Corps de l’article

A rarely studied issue in the family business internationalization literature is international diversification. While internationalization had been studied mainly through the lens of exports seeing family firms as unable or unwilling to opt for more sophisticated entry modes abroad, the strategic option of diversification had unfortunately been neglected. Yet, the study of the hindrances and motivations for this strategic choice in the context of family firms deserves an in-depth investigation because international diversification is one of the vectors of a firm’s growth and performance (Hoskisson and Hitt, 1990; Muñoz-Bullón and Sánchez-Bueno, 2012).

International diversification could be understood under the lens of the Socio-Emotional Wealth[1] perspective and the mixed gamble approach – specifically those gambles that have the potential outcome of gains and losses (Bromiley, 2009). SEW is the set of non-economic and emotional values destined to fulfill the family’s needs for the identification, control, and continuation of the family legacy (Gómez-Mejía et al., 2007). In the available research, there is a general consensus that the family firm has little interest in international diversification so as to avoid a dilution of family control and more generally a loss of the socio-emotional wealth that the firm confers to the family shareholder (Gómez-Mejía et al. 2010; Alessandri et al., 2018; Dou et al. 2020).

Among the many socio-emotional goals pursued by the owning family in a family firm, the identification of family members with the firm is a prominent feature. Indeed, in family firms, there is a strong alignment between the identity of the family shareholder and the identity of the firm (Berrone et al., 2012; Zellweger et al. 2013). For Zellweger et al. (2013), family and organizational identity tend to overlap, creating a mutually shared understanding of ‘who we are’ and ‘what we do’ in ‘our family’s business’. The importance given to this goal might imply that family firms would be reluctant to diversify operations at the international scale. Indeed, the loss of family control and the consequent weakening of the familial identity of the business is the most significant risk that diversification poses to the family firm.

How does this fit to the well-established finding that family firms are strongly oriented towards longevity and many of them leverage diversification as a means to achieve this goal? This question raises the theoretical challenge to reconcile theory and empirical facts. In this paper, we attempt to shed light on the crucial socio-emotional goal of family members’ identification with the firm by investigating the moderating effect of family firms’ learning orientation on the relationship between this goal and a firm’s degree of international diversification. In this paper, learning orientation means a firm’s exploratory and reproductive orientations, as these two variables had been substantiated as clear determinants of a firm’s internationalization in business internationalization research,

To address this challenge, our research calls up the SEW framework and the concept of mixed gambles as well as the learning approaches to a firm’s internationalization (Eriksson et al., 2000; Forsgren, 2002; Teece et al., 1997) to build a conceptual model tested on a sample data of 46 French family SMEs. The model was assessed through the PLS-SEM technique, and the two-step procedure of moderated multiple regressions was used (Akremi and Roussel, 2003). Our findings do not validate the thesis according to which reproductive orientation would reinforce the negative relationship between family members’ identification with the firm and international diversification. However, it was found that the exploratory orientation negatively moderates the relationship between family members’ identification with the firm and international diversification.

This research contributes to the literature on international diversification of family firms (Gómez-Mejía et al., 2010; Stadler et al., 2018) by highlighting the effect of one key socio-emotional goal, namely identification on this strategic decision. In addition, we add to the literature on organizational learning in family firms (Zahra, 2012; Tsang, 2020) by underlining the role of a firm’s learning orientation in shaping the behavior of family members and, consequently, on international diversification. The present research also provides a contribution to the stream of family business literature dealing with exploration and exploitation (Sharma et Salvato, 2011; Moss et al., 2014; Riviezzo et al., 2015), as our research highlights the role of exploration as a capability of the family firm allowing it to grow despite likely hindrances inherent in the owning family characteristics or some of its members.

The next part of this paper reviews the extant literature regarding international diversification in family firms. It then provides a conceptual framework linking the identity-preservation socio-emotional goal with international diversification and organizational learning orientation. The research design, the findings and contributions are presented and then discussed in the last parts of the paper.

Family Firms and International Diversification

The issue of family firms’ corporate strategies such as international diversification received growing attention in recent years. In the family business literature, research on international diversification is positioned at the confluence of diversification strategy research and firms’ internationalization studies. Overall, findings about the problem of family firms and diversification are rather controversial (Anderson and Reeb, 2003; Gómez-Mejía et al., 2010). Despite international diversification’s favorable outcomes (Firm risk reduction, growth, etc.), family firms may diversify less than non-family firms because diversification requires external capital, which may lead to insolvency and dependence on lenders, and external expertise by appointing outside managers, thereby leading to a loss of control (Gómez-Mejía et al. 2010).

