Socioemotional wealth, entrepreneurial orientation and international performance of family firms

Abstract This paper analyses the relationships between the socioemotional wealth, entrepreneurial orientation and international performance of family firms. This research is pioneering in that it seeks to explain the international performance of family firms from the non-economic perspective of entrepreneurial orientation determined by socioemotional wealth. Second generation structural equation modelling (PLS-SEM) using SmartPLS 3.2.8 software was applied to data from 106 Spanish family firms. The study shows that considering socioemotional wealth substantially improves the capacity of entrepreneurial orientation to explain variation in the international performance of family firms. When only entrepreneurial orientation is included in the model, the explained variance of international performance is 34.2%. However, when socioemotional wealth is included in the model as an antecedent of international performance, the explained variance increases to 42.6%.

The family firm has numerous definitions. It is therefore difficult to reach a consensus on how best to provide a generally accepted definition. Nevertheless, two features are common to most definitions of the family firm (Franco & Prata, 2019). The first is ownership of capital. In family firms, most shares belong to one or more family members. The second is management. In family firms, several members of the family participate in the management of the business (Kallmuenzer, Hora, & Peters, 2018). Analysing family firms is a challenge because of the wide variety of family firms that exist ( Cennamo, Berrone, Cruz, & Gomez-Mejia, 2012;Chrisman & Patel, 2012;Llanos-Contreras, & Santos, 2018 ) as a result of varying levels of family involvement in the business (Samara & Berbegal-Mirabent, 2018).
In this study, our goal is to link the aspects mentioned earlier, namely international performance, entrepreneurial orientation and socioemotional wealth. The choice of these variables owes to their strong correlation, which has been explored in previous studies. For example, Hern andez-Perlines (2018) analysed the positive influence of entrepreneurial orientation on the international performance of family firms. Similarly, Hern andez-Perlines, Moreno-Garc ıa, and Y añez-Araque (2019) analysed the positive effect of socioemotional wealth on entrepreneurial orientation. The opportunity in this study is provided by the analysis of how socioemotional wealth is capable of explaining the international performance of family firms through its effect on entrepreneurial orientation. The research question addressed by this study is whether the effect of entrepreneurial orientation on the international performance of family firms differs when socioemotional wealth is considered as an antecedent of entrepreneurial orientation. This question can be broken down into three parts.
First, why consider international performance? The answer is that although research on internationalisation has advanced considerably in recent years, there are still unanswered questions that arise as a result of the increasingly global and competitive environment where companies operate (Werner, 2002).
Second, why study entrepreneurial orientation? Scholars such as Eddleston, Kellermanns, and Zellweger (2012) have reported that one of the main problems with family firms is their survival rate. The question is, how can their survival rate be improved? Kellermanns and Eddleston (2006) claimed that the survival of family firms could be improved through entrepreneurship. There are different proposals to analyse entrepreneurship in family firms, although the most common one is to study entrepreneurship through entrepreneurial orientation (Cruz & Nordqvist, 2012;Lumpkin, Brigham, & Moss, 2010;Naldi, Nordqvist, Sj€ oberg, & Wiklund, 2007;Zahra, 2005). In addition, family firms provide a context that can help us develop a better understanding of entrepreneurial orientation (Hern andez-Linares & L opez-Fern andez, 2018). Accordingly, the number of studies that analyse entrepreneurial orientation in the family business is increasing (Lee & Chu, 2017).
Third, why analyse socioemotional wealth? The justification for considering socioemotional wealth is that it is a key factor in family firm entrepreneurship (Kellermanns & Eddleston, 2006), where it is necessary to consider the interaction between the business and the family (Franco & Prata, 2019). Socioemotional wealth offers a new research perspective on non-economic aspects of family firms (G omez-Mej ıa, Cruz, Berrone, & De Castro, 2011), and, in recent years, it has become a relevant factor in explaining the behaviour of family firms Chua, Chrisman, & De Massis, 2015;G omez-Mej ıa et al., 2007;Mart ınez-Romero & Rojo-Ram ırez, 2016;Miller & Le Breton-Miller, 2014). The socioemotional wealth approach suggests that family firms have non-economic objectives  Mart ınez-Romero & Rojo-Ram ırez, 2017) that affect their purely economic objectives (Chrisman & Patel, 2012). From a socioemotional wealth perspective, it is suggested that family firms make decisions to protect their socioemotional wealth, even if these decisions entail financial losses . Therefore, socioemotional wealth offers a comprehensive theoretical framework that enables analysis of different aspects of the family firm (Kabbach de Castro, Crespi-Cladera, & Aguilera, 2016). Socioemotional wealth also influences the entrepreneurial orientation of the family firm by helping the family achieve its non-economic goals of improving its reputation, guaranteeing the employment of family members and securing family control through generational renewal (Alonso-Dos-Santos & Llanos-Contreras, 2019; G omez-Mej ıa et al., 2011). This study helps close the gap in the literature left by a lack of research that uses the non-economic perspective to explain the entrepreneurial orientation of family firms.

