and the Uppsala model. Asian Business & Management. https://doi.org/10.1057/s41291‐020‐ 00111‐5 Yeoh, P. L. (2005). A conceptual framework of antecedents of information search in exporting. International Marketing Review, 22(2), 165–198. Zahra, S. A. (2005). A theory of international new ventures: A decade of research. Journal of International Business Studies,36(1), 20–28. Zahra, S. A., & George, G. (2002). Absorptive capacity: A review, conceptualization, and extension. Academy of Management Review,27(2), 185–203. Zahra, S. A., Ireland, R., & Hitt, M. A. (2000). International expansion by new venture firms: International diversity, mode of market entry, technological learning, and performance. Academy of Management Journal,43(5), 925–950. How to cite this article: Agustí, M. A., Ramos‐ Hidalgo, E., & Moreno‐Menéndez, A. M. (2021). The role of international knowledge acquisition and absorptive capacity as a predictor of international performance. Canadian Journal of Administrative Sciences, 1–16. https://doi.org/10. 1002/cjas.1651 Appendix A Knowledge acquisition items All items inquired as to the extent to which the respondent agreed with the statements (1 = totally disagree to 7 = totally agree). Congenital knowledge acquisition: International experience of the founders of your business before it was founded. Grafted knowledge acquisition: Sources of prior international learning of managers/executive added to the business after founding. Experiential knowledge acquisition: Experience of the company in international activities (excluding exportation itself). Vicarious knowledge acquisition: International knowledge obtained by the company through observing others. Search knowledge acquisition: Company activities to search for information about internationalization opportunities. Congenital 1 They had international working experience. Congenital 2 They obtained degrees or studied abroad. Congenital 3 They participated in international cooperation networks. Congenital 4 They studied foreign language(s). Grafted 1 They had international working experience. Grafted 2 They obtained degrees or studied abroad. Grafted 3 They participated in international cooperation networks. Grafted 4 They studied foreign language(s). Experiential 1 The company engages in a variety of international activities. Experiential 2 The company is regularly involved in activities related to other exporters. Experiential 3 We interact with many foreign clients. Experiential 4 The company has managers that usually travel for business. Vicarious 1 The company interacts with foreign competitors who have started international expansion. Vicarious 2 The company interacts with domestic competitors who have started international expansion. Vicarious 3 The company pays close attention to companies that are growing internationally. Search 1 The company devotes time and people to the search for opportunities abroad. Search 2 The company has systems to gather information regarding foreign markets. Search 3 The company actively searches for all kinds of information on possible opportunities abroad. Search 4 The company frequently contacts public agencies for export activities. 16 - AGUSTÍ ET AL.