Internationalization and entrepreneurial orientation of family SMEs: The influence of the family character

This is the accepted manuscript of the article that appeared in final form in International Business Review 28(1) : 48-59 (2019), which has been published in final form at https://doi.org/10.1016/j.ibusrev.2018.06.003. © 2018 Elsevier under CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/) 1 Internationalization and entrepreneurial orientation of family SMEs: The influence of the family character Mikel Alayo, Amaia Maseda, Txomin Iturralde, & Unai Arzubiaga Abstract Internationalization is an important entrepreneurial strategy for promoting the long-term growth and survivability of small and medium-sized enterprises (SMEs). Family involvement in top management teams (TMTs) can explain the heterogeneous behaviors of these firms’ international entrepreneurship process. This paper analyzes the moderating effects of the family’s influence on the relationship between entrepreneurial orientation and internationalization with two TMT diversities found only in family firms: the family TMT ratio and generational involvement. An analysis of 191 Spanish family SMEs indicated that entrepreneurial orientation plays a significant role in explaining the degree of internationalization in family firms and that a diversely formed TMT shapes this relationship. A high concentration of family members in managerial positions hinders the international entrepreneurship process. This fact highlights the importance of hiring non-family managers to promote internationalization. The results also reveal that involving multiple generations in decision-making hampers entrepreneurial internationalization, generating control and coordination problems. Keywords: internationalization, entrepreneurial orientation, family firms, family involvement, generational involvement, top management team 2 1. Introduction Entrepreneurial orientation (EO), the firm’s disposition to entrepreneurship, is a key element in businesses’ internationalization process (Jantunen, Puumalainen, Saarenketo, & Kyläheiko, 2005; Javalgi & Todd, 2011; Liu, Li, & Xue, 2011; Ripollés-Meliá, Menguzzato-Boulard, & Sánchez-Peinado, 2007). However, few studies have analyzed factors affecting the relationship between EO and international development in small and medium-sized enterprises (SMEs), and even fewer have analyzed family firms (Calabrò, Campopiano, Basco, & Pukall, 2017; Hernández-Perlines & Mancebo-Lozano 2016; Hernández-Perlines, Moreno-García, & YañezAraque, 2016), although such firms represent the most common form of business organization in the world (Hiebl, Quinn, Craig, & Moores, 2018). The literature has acknowledged that family firms differ in attitudes and behaviors when internationalizing (Graves & Thomas, 2006) and in internationalization strategies (e.g., Fernández & Nieto, 2006; Boellis, Mariotti, Minichilli, & Piscitello, 2016). Family firms may behave differently depending on the extent of family involvement in the business (Chrisman, Chua, & Steier, 2005; Kellermans, Eddleston, Sarathy, & Murphy, 2012; Naldi, Nordqvist, Sjöberg, & Wiklund, 2007). Furthermore, family involvement is a variable used commonly to identify the family’s power to shape a firm’s goals, strategies, and behaviors (Deephouse & Jaskiewicz, 2013; Miller, Le Breton-Miller, & Lester, 2013). Decision makers’ characteristics and attitudes are important determinants of family SMEs’ internationalization (Arregle, Naldi, Nordqvist, & Hitt, 2012; Calabrò, Torchia, Pukall, & Mussolino, 2013; Cerrato & Piva, 2012; Claver, Rienda, & Quer, 2008). However, the relevance of individual managerial levels of knowledge and experience remains largely unexplored in the international business literature (Nielsen, 2010). In family SMEs, there is often an overlap between ownership, the board of directors, and the top management team (TMT), with the same family members involved at all levels (Segaro, 2012). TMT members in family SMEs participate directly in the design and day-to-day implementation of firm strategy (Lubatkin, Simsek, Ling, & Veiga, 2006; Sánchez-Marín & Baixauli-Soler, 2015), and they carry out the entrepreneurship process (Sciascia, Mazzola, & Chirico, 2013). Moreover, it is common for SMEs to have family members in TMT positions (Speckbacher & Wentges, 2012) that can spread the family’s ideas and behaviors, directly influencing strategic choices and decisions. Therefore, the family’s involvement in the firm offers a unique environment in which to analyze whether and to what extent a firm’s family character affects the firm’s internationalization (Fernández-Olmos, Gargallo-Castel, & Giner-Bagües, 2016). 3 The family business literature has overlooked the importance of family involvement in TMT positions despite its relevance (Minichilli, Corbetta, & MacMillan, 2010). Following the upper echelon theory (Carpenter, Geletkanycz, & Sanders, 2004; Hambrick, 2007; Hambrick & Mason, 1984), top managers’ strategic choices are influenced by managers’ cognitive base and values, which influence the decision-making process, shaping organizational outcomes. Organizations are reflections of their TMTs, and firm-level decisions, including internationalization, depend substantially on team background, experience, and values (Hiebl, 2014; Tihanyi, Ellstrand, Daily, & Dalton, 2000). Thus, managers should possess knowledge that allows them to analyze and process information efficiently in the complex and uncertain internationalization process (Nielsen & Nielsen, 2011; Ramón-LLorens, García-Meca, & Duréndez, 2017). Previous research has recognized TMT as an important decision-making group in organizations (Sánchez-Marín & Baixauli-Soler, 2015; Vandekerkhof, Steijvers, Hendriks, & Voordeckers, 2015) and has focused on whether demographic diversity in managerial backgrounds is advantageous for firms, but the results have been inconsistent (Cannella, Park, & Lee, 2008; Rivas, 2012). The family firm context represents an interesting research stream because these organizations’ singularity results in specific TMT diversities and greater complexity than non-family firms (Ling & Kellermanns, 2010; Sciascia et al., 2013). To date, family firm-specific TMT diversities represent under-researched sources (Kraiczy, Hack, & Kellermanns, 2014; Ling & Kellermanns, 2010), and research regarding this aspect remains inconclusive (Sciascia et al., 2013). To shed light on this issue, we analyze the moderating