On another level, the topic of the relationship between family firms and internationalization has given rise to various theoretical and empirical divergences, leading to contradictory results. While a first stream of research argues that family firms are encouraged to internationalize in order to maximize the owning family’s wealth and assure a firm’s renewal (e.g. Kraus et al., 2016; Tsao et al., 2018), another body of research contends that family firms are weakly internationalized (Baronchelli et al., 2016). Recent research adopted the socio-emotional wealth (SEW) perspective (e.g. Pukall and Calabrò, 2014; Kraus et al. 2016; Basly and Saunier, 2020) to address family firms’ heterogeneity in the context of internationalization. The main lesson of this perspective concerning internationalization is that family firms experience diverse forms of risk-taking, and are not generally risk-averse or risk-prone when pursuing internationalization, because family owners might aspire to protect or to develop their SEW.

Within this abundant literature on the internationalization of the family firms, rare research, to our humble knowledge, has dealt with diversification as a particular internationalization strategy, while most of empirical literature has merely focused on exports. Again, the literature on the international diversification of family firms has produced conflicting results. In a pioneering study, Gómez-Mejía et al. (2010) found that, on average, family firms diversify less both domestically and internationally than non-family firms. And, when they do diversify, they tend to opt for domestic rather than international diversification; and those that go the latter route prefer to choose regions that are ‘culturally close’. More recently, the findings of a research by Alessandri et al. (2018) indicate that family firms exhibit lower internationalization in general than nonfamily firms, through lower extent and breadth of internationalization and greater home region orientation. In a similar vein, in their an analysis of firm-level data from 93 countries over 2011-2018, Xu et al. (2020) found that family-dominant firms prefer a low breadth-high depth international diversification strategy.

While this first stream of literature has mostly focused on whether family firms are more or less inclined to diversification than non-family firms, another body of research has focused on the examination of differences in diversification among family firms. Based on fine-grained measures of family involvement in firms, these investigations attempted particularly to highlight the specific influence of family management and ownership on international diversification. For example, examining the impact of ownership on product and international diversification, and using data from 222 European firms between 1994 and 2007, Hautz et al. (2013) showed that family ownership concentration has a positive impact on product and a negative impact on international diversification. For their part, Cerrato and Piva (2012) analyzed data from a sample of 1,324 Italian manufacturing SMEs, and showed that involvement of the owning family in management negatively influences export propensity; but, once the choice to go international has been made, both the degree of internationalization (expressed in terms of sales entropy) and geographical scope in family-managed firms are not significantly different from nonfamily-managed firms.

The present research aims at contributing to the debate on this issue by proposing to refine the acknowledgement of family influence by considering socio-emotional factors. While most of previous studies approached family influence in family firms in terms of involvement in ownership, management or control, they simplify this crucial aspect by neglecting significant factors of family influence on the firm. Family business owners pursue business and family goals as well as financial and non-financial goals. In particular, family-centered non-financial goals (such as family succession, employment of family members) are key in setting organizational goals and shaping a family business behaviors and strategic choices (Chrisman et al., 2012). We argue that the owning family’s priorities and goals need to be related to the problem of international diversification, and more generally to family firms’ strategic management. To achieve our goal, we build upon the theoretical frameworks of socio-emotional wealth and mixed gambles in family firms.

As suggested by Gómez-Mejía et al. (2010), increased complexity (due to diversification) would force owners to hire executives from outside the family, and subsequently to give up some control over the decision-making process; this in turn may corrode the authority and identification foundations of SEW. Despite our agreement with this argument, this rationale had not been empirically tested, to the best of our knowledge. Our focus in this research will be about the specific socio-emotional goal of family members’ identification with the firm. In fact, the extant literature highlights how the main risk related to the loss of control is the denaturation of the firm, which may sooner or later lose its family character, resulting in an erosion (or even an outright loss) of the fit between the owning family’s identity and that of the firm.

Further, as family firms are heterogeneous, we believe that it is necessary to thoroughly investigate the conditions under which the priority given by the family to identification fully exerts its effect and influences its choices concerning international diversification. In total, this paper aims at filling the knowledge gap regarding the role of family members’ identification with the firm as a key socio-emotional goal impacting family firms’ international diversification. In addition, this research aims at investigating the moderating role of family firms’ exploratory and exploitative orientations on the relationship between family members’ identification with the firm and international diversification.