Theory and hypotheses
The increasing globalisation and competitiveness of the market make organisational performance a key concern for firms (Ferreira, Fernandes, & Ortiz, 2018). In the case of family firms, performance, in its different forms, is still a key research area (Franco & Prata, 2019). The primary goal of this study is to determine how best to explain the international performance of family firms. To pursue this goal, we use two concepts that are strongly correlated with this performance, namely entrepreneurial orientation and socioemotional wealth.
Two theoretical frameworks support this study. The first is the theory of dynamic capabilities (Makadok, 2001;Prahalad & Hamel, 1990;Teece, Pisano, & Shuen, 1997). This framework is relevant because entrepreneurial orientation is a capability that helps companies adapt to changes to remain competitive Jaskiewicz et al., 2015). The second framework that supports this study is that of socioemotional wealth, which entails the consideration of non-economic aspects in relation to the family firm (Chrisman & Patel, 2012;G omez-Mej ıa et al., 2011). According to this theoretical approach, the decision-adoption process in the family firm revolves around protecting its socioemotional wealth (Berrone, Cruz, Gomez-Mejia, & Larraza-Kintana, 2010;Cennamo, Berrone, Cruz, & Gomez-Mejia, 2012;Glover & Reay, 2015;G omez-Mej ıa et al., 2007), even if doing so incurs certain financial risks . In this study, we adopt a non-economic view of entrepreneurial orientation. We consider that socioemotional wealth determines entrepreneurship in the family firm by helping the family to achieve its noneconomic objectives such as improving its reputation, guaranteeing the employment of family members and ensuring family control of the firm through generational renewal (G omez-Mej ıa et al., 2011). This perspective helps us understand the process of strategic management of family firms, which is characterised by the interaction between the business and the family (Kallmuenzer, Hora, et al., 2018).