effects on the EOinternationalization relationship of the two main forms of TMT diversities created by the family’s involvement (Arzubiaga, Maseda & Iturralde, 2018a; Kraiczy et al., 2014): the family TMT ratio (the diversity between family and non-family managers) and generational involvement (the diversity among generations of family members that work together in the TMT). We consider that TMT diversities could relate to the firm’s entrepreneurial internationalization, and a separate analysis of these two family involvement measures provides a better understanding of the diverse effects of the family’s influence on family SMEs’ international development. Relying on a sample of Spanish family firms, our results highlight the importance of TMT composition in family SMEs for shaping the relationship between EO and internationalization. Our study contributes in several ways to the research and management practice of family SMEs. First, we increase our knowledge of the upper echelons theory by providing a more fine- 4 grained understanding of the consequences of two family firm-specific TMT diversities. We address the call of previous studies by providing new empirical evidence for the family’s involvement in firms’ managerial positions, in terms of internationalization (Cerrato & Piva, 2012; Mitter, Duller, Feldbauer-Durstmüller, & Kraus, 2014; Segaro, Larimo & Jones, 2014). Existing studies have overlooked the effect of family involvement in TMTs, even though a significant body of research literature suggests that the TMT is an appropriate test of the family’s influence on businesses (Kraiczy et al., 2014; Sciascia et al., 2013; Segaro et al., 2014). This study offers an enhanced theoretical examination of upper echelon theory that allows us to separate the benefits and drawbacks of family-specific TMT diversities (De Massis, Kotlar, Campopiano, & Cassia, 2013) and presents a comprehensive picture of the configurations of family involvement in management that are most favorable or adverse to turning EO into a successful internationalization process. Second, this study extends the literature on internationalization and EO (e.g., Brouthers, Nakos & Dimitratos, 2015; Jantunen et al., 2005) by adding to the small amount of research linking the entrepreneurship and international business fields in the family firm context (Calabrò et al., 2017; Hernández-Perlines & Mancebo-Lozano 2016; Hernández-Perlines et al., 2016). Our study is apparently the first that analyzes how entrepreneurial behavior’s effect in a firm’s internationalization may be moderated due to family-related factors. This study’s setting is noteworthy because Spain features a predominance of family-controlled firms, which comprise around 90% of Spain’s active businesses in 2015 according to the Spanish Family Business Institute (Casillas, López, Meroño, Pons, & Baiges, 2016). Third, this research contributes new insights to the emergent literature on family firm heterogeneity (Chua, Chrisman, Steier, & Rau, 2012; Miller, Minichilli, & Corbetta, 2013), showing that family involvement in TMTs can be a more important driver of variation, in terms of success, when turning EO into internationalization among family SMEs than was previously thought. By focusing on family firm-specific TMT diversity sources, this research extends the knowledge of the family business governance dimensions’ effects, which were identified as major factors in family firms´ heterogeneity (Li & Daspit, 2016; Pittino, Barroso-Martínez, Chirico, & Sanguino-Galván, 2018). The paper is organized as follows. In the next section, the theoretical background is provided, and hypotheses are developed. The third section provides empirical data collected from Spanish family SMEs, and the fourth section validates the hypotheses through structural equation modeling. The fifth section then presents the discussion. Finally, contributions, limitations, and recommendations for future research are provided. 5 2. Theoretical background and hypotheses development 2.1. Entrepreneurial orientation and internationalization Firms’ inclination toward entrepreneurship is broadly conceptualized as EO (Covin & Wales, 2012). Although the EO construct was developed to explain entrepreneurial behaviors in domestic markets (Covin & Slevin, 1991), its use in an international context is justified (Covin & Miller, 2014). Internationalization is an entrepreneurial act (Jantunen et al., 2005; Jones & Coviello, 2005; Liu et al., 2011; Lu & Beamish, 2001) because it entails identifying and exploiting new business opportunities in new environments (Ripollés-Meliá et al., 2007) and combines risk acceptance and the ability to innovate (Fletcher, 2004). Previous studies have demonstrated EO’s positive effect on a firm’s degree of internationalization (Jantunen at al., 2005; Javalgi & Todd, 2011; Liu et al., 2011; RipollésMeliá et al., 2007). These studies conclude that firms with higher EO are more likely to introduce new products, diversify activities, and prosper in unknown environments (Brouthers et al., 2015; Dimitratos, Lioukas, & Carter, 2004). EO provides an important capability to build competitive advantage (Brouthers et al., 2015; Wiklund & Shepherd, 2003) because it facilitates the identification of new business opportunities (Webb, Ketchen, & Ireland, 2010) and contributes to a firm’s continuity and success (Kellermanns & Eddleston, 2006). Firms with higher EO tend to be more proactive and, thus, search for opportunities in new international markets (Jantunen et al., 2005). According to Brouthers et al. (2015, p. 1165), SMEs with greater EO “will perform better in foreign markets because they possess the capabilities needed to develop innovative strategies that provide an advantage in the foreign market, identify and use technologies that better align with foreign market customer needs, and are willing to take business risks associated with adopting new strategies and technologies in foreign markets.” Thus, the degree of internationalization is a condition caused by firms’ entrepreneurial attitudes. Although family firms represent an interesting setting due to their characteristics and EO has been identified as a relevant factor affecting family firms’ strategic decision-making processes (Basco & Voordeckers, 2015; Moreno & Casillas, 2008), the relationship between entrepreneurial behavior and internationalization in the family firm context remains inconclusive, with very little research in this respect (Calabrò et al., 2017; Hernández-Perlines & Mancebo-Lozano 2016; Hernández-Perlines et al., 2016). Some authors argue that family firms are conservative, risk-averse, and reluctant to change (Fernández & Nieto, 2005; Mitter 6 et al., 2014), and, thus, are the opposite of entrepreneurial (Naldi et al., 2007). Nevertheless, family firms possess beneficial attributes such as the ability to make fast decisions, trusted relationships, and a long-term orientation that can make them successful in international environments (Fernández & Nieto, 2006; Kontinen & Ojala, 2010; Lumpkin, Brigham, & Moss, 2010; Mitter et al., 2014). This last attribute is especially important because it can take years to generate profits in international markets (Zahra, 2003). These positive and negative characteristics are generated by the family’s influence on firm management. Furthermore, family firms are not homogenous in terms of strategic behavior because, depending on the degree of family involvement, they may behave in different ways (Botero, Thomas, Graves, & Fediuk, 2013). In this paper, we study the role of the family’s involvement in TMT positions, since different measures of the family’s influence could provide a better understanding of how and to what extent families moderate EO’s effect on internationalization. 2.2. Family involvement in the top management team Heterogeneity among family firms can be appreciated through various factors, such as firms’ ownership structure, governance mechanisms, levels of professionalization, and the generation in charge of the firm (Chrisman et al., 2005; Kellermanns et al., 2012; Sharma, 2004). This heterogeneity can also be perceived in family firms’ attitudes and behaviors when facing internationalization (Segaro et al., 2014; Siebels & Knyphausen, 2012). Calabrò et al. (2013, p. 511) asserted that “family involvement seems to be an important determinant for the internationalization process”. This research focuses on the level of family involvement in firms’ TMT, which is recognized as an important decision-making unit in organizations (Sánchez-Marín & Baixauli-Soler, 2015; Vandekerkhof et al., 2015). Previous research has analyzed family involvement in different governance bodies, such as the board of directors, instead of the TMT. However, not all family SMEs have a board because this is not always compulsory. Although boards may be compulsory in some contexts, the role of the board of directors in SMEs may vary greatly, ranging from a symbolic function to active involvement in value-creating activities, such as strategic decision-making and access to valuable resources (Huse, 2007; McNulty and Pettigrew, 1999; Westphal and Zajac, 1998). This view is corroborated by evidence that many family SMEs do not make full use of their boards (Institute for Family Business, 2016), with the board of directors role usually replaced by informal controls (De Massis, Kotlar, Campopiano & Cassia, 2015). Conversely, in SMEs TMT managers are usually the individuals who participate more directly in the daily design and 7 implementation of a firm’s strategy (Lubatkin et al., 2006; Sánchez-Marín & Baixauli-Soler, 2015), carrying out the entrepreneurial processes (Sciascia et al., 2013) and determining the depth and scope of internationalization and other international strategies (Aharoni, Tihanyi, & Connelly, 2011). Therefore, the family’s influence over managerial positions is an important explanation of firm behavior (Kellermanns et al., 2012; Steier, 2003), and its importance has been overlooked (Minichilli et al., 2010). The upper echelon theory notes that firm outcomes reflect the TMT’s actions (Carpenter et al., 2004; Hambrick, 2007; Hambrick & Mason, 1984). This theory posits that strategic decisions made in the firm are associated with TMT members’ cognition and values, because such members are among the most powerful actors in the organization (Hambrick, 2007). Decisions are influenced by managers’ cognitive mental schemes, which are represented by their educational and functional backgrounds, experiences, perspectives, values, and demographic characteristics (Hiebl, 2014; Tihanyi et al., 2000). Situations are interpreted differently, different actions are taken, and, thus, differences in firm outcomes emerge from TMT composition (Hambrick, 2007). Research on TMTs has analyzed how diversity of managers’ characteristics and backgrounds influences firms’ outcomes (Rivas, 2012), and, therefore, it is argued that the ideas and opportunities pursued are shaped by TMT diversity (Beckman, 2006). Thus, the importance of having diversity among TMT members’ skills, knowledge, and experiences to improve problem-solving, facilitate new ideas, and ensure proactive actions is highlighted (Talke, Salomo, & Kock, 2011). Firms can gain significant competitive advantage from diversely formed TMTs (Bunderson, 2003), because a more diverse TMT provides greater creativity, superior problem-solving capacity, and the ability to innovate. A constructive dialogue exists among members in a diverse TMT that fosters the emergence of diverse ideas, knowledge, and perspectives, which are not achievable in homogeneous TMT groups (Sciascia et al., 2013; Talke et al., 2011). Thus, diverse TMTs are more likely than homogeneous groups to be entrepreneurial and enter new markets. However, excessive diversity can be harmful, as control and coordination problems may arise (Sciascia et al., 2013), increasing a TMT’s emotional conflicts and making cognitive integrations in a firm’s internationalization strategy difficult (Jaw & Lin, 2009). For these reasons, as well as the potential positive and negative outcomes (Dahlin, Weingart, & Hinds, 2005; Joshi & Roh, 2009; Kraiczy et al., 2014; Rivas, 2012), TMT diversity is considered a double-edged sword (Ling & Kellermanns, 2010). 