Theoretical Framework and Hypotheses Development

International Diversification as a Mixed Gamble for Family Firms

While the main theoretical arguments used to explain diversification in family firms rely mainly on agency theory (Anderson and Reeb, 2003; Ducassy and Prevot, 2010), more recent research called up the socio-emotional wealth framework and the mixed-gambles approach in explaining family firms’ international diversification (Alessandri et al., 2018; Dou et al. 2020; Xu et al. 2020). Socio-emotional wealth (SEW) is the set of non-economic and emotional values destined to fulfill the family’s needs for the identification, control, and continuation of the family legacy (Gómez-Mejía et al., 2007). Expressed differently, for Zellweger and Dehlen (2012), SEW is that part of a business value (as perceived by the owner) that is unexplained by financial considerations. Gómez-Mejía et al. (2007) empirically demonstrated that family firms can opt for financially risky strategies, while previous literature had particularly emphasized their prudence and risk aversion (Le Breton-Miller and Miller, 2006). The main thesis of Gómez-Mejía et al. (2007) is that the preservation of this Socio-Emotional Wealth (SEW) is key to understanding the behavior of family firms and that an owning family’s decision is motivated by loss aversion with respect to its stock of SEW (Gómez-Mejía et al. 2007). Accordingly, family entrepreneurs will purportedly make conservative choices to preserve their current stock of SEW and will take additional risks as necessary to prevent that loss (Cruz and Justo, 2017). Applying this rationale to diversification strategies, Muñoz-Bullón and Sánchez-Bueno (2012) argue that family firms are more reluctant to accept the change involved in diversification, given that their goal is to preserve family values and maintain control of the company. Indeed, for Gómez-Mejía et al. (2010), diversification is likely to pose a hazard to SEW because it entails greater uncertainty and delegation, as well as the appearance of new actors (such as creditors) from outside the family circle with the capacity to exert some influence and control over the strategic direction of the firm.

Recent advances by family firms’ research acknowledge that decision-making of family firms is more complex involving various goals, specifically financial as well as SEW goals (Kotlar et al., 2018). This mixed-gamble approach considers that it is essential to simultaneously consider the positive and negative repercussions concerning socio-emotional wealth, as strategic decisions might positively influence this wealth and allow for its development. On the other hand, it is crucial to simultaneously consider the financial and emotional dimension, as family owner-managers have to make trade-offs concerning these two factors, since the positive repercussions concerning one aspect could be compensated for by the other (Bromiley, 2009; Gomez-Mejía et al., 2014). In line with this framework, international diversification decisions might be regarded as mixed gambles for family firms’ owners, as this strategic option might result in positive and negative outcomes for financial and socio-emotional wealth (Table 1). Particularly, international diversification may not systematically reduce family firm’s SEW endowment but might also result in SEW growth and the achievement of various socio-emotional goals. For instance, it is plausible that the owning family might be prompted to grow and internationalize the business, which is considered by business owners as an extension of themselves, in order to acquire or develop notoriety and fame. Additionally, the owners in family firms may have incentive to internationally grow the firms that are intended to be passed on to successors, because internationalization, despite the additional financial risks it entails, is a vehicle for achieving a growth in a family’s wealth. For Alessandri et al. (2018), however, in viewing the trade-off between increased returns and threats to SEW from greater extent and breadth of internationalization and lower home region orientation, family firms appear to not believe the risk is worth the reward, as compared to nonfamily firms. For these authors, challenges, such as changes to existing methods of operations and concerns over potential loss of family control, appear to take precedence in the framing of the decision for family firms, making them more reluctant to expand too much or too broadly into international markets (Alessandri et al., 2018).

Table 1

The mixed gamble of diversification in family firms

The mixed gamble of diversification in family firms

-> Voir la liste des tableaux

A more nuanced conceptualization of SEW disaggregates it into five dimensions that include: Family control and influence, Family Identity, Binding social ties, Emotional attachment, and Renewal of family bonds to the firm through dynastic succession (Berrone et al., 2012). We propose that differences in the logic behind international diversification by family firms can be attributed to the existence of different SEW salient reference points among family principals. Specifically, our model suggests that when the family identity dimension of SEW is the owners’ main reference point, a family firm would be less inclined to international diversification. Achieving and maintaining a fit between the family and the firm’s identities is a key goal pursued by family owners in family firms (Zellweger et al. 2013). Family business research showed that the intertwining of the family and the business gives rise to an inherently unique identity within family firms (Berrone et al., 2012), as the identity of a family firm’s owner is inextricably tied to the organization that usually carries the family’s name (Berrone et al., 2012). This specific socio-emotional goal implies that the owning family cares about the human and social relations maintained with customers, partners and the overall community. Moreover, the firm is seen both by internal and external stakeholders as an extension of the family itself (Berrone et al., 2012). We believe that taking this fundamental socio-motional goal into consideration is essential in analyzing a family firm’s propensity to diversify on the international scale. The extant literature suggests that international diversification might entail significant risks to the family firm’s identity. Indeed, it might challenge family members’ engagement in the strategic direction of the firm, as it may require sharing the firm’s control with outsiders (new investors, creditors or outside directors) and adopting new values and beliefs. For example, Xu et al. (2020) argue that an emphasis on breadth in international diversification can hinder the propagation of family values in a family-dominant firm as it expands to foreign markets that are inevitably culturally distant from its home country. In addition, Hussinger and Issah (2019) suggest that SEW is closely related to the family firm’s core business. With time, family members typically have developed affection and emotional attachment for the core technology, products, and services as well as the domain knowledge and expertise needed to succeed in the industry (Hussinger and Issah, 2019). Yet, diversification may require working with new technologies, developing new markets and new activities, which would result in resistance of family members to this strategic decision. Furthermore, international diversification might imply working with outside partners or incorporating joint activities, as in the case of international joint-ventures. Thus, an owning family would not always be able to effectively monitor partners’ actions, which may be damaging to the firm’s reputation. For this reason, family firms would opt for culturally close international diversification, because such an approach allows the family to monitor international operations more closely without having to resort to extensive use of non-family human capital (Gómez-Mejía et al., 2010). Another rationale for the likely resistance to international diversification is the probable loss of family control resulting from the involvement of outside investors, which may also challenge the strength of the family members’ sense of belonging to the firm. In total, as international diversification poses a hazard concerning the family firm’s identity, and might challenge the family-to-the-firm identity fit (Zellweger et al. 2013), family firms might be less prone to diversify their operations on international markets. It is then expected that the orientation of family members towards the preservation of the fit between their identity and that of the firm would constitute a hindrance for international diversification.