Socioemotional wealth as an antecedent of entrepreneurial orientation
The number of studies that examine the role of emotions in business performance is growing (Rizvi & Oney, 2018). In family firms, the family, ownership and management must live side by side, which makes for a unique decision-making process (Kallmuenzer, Hora, et al., 2018). The behaviour of the family firm is determined by non-economic family goals that create socioemotional wealth (Chrisman & Patel, 2012) and emotional value when focused on the family (Zellweger, Kellermanns, Chrisman, & Chua, 2012). In family firms, the members of the family feel emotional attachment to other family members, strongly identifying with the family firm, which is often seen as an extension of the family (Kallmuenzer, Hora, et al., 2018).
The approach of socioemotional wealth offers a way of analysing non-economic family goals. This theoretical alternative is based on the model of behavioural agency (Chrisman & Patel, 2012) and is suited to the study of the family firm because it integrates family, personal and organisational goals (Llanos-Contreras & Santos, 2018), which occasionally clash with one another (Franco & Prata, 2019).
From the first perspective, Cennamo et al. (2012) proposed that socioemotional wealth allows family firms to adopt a policy of proactive participation of the interested parties, given that the decision-adoption process is determined by identification of the dimensions of socioemotional wealth. From the second perspective, Schepers et al. (2014) analysed the moderating effect of socioemotional wealth on the relationship between the performance and entrepreneurial orientation of family firms, concluding that socioemotional wealth limits this relationship. However, family relationships and innovative capacity provide a strong competitive advantage (Eddleston, Kellermanns, & Sarathy, 2008).
The academic literature offers intense debate over how to measure socioemotional wealth. In early studies, socioemotional wealth was measured as a proxy of ownership and family management (e.g. Berrone et al., 2010;G omez-Mej ıa et al., 2010;Vandemaele & Vancauteren, 2015;Zellweger, Nason, & Nordqvist, 2012). Other studies have tried to capture socioemotional wealth in terms of the strategic orientation of small and medium-sized enterprises (e.g. Goel et al., 2013;Schepers et al., 2014). However, the review of socioemotional wealth by Berrone et al. (2012) is the most widely cited because of its links with other theoretical approaches. Berrone et al. (2012) proposed five dimensions to measure socioemotional wealth: family control and influence, identification of family members with the company, binding social ties, emotional attachment of family members and renewal of family bonds with the company through dynastic succession. These five dimensions are known as FIBER. This socioemotional wealth measurement model was used by Cennamo et al. (2012) to analyse how socioemotional wealth allows family companies to adopt actions for the proactive participation of interested parties. Debicki, Kellermanns, Chrisman, Pearson, and Spencer (2016) reformulated this socioemotional wealth measurement model by evaluating the importance that the owners and managers of family firms attach to the elements of the model. Recent studies have scrutinised the FIBER model, incorporating three aspects of great importance for the family business: the importance of the family in decision making, family continuity and family enrichment (Alonso-Dos-Santos & Llanos-Contreras, 2019).
Control and influence by the family allows family firms to implement proactive strategic decisions that entail innovation and even some risk (Habbershon & Pistrui, 2002;Kellermanns et al., 2012). In this sense, family participation enhances the positive effect of innovative capacity on growth (Casillas & Moreno, 2010) and promotion of the entrepreneurial spirit (Zahra, 2005). Family members tend to seek company behaviour that is prone to improving the social status of the family firm (Davis, Pitts, & Cormier, 2000), and their identification with the company improves performance (Anderson & Reeb, 2003;Zellweger & Nason, 2008). The strong ties between family members affect the acknowledgment of entrepreneurial opportunities (Jack, 2005) and resource accumulation (Khayesi, George, & Antonakis, 2014).
Emotions are another distinctive attribute of family firms (Astrachan & Jaskiewicz, 2008;Berrone et al., 2012;Zellweger & Astrachan, 2008) that could play an important role in company behaviour (Goss, 2005) because of their impact on decision making (Astrachan & Jaskiewicz, 2008). Baron (2008) reported that affect (feelings and emotions) could increase creativity and opportunity recognition in risky environments. Succession is one of the biggest challenges facing family firms (Le Breton -Miller, Miller, & Steier, 2004) because, in these companies, strategic decisions are long-term oriented (Kallmuenzer, Hora, et al., 2018;Miller & Le Breton-Miller, 2005;Miller, Le Breton-Miller, & Scholnick, 2008). Succession in family firms is determined by the degree of involvement of descendants and the way they perceive the rewards obtained by their predecessors (Wang, Wang, & Chen, 2018). Eddleston et al. (2012) showed that long-term orientation is positively related to entrepreneurial spirit. Delmas and Gergaud (2014) reported that family companies with transgenerational intent tend to adopt innovative practices, improving entrepreneurship (Zahra et al., 2004). Mart ınez-Alonso, Mart ınez-Romero, and Rojo-Ram ırez (2018) reported that socioemotional wealth positively affects the ability of family businesses to innovate. Finally, entrepreneurial orientation is affected by economic and non-economic factors ( Pejic Bach et al., 2018) , including socioemotional wealth (Hern andez-Perlines et al., 2019). Jaskiewicz et al. (2015) affirmed that socioemotional wealth is one of the most relevant factors to explain entrepreneurship in family firms. Based on these findings, the following research hypothesis is proposed: H 1 : Socioemotional wealth positively affects the entrepreneurial orientation of family firms.