8 Family firms possess a unique source of TMT diversity not found in other organizations because of family involvement in management positions (Ling & Kellermanns, 2010). Researching TMTs in family firms is important because a family’s influence can be manifested in different forms (Kraiczy et al., 2014). Family firm-specific diversities can be both advantageous and disadvantageous, because each source of family firm-specific TMT diversity can have a different effect (Kraiczy et al., 2014). Therefore, we must consider these diversities separately to form a better understanding of family-induced TMT diversities (Kraiczy et al., 2014). This paper analyzes the two most important family firm-specific diversities created by family involvement in the TMT (Arzubiaga et al., 2018a; Kraiczy et al., 2014): the family TMT ratio or diversity produced by the proportion of family members involved in the TMT (Cruz & Nordqvist, 2012; Minichilli et al., 2010; Naldi et al., 2007) and the diversity produced by generational involvement in the TMT (Kellermanns & Eddleston, 2006; Minichilli et al., 2010; Sciascia et al., 2013) (see Figure 1). Figure 1: The analysis model 2.2.1. The moderating effect of the family TMT ratio The family TMT ratio is the proportion of family members holding TMT positions (Minichilli et al., 2010). Family members often share a common culture, values, and norms transmitted from their parents and relatives (Chua, Chrisman, & Sharma, 2003), which give them a strong emotional attachment to the business that promotes their level of commitment and involvement with the organization (Minichilli et al., 2010). This higher commitment of family members may make interactions among managers more frequent and direct (Gupta, Smith, & Shalley, 2006), which improves information exchange and learning from other firm members (Arzubiaga et al., 2018a; Liu, Chen, & Tao, 2015). However, family members usually possess a common educational background, expertise, and experience acquired within the family firm Entrepreneurial Orientation Degree of Internationalization Family TMT Ratio Generational Involvement in TMT H1 H2 Controls: - Firm Size - Firm Age - Firm Industry - FDI 15 3.3. Variable measurement All variables were measured following the prior literature. However, no broadly accepted measurement exists to measure the dependent variable, the degree of internationalization (DOI). The ratio of export sales is commonly used in SME research as a representative measure of firms’ internationalization (Calabrò & Mussolino, 2013; Graves & Thomas, 2006; Lu & Beamish, 2001). Exporting is considered the most common entry mode in this firm type since SMEs do not usually have subsidiaries abroad. However, internationalization constitutes more than exports, and more comprehensive measures should be used (Mitter et al., 2014). According to Sullivan (1994), measuring the degree of internationalization using multiple items is recommended. Therefore, the DOI construct in this paper was measured using a combination of two items to overcome the limitations associated with the adoption of a single item. We followed the indications of previous family firm research (Arregle et al., 2012; Zahra, 2003), measuring the degree of internationalization as follows: (1) the ratio of foreign sales to total sales, which refers to the scale of firms’ international sales, and (2) the number of foreign countries in which firms operate, which refers to the scope of foreign activities. The independent variable, EO, can be measured using several scales (Covin & Wales, 2012). The choice of measurement must be based on an understanding of the construct’s theoretical nature (Covin & Wales, 2012). We chose to measure EO using Covin and Slevin’s (1989; 1991) three dimensions, because they comprise a broadly accepted and validated measure (e.g., Arzubiaga et al., 2018b; Casillas et al., 2011; Cruz & Nordqvist, 2012; Sciascia et al., 2013). This measure was initially built based on the meaning of “being entrepreneurial” suggested by theory. EO was thus modeled as a second-order reflective construct, formed by three first-order reflective constructs named innovativeness, proactiveness, and risk-taking (Arzubiaga et al., 2018b; Covin & Wales, 2012). EO is a second-order reflective construct that expresses changes in the underlying latent construct as reflected in changes to the indicators (Arzubiaga et al., 2018b; Jarvis et al., 2003). Therefore, the component variables of innovativeness, proactiveness, and risk-taking were first assessed through their corresponding measures, each with three items, and then these three components were assessed as reflective indicators of the second-order construct of EO (Covin & Wales, 2012). Thus, EO was measured using “Type I” second-order factor specification (i.e., reflective first-order and reflective second-order) (Covin & Wales, 2012). Respondents assessed nine items on an 11-point Likert scale to build all firstand second-order constructs. The CEO’s perspective comprises an accepted approach to 16 operationalizing firm-level EO in the literature, especially when analyzing SMEs (Cruz & Nordqvist, 2012; Wiklund & Shepherd, 2003). This study analyzes two moderating variables. The family TMT ratio is constructed from respondents’ indications of the number of family members holding managerial positions and the total number of TMT members. The family TMT ratio is obtained by dividing the number of family managers by the total number of TMT members (Kraiczy et al., 2014; Minichilli et al., 2010). This ratio can assume values of zero to one. However, generational involvement was measured by the number of generations simultaneously holding TMT positions (Chirico et al., 2011; Kellermanns & Eddleston, 2006; Kraiczy et al., 2014). One, two, or three generations can be simultaneously involved in managerial positions, depending on the generations involved in the TMT. Therefore, this variable can assume values from one to three. The control variables used in this study are firm size, age, industry, and foreign direct investment (FDI). Firm size is a proxy of the amount of resources available for the internationalization process; large firms should have a greater amount of managerial, productive, and financial resources that contribute to increasing their presence in foreign markets (Cerrato & Piva, 2012). As in previous SME studies, size was a continuous variable measured as a logarithmic transformation of the firm’s number of employees (Calabrò & Mussolino, 2013; Cerrato & Piva, 2012; Chen, Hsu, & Chang, 2014; Fernández & Nieto, 2005; 2006). Firm age is a commonly used control variable in internationalization studies (Cerrato & Piva, 2012; Fernández & Nieto, 2006; Mitter et al., 2014). Age is primarily used to control for a firm’s business experience, because firms with more years of performance have gathered more experience and knowledge (Cerrato & Piva, 2012). Firm age is a continuous variable measured as a logarithmic transformation of the years a firm has been in existence (Chen et al., 2014). Firm industry is also assumed to affect the degree of internationalization (Segaro et al., 2014). Firms were categorized into six different industries—chemical, food, steel, electric, furniture, and textiles—and controlled using dummy variables (Segaro et al., 2014; Singla, Veliyath, & George, 2014). The FDI variable is used to control for foreign direct investment or assets that firms control abroad. Previous family firm research studies have used the foreign assets to total assets ratio to measure FDI (Bhaumik et al., 2010; Singla et al., 2017). Although our sample firms are family SMEs and their main activity in foreign markets is exporting, the level of FDI might also influence the degree of internationalization. 