In view of these arguments, a first hypothesis is posited:

H1: There is a negative relationship between the identification of family members with the firm and international diversification.

Most studies on family firms’ internationalization have considered their internationalization development as a sequential process (Johanson and Vahlne, 1977) based on a reproductive dynamic (Kontinen and Ojala; Metsola et al. 2020). Tacitly or explicitly, most of the existing studies highlight a sort of preference for exploitation-based internationalization on the part of family firms and provide legitimizing rationales of their less bold, aggressive exploration-based internationalization. Based on this fact, the consideration of exploitation and exploration as variables driving internationalization seems to us essential in studying family firms international diversification. These two variables had been extensively used in previous internationalization research and their effects had been substantiated as clear determinants of a firm’s internationalization. Moreover, previous family business literature highlighted the fact that the long-term ownership of a family business presumes a degree of competency in entrepreneurial exploration and exploitation, without which the family business could not be sustained as a family business (Miller and Le Breton-Miller, 2006). However, scarce studies have dealt with these issues in family business research and their explicit consideration in the investigation of family firms’ internationalization is lacking. In this research, we contend that considering these two features of learning might prove useful in shedding light on the problem of international diversification of family firms and will extend knowledge about the role of SEW in this strategy.

Exploitation, Exploration and International Diversification of Family Firms

A number of learning-based firm’s internationalization theories have mobilized the concepts of exploration and exploitation (eg. Luo, 2000; Forsgren, 2002; Prange and Verdier, 2011). For example, adopting a dynamic capabilities perspective, Prange and Verdier (2011) argue that there are two opposing classes of explorative and exploitative capabilities differentially linked to growth and survival on the international scale. In a dynamic capacities perspective (Teece et al., 1997), it is believed that a firm’s internationalization is founded on both reproductive and exploratory dynamics. Reproductive dynamics imply that capacities and skills are exploited in such a way that firm’s assets are reproduced or simply developed incrementally over a short- or medium-term horizon. Thus, these dynamics are based on learning by experience subject to path dependency (Nelson and Winter, 1982). Conversely, exploratory dynamics consist in the promotion of innovation and the creation of new routines and capabilities (Argyris and Schön, 1978). Thus, exploration-based learning implies a search process based on trial and error, which is less subject to path dependency and more chaotic, thereby allowing for the creation of new skills and routines (Forsgren, 2002).

Historically, the incremental internationalization models have privileged a reproductive conception of internationalization according to which the experiential learning process in the international context is incremental, cumulative and path dependent (Johanson and Vahlne, 1977; Forsgren, 2002). Specifically, these models emphasize that learning is related to current activities on foreign markets. In this setting, a firm would prefer to focus on and learn more about a specific market than to explore new alternatives. The models of early and rapid internationalization rather rely on exploratory dynamics and consider the internationalization decision in itself as an innovation. The reproductive conception is called into question if we consider that instead of simply solving problems, an organization might create and define problems, develop and apply new knowledge to solve them, and also develop more new knowledge through the action of problem-solving (Cyert and March, 1963; March, 1991). Additionally, a number of theoretical and empirical contributions moderate the impact of path dependency and support the exploratory approach by taking into account the “variation” i.e. the diversity of foreign environments to which a firm is exposed (Eriksson et al., 2000).