Entrepreneurial orientation as an antecedent of international performance in family firms
Although research on internationalisation has advanced considerably in recent years, answering the question that arises as a result of the increasingly global and competitive environment where companies operate remains a challenge (Werner, 2002). Internationalisation allows companies to grow and survive in the long term (Alayo, Maseda, Iturralde, & Arzubiaga, 2019). Accordingly, entrepreneurial orientation can be conceived as a powerful construct that explains how companies face an everchanging environment (Hern andez-Linares & L opez-Fern andez, 2018), representing a key element of their internationalisation process (Alayo et al., 2019;Javalgi & Todd, 2011) and a pre-requisite for survival in highly competitive environments (Jaskiewicz et al., 2015). Entrepreneurial orientation is a recurring, cross-cutting concept that can explain internationalisation (Baier-Fuentes, Hormiga, Miravitlles, & Blanco-Mesa, 2019). The entrepreneurial orientation and internationalisation of family firms is determined by the interaction between the firm and the family (Kallmuenzer, Hora, et al., 2018). Entrepreneurial orientation enables identification, evaluation and exploitation of different market opportunities (Banalieva, Puffer, McCarthy, & Vaiman, 2018;Ferreira et al., 2018), even if the outcome is uncertain (Banalieva et al., 2018).
Studies have shown the existence of a positive relationship between entrepreneurial orientation and performance (Barringer & Bluedorn, 1999;Covin & Slevin, 1989;Ferreira et al., 2018;Miller, 1983;Wiklund, 1999;Wiklund & Shepherd, 2005;Zahra, 1991;Zahra & Covin, 1995). Therefore, entrepreneurial orientation is a valuable predictor of business success (Krauss, Frese, Friedrich, & Unger, 2005). The management literature offers different approaches to studying internationalisation. In recent years, the entrepreneurship approach has been an emerging force with a high explanatory power regarding the value creation processes of companies that do business beyond their domestic borders (Joardar & Wu, 2011;Jones & Coviello, 2005;Weerawardena, Mort, Liesch, & Knight, 2007). This focus gives rise to the concept of international entrepreneurial orientation, which offers a fresh, dynamic way to explain why companies internationalise (e.g. Freeman & Cavusgil, 2007;Sundqvist, Kyl€ aheiko, Kuivalainen, & Cadogan, 2012). From this approach, many authors have analysed the influence of entrepreneurial orientation on the international performance of the company by considering that this internationalisation is in itself an act of entrepreneurship

Method
Following the literature review and the presentation of the corresponding hypotheses, we now present the research method.

Data
Data were obtained from a questionnaire sent by e-mail using Limesurvey v.2.5 to the highest-ranking executives of a sample of firms drawn from the records of the Spanish Family Firm Institute. The chosen firms were selected randomly. The questionnaire consisted of Likert-type items rated on a scale ranging from 1 to 5. The sample comprised 1,045 companies associated with the Spanish Family Firm Institute. In total, 106 responses were obtained, representing a response rate of 10.14%. The field work was performed between April and June 2017. Table 1 gives details of the sample.
We also performed retrospective analysis of the statistical power of the sample using Cohen's (1992) test in G Ã Power 3.1.9.2 software (Faul, Erdfelder, Buchner, & Lang, 2009). The sample of family firms for this study had a statistical power of 0.9348, which was above the limit of 0.80 established by Cohen (1992). This value means that significant relationships may be identified in the data (Sarstedt & Mooi, 2019).

Socioemotional wealth
In this study, socioemotional wealth was measured using 27 items divided into the following dimensions as per the indications of Berrone et al. (2012): family control and influence (six items), identification of family members with the company (six items), binding social ties (five items), emotional attachment of family members (six items) and renewal of family bonds through dynastic succession (four items). Socioemotional wealth was conceived using a multidimensional approach, in line with the studies by Filser, De Massis, Gast, Kraus, and Niemand (2018) and Gast et al. (2018), bringing together these multiple dimensions to present socioemotional wealth as a whole (Fitz-Koch & Nordqvist, 2017;Li & Daspit, 2016). Thus, socioemotional wealth was operationalised as a second-order composite of type b.