17 4. Analyses and results 4.1. Validity and reliability The proposed model has been analyzed using structural equation modeling based on a partial least squares method (PLS). The PLS approach is appropriate for investigating complex relationships with moderating variables (Chin, 1998). The SmartPLS version 2 software was applied in the empirical analyses (Kortmann, 2015). The analysis used a two-stage process: (1) assessment of the measurement model’s reliability and validity and (2) appraisal of the proposed structural model (Acedo & Galán, 2011; Barclay, Higgins, & Thompsom, 1995). Three different tests were conducted to analyze the reliability and validity of the reflective constructs’ measurement scales (Acedo & Galán, 2011; Hulland, 1999; Kortmann, 2015): individual item reliabilities, the convergent validity of measures related to individual constructs, and discriminant validity. First, individual item reliability examines each item’s loading for reflective constructs. The items’ loadings should be greater than 0.7, because the item and construct’s shared variance must be greater than the variance between the construct and its error (Carmines & Zeller, 1979; Hulland, 1999). Table 2 illustrates the results. Note that one item loads below the 0.7 cut-off. Thus, item 6 of the EO included in the proactiveness construct was removed. The model also includes a second-order construct, EO. The structural coefficients, in this case, can be taken as factor loadings (Doll, Xia, & Torkzadeh, 1994). The three structural coefficients of the EO construct in the proposed model are greater than 0.7, and, thus, no problems exist with the second-order constructs. Second, convergent validity is analyzed to determine internal consistency through composite reliability (CR) and Cronbach’s alpha (Kortmann, 2015); both statistics must be greater than 0.7 (Hulland, 1999; Nunnally, 1978). The average variance extracted (AVE) must also be analyzed, and its recommended value is greater than 0.5 (Acedo & Galán, 2011; Fornell & Larcker, 1981). The second-order construct’s AVE was calculated following the work of Hair, Hult, Ringle, and Sarstedt (2014). As Table 2 reveals, convergent validity is fulfilled in the proposed model, as all parameters are above the cut-off points. Third, discriminant validity is evaluated through the AVE (Fornell & Larcker, 1981). A construct must share more variance with its indicators than with other model constructs. Thus, the two constructs’ individual AVEs should be greater than the squared correlation between the two constructs. This condition must be fulfilled in the present model between DOI and EO. 18 Table 3 presents the correlation matrix, with values for the AVE’s square root on the diagonal. Note that the required condition is met. Table 2. Construct reliability and convergent validity Table 3. Correlation matrix and discriminant validity 4.2. Assessment of the structural model Once the measurement model has been verified, the structural model can be assessed by examining the relationships between constructs and the model’s predictive capacity. Various tests are then conducted to validate the overall structural model, because the PLS does not supply a specific index (Chin, 1998; Kortmann, 2015). Following Hair et al. (2014), we used 19 “bootstrapping” with 5,000 subsamples to calculate the proposed relationships’ t-statistics. Table 4 illustrates the results. The direct effect of EO on family firms’ degree of internationalization is positive and statistically significant (β = 0.234; t = 3.728; p < 0.001). The results suggest that firms with higher entrepreneurial orientation are more internationalized. However, results partially support the hypotheses regarding family involvement. The moderating effect produced by the family TMT ratio is negative and significant, as expected (β = -0.183; t = 3.306; p < 0.001); thus, Hypothesis 1 is corroborated. It is suggested that EO’s effect on the level of internationalization is lower in firms with a higher proportion of family members in the managerial team. However, the moderating effect produced by generational involvement is negative and statistically significant (β = -0.145; t = 2.651; p < 0.01) and not positive, as hypothesized. Thus, Hypothesis 2 is rejected. The degree of internationalization’s explanatory power (R2 = 0.298) exceeds the recommended value proposed by Falk and Miller (1992), in which R2 should be greater than 0.1. A Q2 test was conducted to evaluate the constructs’ predictive relevance (Chin, 1998) using a blindfolding procedure (Henseler, Ringle, & Sinkovics, 2009); as the corresponding Q2 values are positive, the hypothesized model has a strong predictive capacity overall (Henseler et al., 2009; Kortmann, 2015). 20 Table 4: Results of structural equation modeling Figure 2: Results of structural equation modeling We separately analyzed the moderating effect generated by firms with only one generation in the TMT to increase our understanding of generational involvement. Thus, we created a dummy variable that assumes a value of one when a firm only has members from one generation, and zero when multiple generations hold managerial positions. The results indicate that firms with members from only one generation produce a positive moderation (β = 0.127; t = 2.067; p < 0.025). Thus, when only one generation manages a firm, EO’s effect on the degree of internationalization improves. Alternatively, as tested in Hypothesis 2, when multiple generations simultaneously hold managerial positions, this generates a negative moderating effect. 