Family business literature nurtures the idea that family firms would be more oriented towards exploiting current opportunities rather than exploring new ones. For Sharma and Salvato (2011), family firms have advantages over nonfamily firms in creating and exploiting opportunities closely related to current operations. According to König et al. (2013), family influence aggravates critical sources of organizational paralysis, specifically emotional ties to existing assets and the rigidity of mental models. As a consequence, as explained by Miller and Le Breton-Miller (2014), a family firm’s focus on existing competencies and products or services often leads to a lack of competence when venturing into new markets. A family firm’s exploitative orientation has every chance of being accepted and even encouraged by the shareholding family because its fundamental nature is consistent with the essential values shared by many family businesses. The logic of exploitation defined as “the refinement and extension of existing competencies, technologies, and paradigms exhibiting returns that are positive, proximate, and predictable” (March, 1991: 85) is highly consistent with the goal of SEW preservation in family firms. Exploitation even risks exacerbating the tendency of the owning family to identify with the firm because, all other things being equal, it implies a low openness to external expertise, finance and knowledge. Moreover, exploitation would not necessarily require the firm to adopt new mental models or change its values system. Based on these arguments, it is hypothesized that an exploitative orientation might strengthen the negative relationship between family members’ identification with the firm and international diversification. In other terms, a family firm’s inclination towards exploitation might comfort family members in their SEW loss aversion orientation expressed by their preference to protect the family business identity and consequently to less diversify the business. Conversely, we contend that an exploratory orientation might weaken the negative relationship between family members’ identification with the firm and international diversification. As exploration implies “experimentation with new alternatives having returns that are uncertain, distant, and often negative” (March, 1991: 85), it seems at odds with the goal of maintaining a fit between the family and the firm’s identities. This conflict of logics can, in our opinion, attenuate family members’ identification with the firm because, among other things, exploration-based internationalization would require opening the firm to outside influence and resources through the incorporation of external capital, people, knowledge and technology. Moreover, exploring new markets or regions - especially those that are geographically and psychologically distant - would require understanding, adopting or even embracing new ways of thinking and new organizational values. Even if it is conceivable that family members might resist this orientation, as exploration might reduce the family-to-firm fit, it is also likely that it could push them to perceive the benefits of international diversification particularly regarding the owning family’s identity and long-term continuity. Accordingly, we argue that an exploratory orientation reverses the mixed gamble assessment and drives family members to search for SEW growth by focusing in priority on their firm’s long-term growth and on business opportunities allowing for wealth creation. Therefore, in this research, it is expected that:

H2: Exploitative orientation positively moderates the relationship between the identification of family members with the firm and international diversification.

H3: Exploratory orientation negatively moderates the relationship between the identification of family members with the firm and international diversification.

Method

The hypotheses were tested on a sample data of 46 family SMEs collected through a questionnaire addressed to top management executives. The model was assessed through the PLS-SEM technique. The survey instruments developed to measure the variables of the model are based on previously validated constructs.

Sample

This study was carried out in the French context which is an appropriate setting for studying international diversification of family SMES. Indeed, French family businesses represent one third[2] (40.000 to 50.000 firms) of the overall number of small- and medium-sized enterprises (excluding very small businesses) in the country. Family firms also make up 40% of the 250 largest French firms (McKinsey&Company, 2010)[3]. Furthermore, while 4900 French multinational firms (excluding the banking sector and non-market services) controlled 43600 foreign subsidiaries (in 2017), there is a preponderance of large multinational firms while Intermediate-sized Enterprises and SMEs generate less revenues abroad compared to these large firms (INSEE, 2017)[4]. It is plausible that the low internationalization of French SMEs is due to their owners’ concern to preserve SEW, as suggested by existing theories. Therefore, researching the difficulties hampering their internationalization and suggesting solutions to encourage them to internationalize further is quite relevant.

The present study targeted a population of French family SMEs with a size comprised between 10 and 249 employees and total sales between 2 and 50 million €[5]. The target sample was selected from the French firms’ database Diane (Bureau van Dijk), which gathers data from over 1.4 million French listed or unlisted companies. Obtaining reliable a priori information in order to identify family SMEs is difficult (Vandekerkhof et al., 2018), especially in France where a comprehensive list of family firms is lacking. Following family business literature that considers familiness as a continuum, family businesses were identified in this study as firms in which the primary owners were individuals sharing the same family name, or belonging to the same family or to multiple families, whatever the equity or voting rights (no minimum threshold of ownership was defined in Diane Database) (Tagiuri and Davis, 1996; Moss et al., 2014). In the case of one individual shareholder, there had to be at least one other member of his/her family involved in the management or governance bodies in order for the firm to be qualified as a family business. In order to ensure that the target population was made up of internationalized firms, all firms that had not achieved a level of exports of at least 5% of total sales during the last 3 years were excluded[6]. Public utilities and financial institutions were also excluded from the sample, as their financial characteristics - and particularly leverage - are idiosyncratic[7]. Therefore, the study focused on manufacturing, commerce and service firms. The selection criteria resulted in a sample of 1,443 companies, from which 884 were classified as family businesses.

Data Collection

A regular mail survey was sent to the top family managers exerting executive functions (CEO or President of the management board[8] or top manager as CFO) or a monitoring role (Chairman of the board[9]). The questionnaire was sent in November 2015. A follow-up emailing to the 749 non-respondents was carried out on January, 1st 2016. A total set of 47 responses was received by the end of January 2016, resulting in a response rate (5.3%). The timing of survey administration coincided with a difficult period for firms and individuals in France that could have affected the response rate, as it shortly followed the terrorist attacks in Paris (November 13, 2015). Ex ante precautions regarding the design of the survey were taken to avoid non-response bias. Furthermore, differences between early and late respondents were tested through ANOVA, T-tests for equality of means and Mann-Whitney U tests and resulted in no significant difference, suggesting that nonresponse bias is not a concern. In view of the sample size, and despite few questions being left unanswered in some questionnaires, 46 out of 47 questionnaires were kept and exploited for the statistical analyses.