International performance
We measured international performance using a multi-item scale based on export intensity, which has been used as a measure of international performance by authors such as Zahra, Neubaum, and Huse (1997) and Morgan, Kaleka, and Katsikeas (2004). We also considered perceived satisfaction with export performance, which has been used by authors such as Balabanis and Katsikea (2003), Cavusgil and Zou (1994), Dimitratos et al. (2004) and Zahra et al. (1997). Both variables were measured on 5-point Likert scales. The third item used to measure international performance referred to export performance and has been used by authors such as and Ibeh (2003), Morgan et al. (2004) and Zahra et al. (1997).

Control variables
Size (number of employees), age (number of years since establishment) and the main sector of the family firm were used as control variables. These control variables have appeared on a recurring basis in studies of family businesses (Chrisman, Chua, & Sharma, 2005).

Results
To analyse the results and test the hypotheses, we performed partial least squares structural equation modelling (PLS-SEM) using SmartPLS 3.2.8 software (Ringle, Wende, & Becker, 2015). We used PLS-SEM because it enabled estimation of a complex model with second-order composites (Hair, Risher, Sarstedt, & Ringle, 2018). We were therefore able to deduce certain managerial implications (Hair, Sarstedt, & Ringle, 2019), which, in this case, relate to family firm internationalisation (Richter, Sinkovics, Ringle, & Schl€ agel, 2016). Data were obtained from a questionnaire sent by e-mail to the CEOs of family firms associated with the Spanish Family Firm Institute. The questionnaire was completed between April and June 2017. The process yielded useable data from 106 family companies.
To secure that the proposed measurement scales were valid and reliable, the two steps proposed by Barclay, Higgins, and Thompson (1995) and Hair Jr, Sarstedt, Ringle, and Gudergan (2017) were followed. The first consisted of assessing the measurement model, and the second consisted of assessing the structural model.

Assessment of the measurement model
Following recommendations by Rold an and S anchez- Franco (2012) and Hair Jr et al. (2017), the first step was to analyse the values of composite reliability. Fornell and Larcker (1981) recommend values greater than 0.7 for composite reliability (see Table 2). The values observed in this analysis may be described as good, according to Hair et al. (2018), because they were between 0.7 and 0.9. In addition, they did not present problems of redundancy because in no case did they exceed 0.95 (Diamantopoulos, Sarstedt, Fuchs, Wilczynski, & Kaiser, 2012;Drolet & Morrison, 2001).
The second step was to calculate Cronbach's alpha. Fornell and Larcker (1981) recommend Cronbach's alpha values greater than 0.7. As Table 2 shows, the Cronbach's alpha values for this study exceeded this recommended value. Additionally, we calculated a reliability measure that lies between the two previous extreme values (composite reliability and Cronbach's alpha). This measure is rho A, proposed by Dijkstra and Henseler (2015), which should be greater than 0.7 (Dijkstra & Henseler, 2015) and should lie between the values of composite reliability and Cronbach's Alpha (Hair et al., 2018). The values for this study met these criteria (see Table 2).
The third step, to assess convergent validity, was to calculate the average variance extracted (AVE). Fornell and Larcker (1981) recommend a value greater than 0.5 for the AVE. In our case, this condition was met (see Table 2).
The fourth step was to check discriminant validity by confirming that the correlations between each pair of constructs did not exceed the value of the square root of the AVE of each construct and by examining the Heterotrait-Monotrait ratio (HTMT) for composites of type a (see Table 2). For discriminant validity to hold, HTMT values must be less than 0.85 (Henseler, Ringle, & Sarstedt, 2015). As Table 2 shows, the conditions for discriminant validity were met because the values did not exceed the established threshold.