4.3. Robustness check To examine our results’ robustness, we estimated structural models with some subsamples of firm characteristics (Gruber, Heinemann, Brettel, & Hunling, 2010). First, we divided the sample by age (older and younger firms), taking the mean as a reference. Firms with 32.9 or more years were classified as older firms, obtaining a subsample of 101 firms. Conversely, firms with less than 32.9 years were classified as younger firms, obtaining a subsample of 90 Entrepreneurial Orientation Degree of Internationalization Family TMT Ratio Generational Involvement in TMT H1: -0.183*** 0.234*** H2: -0.145** R2 = 0.298 21 firms. The results in the older firms subsample indicate that the EO–internationalization relationship is significant (β = 0.265; t = 2.599; p < 0.005), the moderating effect of the family TMT ratio is significant (β = -0.280; t = 3.628; p < 0.001), and the moderating effect of generational involvement is significant (β = -0.232; t = 2.130; p < 0.025). The results in the younger firms subsample indicate that the EO–internationalization relationship is significant (β = 0.225; t = 2.535; p < 0.01), the moderating effect of the family TMT ratio is significant (β = -0.193; t = 2.111; p < 0.025), and the moderating effect of generational involvement is significant (β = -0.187; t = 2.150; p < 0.025). Analyzing the results obtained, which produced only minimal differences in the path coefficients for the two subsamples, we can conclude that the robustness test carried out is consistent. Second, the original sample was divided into two groups by industry (low technology and high technology sectors). Low technology sectors include textile, steel, and furniture industries, and the subsample included 96 firms. High technology sectors include chemistry, electric, and food industries, and the subsample included 95 firms. The results of the subsample of low technology sectors indicate that the EO–internationalization relationship is significant (β = 0.286; t = 2.779; p < 0.005), the moderating effect of the family TMT ratio is significant (β = -0.169; t = 2.255; p < 0.025), and the moderating effect of generational involvement is significant (β = -0.156; t = 1.924; p < 0.05). The results of the subsample of high technology sectors indicate that the EO–internationalization relationship is significant (β = 0.275; t = 3.043; p < 0.005), the moderating effect of the family TMT ratio is significant (β = -0.266; t = 3.357; p < 0.001), and the moderating effect of generational involvement is significant (β = - 0.176; t = 2.132; p < 0.025). Analyzing the results obtained, which also produced minimal differences in the path coefficients for the two subsamples, we can conclude that the robustness test carried out is consistent. To sum up, given that the analysis of different subsamples reveals that subsamples are also significant and with minimal differences in the path coefficients, we can assert that our results are robust. 5. Discussion This study attempts to shed new light on the debate regarding the extent to which family TMT involvement influences the relationship between EO and the internationalization of family SMEs. We analyzed family involvement in and influence on the firm by focusing on two family 22 firm-specific diversities in the TMT (i.e., the family TMT ratio and generational involvement in the TMT). As argued during hypotheses development, the upper echelon theory (Carpenter et al., 2004; Hambrick, 2007; Hambrick & Mason, 1984) supports the idea that firm outcomes can be predicted by TMT composition. Diversely formed TMT groups may possess a wider pool of skills, knowledge, experiences, and values, which may positively influence business outcomes such as internationalization. Our results partially support the proposed hypotheses and reveal unexpected and interesting findings, extending our understating of family firms’ entrepreneurial behavior and internationalization. Our results indicate that EO plays a significant role in explaining family firms’ degree of internationalization, as previous studies indicated (Calabrò et al., 2017; Hernández-Perlines & Mancebo-Lozano 2016; Jantunen et al., 2005; Javalgi & Todd, 2011; Liu et al., 2011; RipollésMeliá et al., 2007). Furthermore, family firm-specific TMT diversities shape the effects of the family firm’s entrepreneurial orientation on the degree of internationalization. The family TMT ratio negatively moderates the relationship between EO and degree of internationalization, as the relationship is weaker when a higher proportion of family members hold positions in the TMT. Thus, an excessive level of family involvement in the TMT could limit diversity in knowledge and multiple perspectives about internationalization strategies. The family may not possess sufficient human resources for TMT positions, and, therefore, non-family members’ participation in the TMT, bringing different expertise and backgrounds, might help family firms overcome this shortage (Calabrò et al., 2013; Segaro et al., 2014). Thus, SMEs that wish to internationalize may need to consider the inclusion of external managers’ skills and background to enhance EO’s effect on internationalization. Higher generational involvement was expected to positively moderate EO’s effect on internationalization. Generational involvement increases the knowledge and diversity of experiences in the TMT (Kellermanns & Eddleston, 2006; Ling & Kellermanns, 2010). However, the opposite effect was revealed; a higher number of generations involved generated a negative moderating effect. Consequently, generational involvement does not always act as a catalyst in the relationship between EO and degree of internationalization. Several reasons may explain this situation. One explanation is that vision, interests, and expectations of business outcomes can differ between generations, leading to problems in control and power (Bammens, Voordeckers, & Van Gils, 2008; Ling & Kellermanns, 2010; Sonfield & Lussier, 2004). Conflicts between generations can hamper the constructive debate promoted by a diverse TMT, hindering the assessment, acceptance, and incorporation of others’ ideas 23 (Sciascia et al., 2013). Jaw and Lin (2009) asserted that a heterogeneous TMT may be more creative in solving problems related to foreign markets, but when multiple generations are involved in the decision-making team, this can impede group cooperation and cognitive integration toward a firm’s global strategy. Thus, Chirico et al. (2011) indicated that persistent conflict affects family firms’ entrepreneurial efforts, and, consequently, a participative strategy is crucial to ensure value-creating co-alignment among individuals across generations. Another explanation for generational involvement’s