Statistical Choices

Rather than covariance-based structural equation modeling (SEM), partial least squares SEM (PLS-SEM) was chosen to evaluate the research model, because the latter has no assumptions about data distribution (Chin, 1998). Moreover, being based on a series of OLS regressions, PLS-SEM can be used with small samples (Kyu Kim et al., 2011) and generally achieves high levels of statistical power (Hair et al. 2012). Moreover, it allows analyzing structural models with multi-item constructs as well as in direct and indirect relationships (Vallejo, 2009), which corresponds to the present research model.

Variable Measurement

In the present model, the variables were defined from previously validated constructs then translated into French to be included in the survey instrument. All construct items were measured on Likert-type scales from one to five. The two constructs of the research model are reflective. Factor analysis was performed to assess reflective measurement models, in order to reveal the relations among observable items and provide support for unidimensionality of constructs. In addition, calculating Cronbach’s alpha allowed an examination of the internal consistency of constructs.

Dependent variable: Following previous studies (Gómez-Mejía et al., 2010), we have used the entropy index to measure the international diversification strategy. The entropy measurement of diversification is defined as Σ Pi ln(1/Pi), where Pi is the share of a firm’s total sales attributed to geographic region i, and ln(1/Pi) is the weight of each geographic region i (Europe, Americas, Asia/Pacific, Africa and Other). This measurement takes into account both the number of regions in which a firm operates and the relative importance of each region over a firm’s total sales.

Independent variable: the “identification” construct was measured on the basis of the set of five items suggested by Berrone et al. (2012) (See Appendix). Following PCA (Principal Component analysis), item 4 was dropped because of weak communality. In the explanatory models (see below) tested through Smart PLS, the construct has a Cronbach’s Alpha of 0,847 and an AVE (Average Variance Extracted) of 0,690. All of the items had loadings with values superior to 0,7 (except item 3 whose loading was 0,638).

Moderators: Exploitation and exploration were measured through ten items derived from Bedford (2015). PCA analysis leads to dropping items 1 and 9 (weak communalities and loadings). Items 6, 8 and 10 formed the exploitation construct (item 10 was then dropped in Smart PLS model assessment) and items 2, 3, 4, 5 and 7 measured the exploration construct. Exploitation had a Cronbach’s Alpha of 0,768 and an AVE of 0,732. Exploration had a Cronbach’s Alpha of 0,886 and an AVE of 0,679.

Control variables: In accordance with previous research (Ducassy and Prevot, 2010; Cerrato and Piva, 2012; Gómez-Mejía et al., 2010), the following control variables were used. First, family equity was measured through the percentage of equity held directly or indirectly by the owning family (data obtained from Diane Database); leverage was measured through the financial debt by total medium/long term financial resources (for the year 2014); profitability was measured by the Gross Return on Equity (2014); corporate risk was measured for 2014 by the natural logarithm of (ROA 2014/ROA 2013); finally, financial slack was measured by the current ratio (2014). Firm’s size was also controlled for by including the number of employees (expressed in Ln) in the model. Finally, firm’s age was measured by the number of years since foundation (expressed in Ln).

Results

Sample Description

In the present sample, family firms are significantly controlled by owning families, as the average percentage of equity held is of 93.45 (varying from 60% to 100%). Owning families also appear to be intensely involved in the managing and monitoring organs, as on average 90.96% of board members are family members, and almost 86% of top management team members belong to firms’ owning families. Table 2 also reveals that the sample is mainly made up of first- and second-generation family firms. Finally, the sample consists of 31 manufacturing firms, 10 commercial firms and 5 services firms.

Table 2

Generations involved in ownership, management and governance

Generations involved in ownership, management and governance

-> Voir la liste des tableaux

Findings and Discussion

To test our hypotheses, the two-step procedure of moderated multiple regressions was used (Akremi and Roussel, 2003). First, a model including control variables, independent and moderator variables was tested. In a second step, the product of independent and moderating variables representing the non-linear interaction effect was included. The moderator role is established if the regression coefficient associated with the interaction effect is statistically significant and the determination coefficient (R2) associated with the second regression is higher than the first.

A collinearity analysis was performed regarding the independent variables; it resulted in variance inflation factors all less than 5.0, which suggests that the estimated path coefficients are well established in the model. The results are exhibited in Table 5, 6, 7 and 8 and show explained variance (R2) of the dependent variables and the path coefficients of the model. In agreement with Chin (1998), a bootstrapping (5.000 samples) was used to generate standard errors and the t-statistics.