Structural model analysis
After confirming the convergent and discriminant validity of the measurement model, the hypotheses were tested by observing the path coefficient values and their significance (applying the bootstrapping procedure of 5,000 sub-samples). The results confirm that socioemotional wealth is a positive antecedent of entrepreneurial orientation, with socioemotional wealth explaining 56.3% of the variance of entrepreneurial orientation. Therefore, the results show that the explanatory power of socioemotional wealth to explain entrepreneurial orientation is moderate (Hair, Ringle, & Sarstedt, 2011). This finding implies that if family firms want to improve their entrepreneurial orientation, they must enhance their socioemotional wealth (see Table 3 and Figure 2). Entrepreneurial orientation in turn exerts a positive influence on the international performance of family firms, explaining 42.6% of the variance. This result implies that family firms that wish to improve their international performance should consider their entrepreneurial orientation.
No control variable had a significant influence on international performance. All path coefficients were less than 0.2 and were non-significant (see Table 4).
To complete the structural model analysis, we calculated the goodness of fit using the standardised root mean square residual (SRMR). The proposed model had an SRMR value of 0.071, which is lower than the maximum recommended by Henseler et al. (2015) of 0.085. Therefore, the model had a good fit to the data (Henseler, Hubona, & Ray, 2016).

Conclusions
The first conclusion relates to the composites used in the study and is therefore methodological. The use of PLS-SEM showed that socioemotional wealth, entrepreneurial orientation and international performance have suitable reliability and validity values (Henseler et al., 2016). Thus, all three composites used in this research were correctly measured as second-order composites. These results are consistent with previous studies (Hern andez-Perlines, 2018;Hern andez-Perlines et al., 2019;Llanos-Contreras & Santos, 2018).
The second conclusion is that entrepreneurial orientation has a positive significant effect on the international performance of family firms. The results show that entrepreneurial orientation explains 42.6% of the variance of international performance. Therefore, family firm internationalisation is determined by innovation, proactiveness and risk-taking. The results indicate that these three aspects together encourage family firms to internationalise (Hern andez-Perlines, 2018).
The third conclusion is that socioemotional wealth positively and significantly affects the entrepreneurial orientation of family firms, explaining 56.3% of its variance. The results thereby corroborate Berrone et al. (2012) FIBER model and its influence on the multidimensional composite of entrepreneurial orientation (Hern andez-Perlines et al., 2019).
These findings indicate that socioemotional wealth is an antecedent of entrepreneurial orientation and that this entrepreneurial orientation in turn influences the international performance of family firms. If the effect of socioemotional wealth were not considered, then entrepreneurial orientation would explain only 34.2% of the international performance of family firms. In short, for family firms to improve their internationalisation through entrepreneurial orientation, they need socioemotional wealth. This study shows the indirect effect of socioemotional wealth on the international performance of family firms.
The first major limitation of this study is the use of a single informant for data collection. We tried to overcome this limitation by following the recommendations by Rong andWilkinson (2011), Woodside (2013) and Woodside, Prentice, and Larsen (2015). Accordingly, we contacted the highest-ranking executive of each firm (Dal Zotto & Van Kranenburg, 2008). Also, following Torchiano, Tomassetti, Ricca, Tiso, and Reggio's (2013) suggestions, a computerised process was used to send the surveys. In the e-mail sent to respondents, a cover letter was included explaining the study's aims, soliciting participation and providing a contact e-mail address. Also, a reminder was sent to recipients who had not completed the questionnaire. The second limitation relates to the sample, which comprised companies associated with the Spanish Family Firm Institute. To overcome this limitation, we recommend using other databases such as SABI (Iberian Balance Sheets Analysis).
Regarding future research, we would underscore the need to perform longitudinal studies to analyse these effects over time (Cennamo et al., 2012). We encourage studies that compare results across countries using the same measurement scales to check whether differences arise depending on the context of the analysis. Another line of research in the future is to analyse the influence of human capital as a long-term generator of competitive advantage (Vargas, Lloria, & Roig-Dob on, 2016). It would also be of interest to analyse the effect of the interaction of different structures of goals and corporate governance (Chrisman & Patel, 2012).
From a methodological perspective, we propose conducting studies that combine qualitative and quantitative techniques to improve our understanding of the proposed models (Llanos-Contreras & Santos, 2018). It would also be of interest to perform studies in the future that consider the individual characteristics and personality traits of the family entrepreneurs (Franco & Prata, 2019) or the characteristics of the family firms themselves (Baier-Fuentes et al., 2019). Finally, we encourage scholars to examine the effects of gender, generation and non-family board members (Ferreira et al., 2018;Samara & Berbegal-Mirabent, 2018).

Disclosure statement
The authors have no conflict of interest.