negative effect may rely on managers’ different visions and perspectives resulting from their age. Firms with more than one generation involved in management positions may not have gone through a succession process yet, and, thus, senior managers in these firms are likely to be older. Conversely, firms with only one generation might have already gone through a succession process, and, thus, it is expected that these firms’ managers will be younger. Previous research indicates that managers’ age influences strategic decision-making (Herrmann & Datta, 2005; Tihanyi et al. 2000). Age can indicate the extent of experience, but it can also signal managers’ propensity for risk-taking (Herrmann & Datta, 2005). TMT members’ age has been negatively associated with high-risk decision-making and with the capacity to analyze new information (Suárez-Ortega & ÁlamoVera, 2005). However, younger directors are more likely to adopt riskier strategies (Hambrick & Mason, 1984). Empirical evidence suggests that a lower age of managers is positively related to internationalization strategies (Fernández-Ortíz & Fuentes-Lombardo, 2009; Herrmann & Datta, 2005; Tihanyi et al., 2000). Thus, it is expected that firms with managers from different generations will have older managers that are likely to be more averse to implement risky strategies such as internationalization. Therefore, generational involvement could negatively affect the relationship between EO and a firm’s internationalization. 6. Contributions, limitations, and future research This study contributes to the literature on family firms and management practices in several ways. First, we contribute theoretically to increasing our knowledge of the upper echelons theory by increasing our understanding of two family firm-specific TMT diversities induced by family influence. Although the TMT represents a good way of testing family involvement in the business, previous research has largely overlooked the influence of family involvement in the TMT (Kraiczy et al., 2014; Sciascia et al., 2013; Segaro et al., 2014). We have focused on the family TMT ratio and generational involvement as two important TMT diversities found 24 only in family firms. These two family firm-specific diversities were previously used to analyze innovation issues in family firms (Arzubiaga et al., 2018a; Kraiczy et al., 2014), but not in the relationship between EO and internationalization. Accordingly, we have contributed to enhancing our knowledge of upper echelons theory, distinguishing between the benefits and drawbacks of family firm-specific TMT diversities (De Massis et al., 2013). The family’s influence plays an important role in strategic decision-making and shapes the existing positive relationship between EO and the firm’s degree of internationalization. The diverse formation of TMTs with different expertise, experience, and knowledge may not result always beneficial, as in the case of generational involvement. The expected positive effect was negative. This is an example of how diversity in the TMT can be a double-edged sword. More research is needed to address generational involvement’s different implications in strategic decision-making, such as internationalization. Second, our study contributes to the international business and entrepreneurship literature by increasing knowledge in this joint research field. There has been very little research linking firms’ entrepreneurial behaviors and the internationalization of family firms (Calabrò et al., 2017; Hernández-Perlines et al., 2016). However, these studies do not analyze how the effect of entrepreneurial orientation in internationalization may vary because of family influence. Due to the overlap of owners and managers in family SMEs, these organizations represent an interesting setting, because they can behave differently when making an entrepreneurial strategic decision, such as internationalization. Therefore, future research should extend our knowledge of the entrepreneurial internationalization of family firms. Third, our research contributes to the emergent research stream of family firm heterogeneity (Chua et al., 2012; Miller et al., 2013). Family involvement in the TMT can be an important determinant explaining the heterogeneous behaviors of family firms. Our study adds evidence of the significant role of the TMT in important governance dimensions of family firms, which have been previously identified as the most important factors in family firms’ heterogeneity (Li & Daspit, 2016; Pittino et al., 2018). Fourth, our study contributes to the research on family SMEs. Family SMEs are the most common type of business, but their process of internationalization has not been deeply developed despite its importance (Mitter et al., 2014; Segaro et al., 2014). We test our hypotheses with a sample of firms from Spain, a country where almost 90% of firms are considered family firms (Casillas et al., 2016). It should also be interesting to analyze the 31 Ling, Y., & Kellermanns, F. W. (2010). The effects of family firm specific sources of TMT diversity: The moderating role of information exchange frequency. Journal of Management Studies, 47(2), 322-344. Liu, J., Chen, J., & Tao, Y. (2015). Innovation performance in new product development teams in China’s technology ventures: The role of behavioral integration dimensions and collective efficacy. Journal of Product Innovation Management, 32(1), 29-44. Liu, Y., Li, Y., & Xue, J. (2011). Ownership, strategic orientation and internationalization in emerging markets. Journal of World Business, 46 (3), 381-393. Lu, J. W., & Beamish, P. W. (2001). The internationalization and performance of SMEs. Strategic Management Journal, 22(6-7), 565-586. Lubatkin, M., Simsek, Z., Ling, Y., & Veiga, J. F. (2006). Ambidexterity and performance in small-to medium-sized firms: The pivotal role of top management team behavioral integration. Journal of Management, 32(5), 646–672. Lumpkin, G. T., Brigham, K., & Moss, T. (2010). Long-term orientation: implications for the entrepreneurial orientation and performance of family business. Entrepreneurship and Regional Development, 22(3), 241-264. McNulty, T., & Pettigrew, A. (1999). Strategists on the board. Organization Studies, 20(1), 4774. Michie, S., Dooley, R. S., & Fryxell, G. E. (2006). Unified diversity in top-level teams: Enhancing collaboration and quality in strategic decision making. International Journal of Organizational Analysis, 14(2), 130-149. Miller, D., Le