Table 3

Descriptive statistics

Descriptive statistics

-> Voir la liste des tableaux

Table 4

Correlation matrix

Correlation matrix

-> Voir la liste des tableaux

Table 5

Effect of exploitation (Step 1) – R2 = 0,380 / Adjusted R2 = 0,230

Effect of exploitation (Step 1) – R2 = 0,380 / Adjusted R2 = 0,230

-> Voir la liste des tableaux

Table 6

Effect of exploitation (Step 2) – R2 = 0,391 / Adjusted R2 = 0,222

Effect of exploitation (Step 2) – R2 = 0,391 / Adjusted R2 = 0,222

-> Voir la liste des tableaux

Table 7

Effect of exploration (Step 1) – R2 = 0,396 / Adjusted R2 = 0,249

Effect of exploration (Step 1) – R2 = 0,396 / Adjusted R2 = 0,249

-> Voir la liste des tableaux

Table 8

Effect of exploration (Step 2) – R2 = 0,472 / Adjusted R2 = 0,325

Effect of exploration (Step 2) – R2 = 0,472 / Adjusted R2 = 0,325

-> Voir la liste des tableaux

In the two main models, our results validate the hypothesis put forward by previous studies regarding the negative relationship between the goal of family members’ identification with the firm and international diversification, which corroborates the thesis of socio-emotional wealth preservation goal (Gómez-Mejía et al., 2010; Dou et al., 2020). The tested models show the positive influence of exploitation and exploration on international diversification corroborating the need for these two capabilities to achieve international growth. This interesting result also supports the thesis of family firms’ ambidexterity (Riviezzo et al., 2015) by showing that these two orientations are not exclusive but need to be complementary so that the family firm achieves its strategic goals. This result supports the view of Riviezzo et al. (2015) according to whom the “exploration of new opportunities while maintaining the ability to leverage the existing market is the result of a strong identity perceived by all members of the family, which results in deeply held values, passed from generation to generation, and in a reputation to defend at all costs” (Riviezzo et al., 2015). However, it is worth noting that the positive influence of exploration on international diversification is stronger than that of exploitation. This is not surprising, as exploration inherently implies broadening the scope of a firm’s operations on different markets. This finding is consistent with Prange and Verdier (2011) who suggest that growth rates are higher for firms pursuing an explorative internationalization than firms pursuing an exploitative internationalization.

No significant moderation effect is found as for the exploitation variable (Table 6). Therefore, this finding does not substantiate our second hypothesis. Based on the step 1 model, the main finding is that family members’ identification might have a detrimental effect on international diversification while a firm’s exploitative orientation might positively affect this latent variable. It is plausible, as suggested by Gómez-Mejía et al. (2010), that when family firms do diversify, they tend to opt for domestic rather than international diversification, and those that go the latter route prefer to choose regions that are ‘culturally close’. Family firms are known to be strongly rooted in their immediate territories and communities, and likely direct their investments towards the opportunities that come from these settings. It can therefore be concluded that there would be a natural inclination of family firms towards exploiting international opportunities that are less distant, less risky and easier to exploit (all other things being equal). This is all the more important that family firms, especially when they are small or medium-sized, have to make a trade-off in terms of resource allocation and voluntarily prefer opting for local at the expense of international growth because of resource scarcity. To sum up, these first results substantiate Sharma and Salvato’s argument (2011) according to which family firms have advantages over nonfamily firms in creating and exploiting opportunities closely related to current operations.

As far as exploration is concerned, a negative moderation effect is observed at the significance threshold of 10%[10]. This finding corroborates our third hypothesis according to which the more the firm is oriented towards exploring new activities and business opportunities, the less family members’ identification with the firm would negatively influence international diversification. Therefore, a family firm’s exploratory orientation appears to be widening the horizons of the business and allows for envisioning a riskier development of the firm’s operations on an international scale. By exploring new ideas, development paths and business opportunities, the owner-manager is likely to succeed in curbing the negative effect of family members’ identification with the firm on international diversification choices. An important conclusion is that exploratory orientation seems to reverse the mixed-gambles assessment by minimizing concerns over international diversification and emphasizing its positive outcomes. A firm’s exploratory orientation might reinforce the interest of the family firm’s owner-managers and family members in benefiting from the advantages of international diversification in terms of firm’s growth potential and business generational sustainability. In addition, international diversification could offer an opportunity to delegate responsibilities to potential successors as a training for them before taking the firm over. Furthermore, international diversification would make it possible to meet the family’s socio-emotional need to forge new relationships and to develop social relations beyond national borders. To achieve this goal, exploratory orientation would help the owning-family carrying out social networking activities, the benefits of which could be reflected on the business. Finally, some family firms choose to operate jointly with other family businesses which are similar to them and which share the same creeds and values (Swinth and Vinton, 1993). Thus, the risk of identity loss in the context of international diversification can be minimized in the context of alliances or joint operations with foreign family businesses. International diversification might even favor the owning family member’s identification with the firm considering that this strategic option might allow for developing the owning family’s reputation and notoriety beyond national borders.