Breton-Miller, I, & Lester, R. H. (2013). Family firm governance, strategic conformity, and performance : Institutional vs. strategic perspectives. Organization Science, 24(1), 189-209. Miller, D., Minichilli, A., & Corbetta, G. (2013). Is family leadership always beneficial? Strategic Management Journal, 34(5), 553-571. Minichilli, A., Corbetta, G., & MacMillan, I. C. (2010). Top management teams in familycontrolled companies: "familiness", "faultlines", and their impact on financial performance. Journal of Management Studies, 47(2), 205-222. Mitter, C., Duller, C., Feldbauer-Durstmüller, B., & Kraus, S. (2014). Internationalization of family firms: The effect of ownership and governance. Review of Managerial Science, 8(1), 1-28. Moreno, A. M., & Casillas, J. C. (2008). Entrepreneurial Orientation and Growth of SMEs: A Causal Model. Entrepreneurship Theory and Practice, 32(3), 507-528. Muñoz-Bullón, F., & Sánchez-Bueno, M. J. (2012). Do family ties shape the performance consequences of diversification? Evidence from the European Union. Journal of World Business, 47(3), 469-77. Naldi, L., Nordqvist, M., Sjöberg, K., & Wiklund, J. (2007). Entrepreneurial orientation, risk taking, and performance in family firms. Family Business Review, 20(1), 33-47. Nielsen, S. (2010). Top management tem internationalization and firm performance. The mediating role of foreign market entry. Management International Review, 50(2), 185206. Nielsen, B. B., & Nielsen, S. (2011). The role of top management team international orientation in international strategic decision-making: The choice of foreign entry mode. Journal of World Business, 46(2), 185-193. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. 32 Pittino, D., Barroso-Martínez, A., Chirico, F., & Sanguino-Galván, R. (2018). Psychological ownership, knowledge sharing and entrepreneurial orientation in family firms: The moderating role of governance heterogeneity. Journal of Business Research, 84, 312-326. Podsakoff P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12(4), 531–544. Ramón-Llorens, M. C., García-Meca, E., & Duréndez, A. (2017). Influence of CEO characteristics in family firms internationalization. International Business Review, 26(4), 786-799. Ripollés-Meliá, M., Menguzzato-Boulard, M., & Sánchez-Peinado, L. (2007). Entrepreneurial orientation and international commitment. Journal of International Entrepreneurship, 5(3-4), 65-83. Rivas, J. L. (2012). Diversity & internationalization: The case of boards and TMT’s. International Business Review, 21(1), 1-12. Sánchez-Marín, G., & Baixauli-Soler, J. S. (2015). TMT pay dispersion and firm performance: The moderating role of organizational governance effectiveness. Journal of Management & Organization, 21(4), 436–459. Sciascia, S., Mazzola, P., & Chirico, F. (2013). Generational involvement in the top management team of family firms: Exploring nonlinear effects on entrepreneurial orientation. Entrepreneurship Theory and Practice, 37(1), 69-85. Segaro, E. (2012). Internationalization of family SMEs: The impact of ownership, governance, and top management team. Journal of Management and Governance, 16(1), 147-169. Segaro, E. L., Larimo, J., & Jones, M. V. (2014). Internationalisation of family small and medium sized enterprises: The role of stewardship orientation, family commitment culture and top management team. International Business Review, 23(2), 381-95. Sharma, P. (2004). An overview of the field of family business studies: Current status and directions for the future. Family Business Review, 17(1), 1-36. Siebels, J. F., & Knyphausen, D. (2012). A review of theory in family business research: The implications for corporate governance. International Journal of Management Reviews, 14(3), 280-304. Singla, C., George, R., & Veliyath, R. (2017). Ownership structure and internationalization of Indian firms. Journal of Business Research, 81, 130-143. Singla, C., Veliyath, R., & George, R. (2014). Family firms and internationalizationgovernance relationships: Evidence of secondary agency issues. Strategic Management Journal, 35(4), 606-616. Sonfield, M. C., & Lussier, R. N. (2004). First-, second-, and third-generation family firms: A comparison. Family Business Review, 17(3), 189-201. Speckbacher, G., & Wentges, P. (2012). The impact of family control on the use of performance measures in strategic target setting and incentive compensation: A research note. Management Accounting Research, 23(1), 34-46. Steier, L. (2003). Variants of agency contracts in family financed ventures as a continuum of familial altruistic and market rationalities. Journal of Business Venturing, 18(5), 597-618. 33 Suárez-Ortega, S. M., & Álamo-Vera, F. R. (2005). SMEs’ internationalization: firms and managerial factors. International Journal of Entrepreneurial Behavior & Research, 11(4), 258-279. Sullivan, D. (1994). Measuring the degree of internationalization of a firm. Journal of International Business Studies, 25(2), 325-342. Talke, K., Salomo, S., & Kock, A. (2011). Top management team diversity and strategic innovation orientation: The relationship and consequences for innovativeness and performance. Journal of Product Innovation Management, 28(6), 819-832. Tihanyi, L., Ellstrand, A. E., Daily, C. M., & Dalton, D. R. (2000). Composition of the top management team and firm international diversification. Journal of Management, 26(6), 1157-1177. Vandekerkhof, P., Steijvers, T., Hendriks, W., & Voordeckers, W. (2015). The effect of organizational characteristics on the appointment of nonfamily managers in private family firms: The moderating role of socioemotional wealth. Family Business Review, 28(2), 104122. Webb, J. W., Ketchen Jr. D. J., & Ireland, R. D. (2010). Strategic entrepreneurship within family-controlled firms: Opportunities and challenges. Journal of Family Business Strategy, 1(2), 67-77. Westphal, J. D., & Zajac, E. J. (1998). The symbolic management of stockholders: Corporate governance reforms and shareholder reactions. Administrative Science Quarterly, 127153. Wiklund, J., & Shepherd, D. (2003). Knowledge-based resources, entrepreneurial orientation, and the performance of small and medium-sized businesses. Strategic Management Journal, 24(13), 1307-1314. Zahra, S. A. (2003). International expansion of U.S. manufacturing family businesses: The effect of ownership and involvement. Journal of Business Venturing, 18(4), 495-512. 34 Appendix A. Harman’s Single-Factor Test Appendix B. Common Method Bias Analysis