Contributions and Implications

This research contributes to the literature on family firms’ international diversification by highlighting the effect of one key socio-emotional goal, namely “identification” on this strategic decision. While most of previous research analyzed family influence in terms of involvement in ownership, management or control and sometimes indirectly approached SEW concerns (Gómez-Mejía et al., 2010; Muñoz-Bullon et al., 2018), this study is the first, to the best of our knowledge, to explicitly focus on the key SEW goal of family members’ identification and its influence on international diversification. Building on previous arguments, we show that while the pursuit of identity preservation and identification with the firm might negatively influence international diversification (Gómez-Mejía et al., 2010), a firm’s exploratory orientation negatively moderates the relationship between identification and international diversification. Therefore, this study suggests that the heterogeneity of family firms in face of a complex strategic decision such as international diversification needs to take account of factors related to strategic behavior and learning orientation. More generally, more than the type and structure of shareholding (Hautz et al., 2013; Majocchi et al., 2013), our study underscores the need to further push the analysis of a firm’s shareholding by taking into account non-economic factors such as emotions and affect-related aspects in explaining international diversification. Therefore, this research brings its stone to the building of research on firms’ diversification and particularly on a significant element of organizational processes, which is organizational learning (Hitt et al., 2006).

In addition, we add to the literature on organizational learning in family firms (Zahra, 2012; Tsang, 2020) by underlining the role of a firm’s learning orientation in shaping the behavior of family members and, consequently, on a key strategic move such as international diversification. We show that the exploratory learning orientation promoted by the firm’s management might constitute a counterweight to the resistance and strategic inertia manifested by the shareholding family when it strongly identifies with the firm. More generally, our study provides an illustration of how an owning family characteristics - and particularly its inclination to pursue SEW goals such as identification - might interact with the firm’s organizational learning predispositions affecting international diversification. While not much is known about how organizational learning occurs in family firms, our study contributes to this regrettably meagre literature by showing that family firms are not inherently learning-oriented or “learning-averse” but that socio-emotional goals pursuance might play a role in explaining exploration/exploitation orientations in family firms[11].

The present research provides also a contribution to the stream of family business literature dealing with exploration and exploitation (Sharma et al., 2011; Moss et al., 2014; Riviezzo et al., 2015). Recent research advances that the characteristics of family firms allow them to simultaneously promote processes of exploration and exploitation leading to organizational ambidexterity (Riviezzo et al., 2015; Ben Mahmoud-Jouini and Mignon., 2016). In line with this research, our study highlights the role of exploration as a capability of the family firm allowing it to grow despite likely hindrances inherent in the characteristics of the owning family or some of its members. In addition, our results do not neglect the role of exploitation as a determinant of international diversification of family firms. Therefore, we show that family firms use both of these capabilities to achieve their internationalization strategic goals. In other words, our study corroborates the thesis of the organizational ambidexterity of family firms in the specific context of international diversification.

Based on these findings, family business owners are invited to encourage the building of their firms’ exploratory orientation, particularly by appointing external managers to strategic decision-making positions, since their role in international diversification may prove decisive. This is because a high concentration of family members in managerial positions might hinder the international entrepreneurship process (Alayo et al., 2019). Otherwise, family managers can be a source of openness and new initiatives if, in particular, they were efficiently trained and exposed to the culture of internationalization, especially during their prior studies or journeys abroad. In sum, while exploratory orientation is likely to originate from outside managers, it is not excluded that family managers might be motivated to explore new business opportunities abroad despite a likely overwhelming familial orientation towards identity preservation.

Family firms’ owners sometimes misjudge the impact of international diversification by considering, due to cognitive bias, its perceived risks as more certain than its hypothetical benefits (Alessandri et al., 2018). Following recent studies, our research invites them to consider both the advantages and disadvantages of this strategy for the family firm, particularly in terms of the strategic renewal necessary to ensure business continuity and its transfer over generations. In addition, we invite them to consider the goal of identification and identity preservation no longer as a hindrance but rather as a vector for the firm’s development, considering that the family business (and the owning family) has many things to gain by developing activities on an international scale.

Limitations and Future Research

This study is not without limitations. The sample’s size may prevent from generalizing the findings, as our sample size did not meet the minimum size calculated by the rules of thumb (Chin, 1998; Hair et al., 2012). Thus, our findings should be interpreted with caution, as their statistical power could be considered moderate. The scope of this study can be broadened to include other explanatory and contingency factors, as exploration and exploitation are subject to the influence of external factors, such as environmental complexity and uncertainty. Further, future research can more deeply explore how socio-emotional goals such as identification might influence organizational learning orientation in family firms.